LIBRARY 
 
 THE UNIVERSITY 
 OF CALIFORNIA 
 
 SANTA BARBARA 
 
 PRESENTED BY 
 
 MR. & MRS. MILES TINKER
 
 SILENT READING
 
 THE MACMILLAN COMPANY 
 
 NEW YORK BOSTON CHICAGO DALLAS 
 ATLANTA - SAN FRANCISCO 
 
 MACMILLAN & CO., LIMITED 
 
 LONDON BOMBAY CALCUTTA 
 MELBOURNE 
 
 THE MACMILLAN CO. OF CANADA, LTD. 
 
 TORONTO
 
 SILENT READING 
 
 A STUDY IN THE PSYCHOLOGY AND PEDAGOGY 
 OF READING 
 
 BY 
 
 JOHN ANTHONY O'BRIEN, PH.D. 
 
 PROFESSOR IN THE COLUMBUS FOUNDATION 
 AT THE UNIVERSITY OF ILLINOIS 
 
 Neto gorfe 
 
 THE MACMILLAN COMPANY 
 1922 
 
 fffUt ruervtd 
 
 I
 
 COPYBIGHT, 1921, 
 
 BY THE MAOMILLAN COMPANY. 
 
 Set up and elect retyped. Published March, 1921. 
 
 Nortoooli \3rcss 
 
 J. S. Gushing Co. Berwick <fc Smith Co. 
 Norwood, Mass., U.S.A.
 
 So 
 BURDETTE ROSS BUCKINGHAM 
 
 In grateful acknowledgment 
 of my indebtedness to him 
 
 THIS BOOK IS DEDICATED
 
 PREFACE 
 
 THE processes and purposes involved in silent reading 
 have lately been subjected to searching analysis. The 
 former are revealed as both physiological and psycho- 
 logical, the latter as both complex and various. Each 
 has its lessons for the guidance of the teacher; and 
 each exalts the importance of reading as a school 
 subject. 
 
 Reading processes on their psychological side are 
 essentially associative associative of impressions with 
 meanings and of meanings with each other. In the 
 latter sense they are frequently identified with study. 
 Indeed, silent reading of the reflective type differs in 
 no practical sense from study. The fact that in study 
 we repeat and recapitulate adds but little of a qualita- 
 tive nature to the reading act. Processes on the 
 physiological side consist in large part of eye-move- 
 ments. The number, length, groupings, and progres- 
 sive or recessive character of these movements are at 
 once symptoms and conditions of reading efficiency. 
 Moreover, they take place according to patterns which 
 are peculiar to each individual, being the product of 
 original nature and of practice. So far as they depend 
 on practice, they are capable of improvement ; and this
 
 VU1 PREFACE 
 
 improvement is a matter of habit formation. Since 
 each eye-movement is definitely terminated by a pause 
 or point at which the eye rests and during which seeing 
 takes place, the analysis of eye-movement becomes in 
 fact the analysis of pauses. Better habits of eye-move- 
 ment, therefore, consist in better habits of eye pausing. 
 
 The purposes of silent reading have been enumerated 
 at some length by various writers. For example, we 
 may read for appreciation, for entertainment, or for 
 information ; and each of these purposes may be com- 
 bined to a varying degree of intensity with one or both 
 of the others. Moreover, each is capable of subdivision. 
 In reading for information we may seek to acquire as 
 much as possible of what the author has written, or we 
 may try merely to apprehend his main points, or again 
 we may wish to find out what, if anything, he has to 
 offer on a topic which we have in mind before we begin 
 the reading. Moreover, with reference to these in- 
 formational purposes, we may read for immediate use, 
 or for use on a definite and early occasion ; or we may 
 read to make the material as permanent a possession as 
 possible. 
 
 It is clear that in the teaching of reading the pur- 
 poses for which people read must receive consideration. 
 In particular, pupils should be placed in situations 
 where certain purposes consciously or unconsciously 
 become appropriate ; and they should be led to adopt 
 attitudes and modes of action with reference to these 
 purposes. Thus the method by which reading is taught
 
 PREFACE ix 
 
 will vary with the object in view. But no matter what 
 the object may be, the material must be understood, or 
 the process is not reading at all. The degree of com- 
 prehension, in the light of the reader's purpose, is 
 rightly taken as a test of good reading. 
 
 It is not, however, the only test. No matter what 
 the reading purpose may be at any given time, the 
 effectiveness of the reading act is dependent not only on 
 the degree of the reader's comprehension but also upon 
 the readiness or speed with which the act takes place. 
 Moreover, rate' and comprehension of reading are 
 related in such a way that on the average fast readers 
 comprehend better than slow readers. This does not 
 mean that a given individual comprehends better the 
 faster he reads. It means that as between two indi- 
 viduals one of whom has a natural or habitual rate 
 higher than that of the other, the faster reader will 
 probably read more understandingly. It also means 
 that if the same individual can raise his rate of habitual 
 reading, he will probably improve his comprehension 
 of what he reads. 
 
 There are reasons why this should be true. In the 
 first place, since comprehension is an associative process, 
 it will be facilitated, as association is always facilitated, 
 by anything which brings the associated elements into 
 closer sequence. Rapid reading brings into operation 
 the law of contiguity by bringing the author's ideas 
 into the reader's consciousness in immediate connection 
 with each other. The child who reads haltingly under-
 
 X PREFACE 
 
 stands little because he receives the ideas in isolation. 
 In the second place, the actual process of reading should 
 be in the background of consciousness in order that 
 attention may be concentrated on the thought. The 
 rapid reader is the one who has so mastered the 
 mechanics that this concentration of attention can 
 take place. 
 
 Not only are the reasons plausible for a connection 
 between comprehension and rate of reading, but the 
 evidence on the question is conclusive. On numerous 
 occasions data have been published. For example, in 
 an article in the March, 1920, number of the American 
 Schoolmaster a brief but clear-cut inquiry was made 
 into the relation between rapid reading and under- 
 standing of the material read. A class of college stu- 
 dents, reading a psychological article for one minute, 
 varied in the number of words read from 174 to 520. 
 Immediately after the reading, questions were asked 
 on the first paragraph which, since it contained only 
 154 words, had been read by all the students. There 
 was a tendency for the fast readers to answer more 
 questions correctly. The strength of this tendency may 
 be expressed by the fact that the person who ranked 
 above the average in rate had five out of eight chances 
 to rank above the average in comprehension. 
 
 Therefore, both because rate is an important factor 
 in every type of reading and also because it tends to 
 vary with comprehension, rate of reading is an impor- 
 tant test of good reading. Indeed, one may plausibly
 
 PREFACE xi 
 
 contend that a person's habitual rate of reading is the 
 one best measure of his ability as a reader. Not only 
 does it go with comprehension but it also strictly con- 
 ditions the volume of reading. Everything we do is 
 done in time. I have a certain time to devote to the 
 morning paper. The extent of my comprehension of 
 the news of the day depends to a large extent on the 
 amount I can cover in the time at my disposal. In a 
 large sense the amount which one can read measures 
 the extent to which one is well informed. In school 
 the pupil who can read rapidly reads much, and by that 
 fact alone has a better understanding. In adult life 
 praise is accorded to the "well-read" man. Compre- 
 hension, therefore, as it concerns competence in reading, 
 is definitely related to the amount read ; and the latter 
 is in turn definitely related to the rate at which reading 
 takes place. 
 
 The conditions which favor rapid reading are there- 
 fore of importance to school people. If eye-movement 
 is a matter of habit, then the laws of habit formation 
 apply. If eye-movement conditions rate of reading, 
 then these laws may be invoked to improve not only 
 rate of reading but also the entire reading process. 
 The present volume is an attempt and it seems to me 
 an unusually successful one to realize under class- 
 room conditions some of the possibilities for improving 
 rate of reading which have been suggested in the 
 laboratory. A practicable method has been devised 
 and put into operation with regular teachers and with-
 
 Xll PREFACE 
 
 out requiring more than the usual amount of class time. 
 Indeed, the method is simplicity itself. Taking a cue 
 from the so-called psychology of skill, and regarding 
 the improvement of eye-movements as analogous to the 
 improvement of hand movements in typewriting and 
 telegraphy, the author has provided for intensive 
 effort during short periods. The effort is specifically 
 directed toward the most rapid reading which is con- 
 sistent with the understanding of what is being read. 
 The details of the method are the devices used to stim- 
 ulate and maintain effort. 
 
 It is recorded that under the pressure of this effort the 
 pupils broke to new high levels of reading rates. These 
 levels were no doubt accompanied by integrations of 
 short eye-movements into longer movements just as 
 under stress Professor Book's typists integrated their 
 movements and attained "higher order habits." 
 
 As a contribution to methods of teaching silent 
 reading, this book will be regarded as significant. It 
 does not pretend to exhaust the subject. The plan 
 has rather been to "divide and conquer" -a maxim 
 as valid in education as in war. The author has recog- 
 nized the value of speed and has set up a way to attain 
 it. Although he has also provided for a record of 
 comprehension, he has done so chiefly in order to pre- 
 vent its deterioration while speed was being developed. 
 He gives evidence tending to show that in this he has 
 succeeded. His method is so readily adaptable to 
 conditions that it may be utilized in conjunction with
 
 PREFACE Xlll 
 
 a complete method of teaching silent reading which 
 takes account of the varying purposes for which reading 
 is done. Reduced, for example, to a ten-minute prac- 
 tice period a day, it will leave time for the different 
 kinds of reading which have been identified with study ; 
 and it should make all of these more productive. 
 
 B. R. BUCKINGHAM.
 
 ACKNOWLEDGMENTS 
 
 THE present study is a cooperative investigation. 
 It represents the work of many minds and many hands. 
 The investigation has been conducted by the author, 
 under the auspices of the Bureau of Educational Re- 
 search at the University of Illinois. To the Bureau 
 the writer acknowledges his indebtedness for assistance 
 of many and varied kinds. 
 
 To the superintendents, principals, teachers, and 
 pupils participating in this investigation, the author's 
 gratitude is due. To their painstaking labors and 
 careful execution of directions, much of the success of 
 the experiment is attributable. 
 
 The author acknowledges with pleasure the courtesy 
 of the College of Education of the University of Chicago 
 for graciously permitting the use of the apparatus for 
 photographing the eye-movements of readers. His 
 thanks are likewise due to Dean W. S. Gray, of the 
 University of Chicago, for valuable suggestions, and 
 to Mr. G. T. Buswell, of the same institution, for his 
 work in developing the photographic records. 
 
 For assistance in this research the author ac- 
 knowledges his indebtedness to Professor B. R. Buck- 
 ingham, Professor D. S. Hill, Professor W. W. Charters, 
 Professor'W. S. Monroe, then members of the College 
 of Education of the University of Illinois. The author 
 is also grateful to Professor I. M. Bentley, of the De- 
 
 XV
 
 XVI ACKNOWLEDGMENTS 
 
 partment of Psychology, and to Professor E. C. Cam- 
 eron of the College of Education of the University of 
 Illinois, for reading many chapters of the manuscript 
 and for helpful suggestions and criticism. 
 
 To Professor W. W. Charters, formerly Dean of the 
 College of Education of the University of Illinois, the 
 author is indebted for his sympathetic interest and 
 encouragement, and for a kindness of a nature quite 
 apart from our academic relations. He wishes to ac- 
 knowledge here also his gratitude to Professor G. M. 
 Whipple, who first interested him in the experimental 
 study of problems in educational psychology. 
 
 In a most especial manner is the author indebted to 
 Professor B. R. Buckingham, Director of the Bureau 
 of Educational Research, under whose immediate super- 
 vision this investigation has been conducted. Despite 
 the pressure of multitudinous duties, Professor Buck- 
 ingham's guidance and assistance have been constant 
 and unfailing. Whatever of merit may be found in 
 the present study is due in a large measure to his in- 
 valuable directions and guidance ; the shortcomings 
 represent merely the author's inability to profit fully by 
 his guidance. 
 
 There are many others who have assisted the author 
 in various ways. Their names, though too numerous 
 to be mentioned here, are not too numerous to be re- 
 tained in the grateful memory of the author. 
 
 J. A. O'B. 
 
 THE UNIVERSITY OF ILLINOIS 
 May 14, 1920
 
 TABLE OF CONTENTS 
 
 CHAPTER PAGE 
 
 I. INTRODUCTION: RESUME OP PREVIOUS INVESTIGA- 
 TIONS . 1 
 
 II. THE PROBLEM 
 
 III. FACTORS INFLUENCING THE RATE OF SILENT READING 
 
 IV. TYPE I. TRAINING IN RAPID SILENT READING 
 
 V. TYPE II. TRAINING TO DECREASE VOCALIZATION . 
 
 VI. TYPE III. TRAINING IN PERCEPTION 
 
 VII. STATEMENT OF PROCEDURE 
 
 VIII. STATEMENT OF RESULTS 
 
 IX. INTERPRETATION OF RESULTS . . 
 
 X. PHOTOGRAPHIC RECORD OF EYE-MOVEMENTS DURING 
 
 READING 232 
 
 XI. SUMMARY OF CONCLUSIONS 270 
 
 XII. BIBLIOGRAPHY 275 
 
 XIII. APPENDIX . 281 
 
 xvil
 
 SILENT READING 
 
 CHAPTER I 
 
 INTRODUCTION: RfiSUMfi OF PREVIOUS 
 INVESTIGATIONS 
 
 SINCE the invention of the art of printing by Johann 
 Gutenberg about 1448, the reading of printed sym- 
 bols has continued to grow in importance until to-day 
 it constitutes the fundamental subject of the elementary 
 school curriculum. More than one-fourth of the total 
 time in our grade schools is now devoted to the teaching 
 of reading, according to a recent investigation of time 
 distribution (l). 1 Indeed, in the lower grades, little 
 else but the mastery of the mechanics of reading is 
 attempted. Dramatization, games, picture cards, 
 blackboard exercises, charts, etc., find their essential 
 raison d'etre in the assistance they render to the child 
 in his attempt to interpret the printed symbols. Prob- 
 ably 70 per cent of the time in the primary grades is 
 devoted to exercises which have as their aim the inter- 
 
 1 Numbers in parentheses refer to the corresponding numbers in the 
 bibliography at the end of the book.
 
 2 SILENT READING 
 
 pretation of the written word. This is but natural, 
 since reading constitutes the basic instrument which 
 enables one to penetrate the mines of information 
 not only in all the branches of the curriculum, but on 
 every topic that has been, at some time or other, the 
 object of human study and investigation. This 
 information remains locked up securely within the 
 printed letters of the alphabet, forever hidden from 
 the illiterate, for it can be reached only through the 
 medium of reading. 
 
 The method of teaching the mechanics of reading 
 which held undisputed sway throughout the centuries 
 was inherited from the ancients. Tracing with the 
 stylus the various characters of their alphabet upon 
 waxen tablets and on papyrus, the pedagogues of an- 
 cient Greece and Rome taught their pupils the names 
 and phonic properties of each letter as the first and 
 indispensable step in the process of learning to read. 
 What could appear more logical than to postulate a 
 priori that before a child could read or pronounce any 
 word he must know the names and sounds of the 
 constituent letters of the word? How could a whole 
 word, consisting, as it does, of several letters, possibly 
 be pronounced unless one had first mastered the pro- 
 nunciation of each of the letters making up the total 
 word ? The fact that the word itself might be as simple 
 and as clearly a unit of sound as the letter, seems 
 never to have disturbed their serene acceptance 
 of the above principle. Unchallenged and unques-
 
 INTRODUCTION 3 
 
 tioned the alphabetic method remained in almost uni- 
 versal use in the schools of the world until the last 
 quarter of the nineteenth century. It remained for 
 the new science of experimental psychology to demon- 
 strate conclusively that the mode of procedure in the 
 learning process which would appear to the adult as 
 the logical order is not de facto always the mode of 
 procedure which the mind of the child actually takes. 
 In other words, there has been shown to be a psycholog- 
 ical order of procedure which is totally distinct and 
 different from the order of procedure deduced by the 
 adult from the laws of formal logic (2). 
 
 PSYCHOLOGICAL FINDINGS MODIFY METHODS OF 
 TEACHING THE MECHANICS OF READING 
 
 The studies of Cattell (3) at Leipzig in 1886 
 on the perceptual process resulted in producing 
 strong evidence that, as a general rule, reading proceeds 
 not by successive letters, but rather by word-, phrase-, 
 and sentence-wholes. This conclusion was based upon 
 the fact that the time required to perceive a whole word 
 and, in some cases, whole phrases and short sentences, 
 was no greater than the time required to perceive a 
 single letter. Consciousness grasped whole words, 
 phrases, and short sentences as units hi the perceptual 
 process. Goldscheider and Mueller's investigation 
 (4) showing that an enlargement of the perceptual span 
 resulted when meaningless geometrical lines were ar- 
 ranged in a meaningful geometrical figure, such as a
 
 4 SILENT READING 
 
 rectangle, likewise offered corroborative evidence. 
 But the convincing confirmation of the above con- 
 clusion was presented when Erdmann and Dodge 
 (5) after many tachistoscopical exposures, found that 
 "words may be recognized when the individual letters 
 are too small to be identified or when parts of the word 
 are out of the range of clear vision." 
 
 Along with the wealth of corroborative data from 
 similar experimental investigations, the above studies 
 show that the assumption underlying the traditional 
 alphabetical method, namely, that the reading process 
 proceeds by successive letters, and not by word-, 
 phrase-, and sentence-wholes, is without scientific 
 foundation. Showing indirectly, therefore, that the 
 word and the sentence methods more closely approxi- 
 mate the perceptual process, these findings are largely 
 responsible for the discard of the time-honored alpha- 
 betical method the result of tradition and crude em- 
 piricism and the adoption of the word and sentence 
 methods with consequent greater fruitfulness by the 
 more progressive schools. These and similar studies, 
 conducted under the strong stimulus of the new science 
 of experimental psychology, have exercised a powerful 
 influence upon the teaching of reading and are largely 
 responsible for the fundamental and far-reaching 
 changes that have been wrought in the technique of 
 teaching the mechanics of reading during the last 
 forty years. They have shown, furthermore, the 
 necessity of submitting all the traditional methods
 
 INTRODUCTION 
 
 and conventional assumptions underlying the teaching 
 of reading to critical experimental examination. 
 
 MOVEMENTS OF THE EYES DURING THE READING 
 PROCESS 
 
 Probably not less rich in promise of supplying 
 valuable data for the development of types or methods 
 of effective training in speed and comprehension in 
 silent reading, in the stages subsequent to the mastery 
 of the mechanics of reading than the above mentioned 
 studies of the perceptual process, are the experimental 
 investigations into the nature of the motor behavior 
 of the eyes during the reading process. The origin of 
 these investigations, like those concerned with the 
 nature of the perceptual process, dates largely from 
 the rise of the science of experimental psychology in 
 1870. 
 
 The history of the evolution of the mechanical 
 apparatus necessary for the experimental study of eye- 
 movements has been narrated by Huey (6), who him- 
 self became one of the most prominent investigators 
 through devising and perfecting a mechanical appara- 
 tus which registered, with some degree of success, the 
 movements of the eyes during the reading process. 
 More recently, W. A. Schmidt (7) has sketched in 
 briefer fashion the evolution of this equipment, supple- 
 menting Huey's account with the story of the develop- 
 ments down to 1917. The present investigation 
 actually utilized the mechanical apparatus that was
 
 6 SILENT READING 
 
 finally evolved to photograph eye-movements during 
 reading. Since the types of training for the develop- 
 ment of efficiency in silent reading, which will be pre- 
 sented in this book, are based very largely upon the find- 
 ings of the experimental investigations of the behavior of 
 the eyes during reading, as well as upon the tachisto- 
 scopic studies of the perceptual process, it is thought 
 that a brief outline of the chief stages of the development 
 of this apparatus will help to supply the necessary 
 background for any who may not be familiar with it. 
 
 Javal (8) of the University of Paris in 1879, employ- 
 ing direct observation by means of a mirror, in contrast 
 to the after-image method of Volkmann and Lamansky, 
 was the first to discover that the passage of the eyes 
 across a line in reading is not a continuous movement, 
 but is broken into a number of pauses and movements. 
 In 1897 Erdmann and Dodge (5) in their experimenta- 
 tion at the University of Halle, besides the mirror 
 method of observation, introduced to advantage the 
 use of the telescope. Lamare (9), working in conjunc- 
 tion with Javal, fastened a microphone to the upper 
 eyelid, thus enabling him to determine, in a general 
 way, from the sounds, the number of fixations in read- 
 ing a given line. Doctor Ahrens (10), at the University 
 of Rostock in 1891, fastened a small ivory cup to the 
 cornea of the eye and sought by means of a rod at- 
 tached to the cup to have the eye movements recorded 
 on a smoked drum. Though unsuccessful, the attempt 
 was instrumental in suggesting to subsequent investi-
 
 INTRODUCTION 7 
 
 gators, a method which, if perfected, would yield rich 
 results. A plaster of Paris cup was later substituted 
 for the ivory cup by Professor Delabarre (11) of Har- 
 vard. 
 
 It remained for Huey, however, to perfect this general 
 type of apparatus and to secure permanent records of 
 eye-movements. After considerable experimentation 
 in directly photographing the eyes, Dodge (13) finally 
 succeeded in photographing "the movement of a bright 
 vertical line as it was reflected from the cornea," thus 
 giving a fairly reliable record of eye-movements un- 
 disturbed by the unnatural condition of a cup or other 
 device pressing upon one of the eyes. Dearborn (14) 
 introduced the use of double lenses, thus enabling him 
 to obtain records of binocular reading. Judd (15) 
 further improved the apparatus by employing a kine- 
 toscope double camera, thus enabling records of eye- 
 movements in both the horizontal and vertical planes 
 to be secured. Various other refinements have been 
 introduced by Freeman, C. T. Gray, and W. A. Schmidt 
 while experimenting at the University of Chicago. 
 
 The evolution of this mechanical . apparatus has 
 rendered possible the carrying out of many investiga- 
 tions which have thrown considerable light on such 
 points as the number of fixations per line, the duration 
 of the fixations, the length of the eye-movements, and 
 the relative time consumed in pauses and movements. 
 Only those investigations will be referred to here 
 which have in some way afforded clues for the develop-
 
 8 SILENT READING 
 
 ment of the types of training that will be described 
 later, or which have presented problems upon which 
 the present study may be expected to yield additional 
 data. Since the total time spent in reading consists 
 of the time consumed by the eye-movements, plus the 
 time required by the fixations, an investigation seeking 
 to economize the length of time required in reading 
 will naturally be concerned in determining at which 
 of the constituent time-consuming processes move- 
 ments or fixations the economy can be most satis- 
 factorily effected. 
 
 /. Inter fixation Movements 
 
 Using the after-image method, Javal, the discoverer 
 of the discontinuous movement of the eyes in reading, 
 estimated the velocity of the interfixation movements 
 to be so great as to prevent clear vision. On the basis 
 of certain investigations, Erdmann and Dodge (5) 
 calculated that only from one-thirteenth to one-twenty- 
 fourth of the total time required for the eyes to cross 
 a given line was consumed by the interfixation move- 
 ments. Quite the reverse of the popular conception 
 that had long prevailed that the total time spent in 
 reading was consumed during the passage of the eyes 
 across the printed line, Erdmann and Dodge found that 
 almost the entire time was taken up by the pauses, or 
 fixations of the eyes at various points in their journey 
 across the line. The investigations made by Huey (12), 
 Dodge and Kline, Dearborn, and more recently by C. T.
 
 INTRODUCTION 9 
 
 Gray and W. A. Schmidt, in measuring the rapidity 
 of the interfixation movements, have conclusively 
 confirmed the findings of Erdmann and Dodge. With 
 such lightninglike velocity do the eyes move from 
 fixation-point to fixation-point, that it is now generally , 
 agreed that fusion of stimuli occurs, so that, as far as/ 
 purposes of clear perception are concerned, the eyes/ 
 are practically "blind" during these movements. 
 
 The experimentation of Holt (16), leading him to the 
 conclusion that the movements of the eyes produced 
 "a momentary visual central anaesthesia," confirmed 
 the findings of Dodge, which were again corroborated 
 by Dearborn (17). From these findings, it is evident 
 that the interfixation movements, consuming, as they 
 do, such a small fraction of the total reading time, 
 do not constitute the processes in the reading complex 
 wherein any great economy of time can be effected. 
 The question is, however, whether the development 
 of a habit of increased rapidity of silent reading 
 would measurably affect the speed of the movement of 
 the eyes from fixation to fixation, as well as the return 
 sweep from the end of one line to the beginning of the 
 following, or whether, in spite of a considerable increase 
 in speed of reading, the rate of these movements would 
 remain unchanged. By measuring the rate of 
 these movements before giving the training which 
 resulted in the establishment of habits of greatly 
 increased speed in silent reading, and then by 
 measuring the rate of these movements subsequent
 
 10 SILENT READING 
 
 to the acquisition of this increased rapidity in reading, 
 data have been secured which are of value in the 
 formulation of at least a tentative answer to the above 
 query. 
 
 //. The Fixation-Pauses 
 
 The processes of the reading complex which consume 
 the bulk of the time, in fact from twelve-thirteenths 
 to twenty-three-twenty-fourths of the total reading 
 time, are the fixations or pauses which the eyes make at 
 various points in their journey across the printed line. 
 These pauses, the duration and number of which have 
 been disclosed to us by experimentation, constitute 
 the real "core" of the physiological process under- 
 lying the mental act of reading. Since the eyes 
 while moving are incapable of clear vision, owing 
 to the velocity with which they move from 
 fixation to fixation and sweep back from the end of one 
 line to the beginning of the next, it follows that what- 
 ever clear vision occurs must take place during the 
 brief momentary fixation-pauses. Upon these fixation- 
 pauses a large number of painstaking investigations 
 requiring an elaborate equipment have been focused, 
 with a view to determining chiefly their location, 
 duration, and number per line. 
 
 The investigations of Dodge, of Huey, and especially 
 of Dearborn have shown that fixations occur in every 
 part of a word as well as between words, and are ap- 
 parently subject to no definite law of occurrence. It
 
 INTRODUCTION 11 
 
 is the number and the duration of pauses that are of 
 chief interest in a study that seeks to accelerate the 
 reading rate. For the total reading time may be said 
 to be the product of the number of pauses, multiplied 
 by their average duration, plus, of course, the very 
 slight fraction of the total time one-thirteenth to 
 one-twenty-fourth consumed by the eye-movements. 
 
 (a) Number of Fixation Pauses. Investigations have 
 revealed in this case, as elsewhere in educational psy- 
 chology, the existence of great individual differences. 
 Dearborn reports that his subjects in reading silently the 
 same selections from a newspaper averaged from 3.0 to 
 7.1 pauses per line. In number of words perceived per 
 fixation, the average ranged from 1.9 to 1.0 words. 
 W. A. Schmidt reports a range of averages from 4.1 
 to 10.8 pauses per line in silent reading, and 6.1 to 11.5 
 pauses per line in oral reading. For Schmidt's adult 
 silent readers, the averages ranged from 2.15 to 0.93 
 words per fixation, and in the case of oral reading the 
 same subjects averaged from 1.52 to 0.87 words. It is 
 obvious that the number of pauses per line is depend- 
 ent upon the number of words perceived per fixation 
 the fewer the number of pauses the more extensive must 
 be the perceptual span. 
 
 (6) Duration of Fixation-Pause. Of equal impor- 
 tance in determining the total amount of time consumed 
 in reading, is the duration of the pauses. The number 
 and the average duration of the pauses form, as it were, 
 the multiplier and the multiplicand of approximately
 
 12 SILENT READING 
 
 the total reading time product, which latter cannot, 
 of course, be computed, if either of the preceding 
 are undetermined. Only comparatively recently has 
 the experimental apparatus been evolved to secure 
 accurate measurements of the duration of pauses. 
 Huey reports that his subjects averaged approximately 
 185<7. * Dearborn, with more accurate apparatus, re- 
 ports a range in averages for his five subjects from 160.8<r 
 to 401. 9cr. Schmidt (18) reports a range of averages 
 for his silent readers from 214<r to470<r, his elementary 
 pupils averaging from 264<r to 438o- 
 
 Types of training to develop habits of speed which 
 result in a considerable reduction of the total reading 
 time obviously must either diminish the number of 
 pauses or reduce their duration, or else effect a decrease 
 in both. At the expense of which one of these processes 
 is the economy of time to be secured? 
 
 From his elaborate experimental investigation of eye- 
 movements, fixation-pauses, etc., Schmidt concludes 
 "that a momentary silent reading at maximal rather 
 than normal rate involves a shortening of the duration of 
 the pauses rather than a lessening of their number. But 
 this leaves unsolved the problem as to what would 
 happen if the subjects were to engage in practice extend- 
 ing over a considerable period of time, gradually forming 
 habits of reading at maximal rate." This he classes as 
 one of the "interesting but as yet unsolved problems" 
 in the psychology of reading. The present study, 
 
 1 Sigma (<r) is ^^ of a second.
 
 INTRODUCTION 13 
 
 through its formulation of types of training for the 
 development of speed in silent reading and through its 
 application of these methods of training to pupils in 
 various grades of the elementary school for a period 
 of two months, has sought to establish just such habits 
 of reading at maximal rates. It is thought that the 
 photographic records of the eye-movements of some 
 of these pupils taken before and after the acquisition 
 of this habit of rapid silent reading, will yield some in- 
 teresting and valuable data upon this particular problem. 
 
 COMPARISON OF SILENT AND ORAL READING 
 
 /. By Means of Photographic Records of 
 Eye-Movements 
 
 The most important of the recent contributions 
 to the experimental study of the reading process 
 have come from investigators working at the 
 University of Chicago. An elaborate experimental 
 study of the motor behavior of the eyes during the 
 reading process has been made by W. A. Schmidt (7). 
 His comparison of the photographic records of oral 
 and silent reading by adults, high school pupils, 
 and elementary school pupils is significant and 
 instructive. His records reveal the physiological basis 
 of the superiority of silent over oral reading in point 
 of both speed and comprehension. Thus, his group 
 of 45 adult persons averaged 1.7 more pauses per line 
 in oral than in silent reading, while the average dura-
 
 14 SILENT READING 
 
 tion per pause was 72.6<r greater in the former than in 
 the latter. This made the average fixation time per 
 line 1119.1er longer for the same subjects when read- 
 ing orally than when reading silently. 
 
 The smaller number of fixations per line and the 
 shorter duration of the fixations may be said to con- 
 stitute, in a general way, the chief physiological basis of 
 the superior speed of silent over oral reading. The clos- 
 est index to the physiological basis of the superior com- 
 prehension in silent reading may be said to be the smaller 
 number of refixations per line. A refixation is a regres- 
 sion of the eyes to an anterior portion of the line, which 
 portion has been already once fixated but which has 
 evidently not been comprehended. The refixation ap- 
 pears necessary to complete the understanding of the 
 line. The group of adults averaged 74 per cent more 
 refixations per line in oral than in silent reading ; the 
 high school pupils 55 per cent more ; and the elementary 
 school pupils 62 per cent more. Schmidt reports that 
 his subjects "read 57 per cent more material silently 
 than orally." 
 
 On the strength of his investigation Schmidt draws 
 the following significant conclusion : 
 
 "There is in evidence in connection with silent reading a 
 very definite tendency towards positive correlation between 
 the rate and comprehension, rapid readers representing a 
 higher quality of comprehension than the slow. No such 
 correlation is in evidence in the case of the oral reading. In 
 fact, there are some indications that an increase in rate may
 
 INTRODUCTION 15 
 
 be inconsistent with good quality of comprehension. The 
 difference between the two types of reading would un- 
 doubtedly have been more marked had the subjects been 
 trained to distinguish between the two. As a matter of fact 
 reading has until recently been taught so poorly that it is 
 extremely hazardous to speak of standards for either type 
 of reading as well as standard differences between the two 
 types. Training in silent reading has been almost unthought 
 of and whatever may have been accomplished in the case of 
 oral reading during the earlier grades has not infrequently 
 been undone because of indifferent practices in connection 
 with the later grades." 
 
 C. T. Gray (19), in an investigation recently con- 
 ducted at the University of Chicago, has contributed 
 additional valuable data on the types of reading 
 ability determined through standard tests and labora- 
 tory experiments. His findings likewise emphasize 
 the superior efficiency of silent over oral reading. 
 The methods employed by him in the experi- 
 mental laboratory to increase the speed and compre- 
 hension of two individual subjects have largely fur- 
 nished the clues for the formulation and development 
 of types of training which have been adapted for class- 
 room procedure under the ordinary conditions of the 
 school. In the eighteenth yearbook, W. S. Gray (20) 
 elaborates these and other factors tending to condition 
 the efficiency of silent reading. Though the types of 
 training to be presented in this book had been 
 formulated before the appearance of the yearbook, 
 it is believed that all of Gray's main suggestions
 
 16 SILENT READING 
 
 have been incorporated in the development of the types 
 of training in the present study. Judd (21) has also 
 greatly stimulated interest in the nature of the reading 
 process by his recent monograph in which he interprets 
 the meaning of the photographic records of eye- 
 movements, fixation-pauses, etc., and points out their 
 diagnostic value to teachers of reading. The present 
 study has attempted to utilize many of his suggestions 
 for increasing efficiency in silent reading. 
 
 //. By Means of Standard Tests 
 
 Significant as are the findings of the psychological 
 investigation and the laboratory experiments in dis- 
 closing the functioning of the various constituent pro- 
 cesses of the reading complex, showing, as they do, 
 the physiological basis of the superior efficiency of 
 silent over oral reading, it is chiefly the mass of investi- 
 gations wherein the efficiency of school products has 
 been measured by the simple standardized tests that 
 have convinced and are still convincing practical 
 school men and women of the superiority of silent over 
 oral reading, both in point of speed and comprehension. 
 The application of a variety of scientific standardized 
 reading tests to pupils in the elementary school, high 
 school, and college has resulted in the accumulation of 
 voluminous data confirming the above conclusions of 
 the experimental laboratory. 
 
 (a) Comparison of Rates. In 1913 Pintner (22) 
 conducted an investigation with twenty-three pupils
 
 INTRODUCTION 17 
 
 in the fourth grade in which he found that the pupils 
 reading silently averaged eight lines more per minute 
 than when reading orally a superiority of 40 per cent 
 in point of speed. Oberholtzer's (23) investigation 
 of the rate of silent and oral reading, as shown by test- 
 ing 1800 elementary school pupils in grades three to 
 eight, yielded superior averages for silent reading in 
 all of the six grades examined. Starting in the third 
 grade with an average superiority of 0.2 more words 
 per second when reading silently than when reading 
 orally, the superiority of silent reading in point of speed 
 continued to manifest itself increasingly through all the 
 grades, until, in the eighth grade, the pupils read ap- 
 proximately one complete word more per second silently 
 than they did orally the averages being 3.9 and 4.8 
 respectively. 
 
 The investigation of Meade (24) with 100 sixth grade 
 pupils, and especially the extensive study reported by 
 Judd covering all the grades in 44 different schools, 
 prove the greater rapidity of silent over oral reading 
 beyond all shadow of doubt. This conclusion, after 
 all, is only what one would naturally be led to expect. 
 For vocalization involving rather elaborate movements 
 of the lips, tongue, vocal chords, and general throat 
 mechanism is a much slower process than visualization. 
 Consequently, in oral exhibition, the reader is obliged 
 to retard the rate of visualization in order that the 
 more tedious process of vocalization may keep pace 
 with it. Synchronization of the vocal and visual
 
 18 SILENT READING 
 
 records of oral readers, as secured through an elaborate 
 mechanical equipment in an investigation, the results 
 of which are soon to be published by G. T. Buswell in 
 his dissertation for the doctorate at the University of 
 Chicago, offer corroboration of the above conclusion. 
 
 (6) Comparison of Comprehension. Not only is silent 
 reading more rapid than oral reading but it is also 
 superior in point of comprehension of the matter read. 
 Though oral reading secures more time for the process 
 of comprehension, yet the division of attention which 
 oral exhibition necessitates, drawing part to the task 
 of articulation and enunciation, and part to the under- 
 standing of the content, seems more than to counter- 
 balance the first apparent advantage. The effort to 
 enunciate correctly, to stress in accordance with the 
 laws of the rhetorician, renders difficult that strong 
 concentration of attention upon the content side which 
 is the sine qua non of effective thought mastery, and 
 which is more readily obtainable in simple visualiza- 
 tion. 
 
 The results of testing by means of various forms 
 of reproduction of the matter read serve to ree'nforce 
 the above conclusion. Meade reports that all the 
 sixth grade classes reproduced more effectively when 
 they read silently than when they read orally. Pintner 
 reports that a greater percentage of points were re- 
 produced by his fourth grade pupils when they read 
 silently. As a result of their investigation in 1916, 
 Pintner and Gilliland (25) draw the following conclu-
 
 INTRODUCTION 19 
 
 sion : "the silent reading of the adult is quicker than 
 the oral reading and at the same time the number of 
 ideas remembered is slightly greater, certainly much 
 greater per unit of time." 
 
 (c) Correlation between Speed of Reading and Quality 
 of Comprehension. Besides demonstrating the greater 
 efficiency of silent over oral reading in point of speed 
 and comprehension a number of investigations have 
 shown the existence of a positive correlation between 
 rapidity of reading and quality of comprehension. 
 As far back as 1897, Quantz's (26) study led him to 
 conclude that the rapid readers average approximately 
 37 per cent superiority over slow readers in the quality 
 of their work. He observed then, what later researches 
 have confirmed, that the memory in the case of rapid 
 readers seems to be more accurate, introducing in the 
 reproduction of the matter read less extraneous matter 
 than in the case of slower readers. As a result of his 
 study reported in 1911, Hendricks (27) concluded: 
 "in the percentage of thought reproduced the rapid 
 readers excel, giving 91 per cent of the thought as com- 
 pared with 76 per cent reproduced by slow readers." 
 After an extensive study of 1831 pupils Judd (28) re- 
 ports : "These figures serve to emphasize the fact that 
 good readers are usually not slow and poor readers are 
 usually not fast. It will be shown later, in detail, that 
 rate is parallel in a majority of cases with general effi- 
 ciency in reading. For the moment, it is enough to 
 point out the chief reason why ability to read rapidly
 
 20 SILENT READING 
 
 is a fair measure of the mastery which the reader has of 
 the printed page. The poor reader is one who is unable 
 to pass readily from the printed symbol to the meaning. 
 For the poor reader the mere mechanical processes are 
 obstacles and he loses time in trying to perform the 
 preliminary acts which are necessary before he can com- 
 prehend the passage. In the case of the good reader, 
 on the other hand, the mechanics of the process are 
 very fluent and rapid. The proficient reader has mas- 
 tered the words and moves on without hesitation to the 
 meaning." W. S. Gray (20) thus aptly epitomizes the 
 experimental findings on this point : "The rapid reader 
 is usually more efficient than the slow reader." 
 
 SILENT READING THOUGH MOST IMPORTANT Is 
 NEGLECTED BY THE SCHOOL 
 
 Though investigations have shown the superior 
 speed of silent reading and its greater efficiency as a 
 basic instrument for the gathering of thought and ideas 
 from the printed symbols, yet one finds that the schools 
 not only teach reading as an oral exercise but stress 
 that side to the almost complete exclusion and dis- 
 regard of silent reading. There is no place in the cur- 
 riculum in the overwhelming majority of our schools 
 for silent reading. But may not this apparent anomaly 
 find its justification on the grounds of the greater use 
 to which oral reading is put? Frequency or ex- 
 tensiveness of use has come to be considered probably 
 the strongest criterion determining not only the inclu-
 
 INTRODUCTION 21 
 
 sion or exclusion of the various parts of a subject 
 but also the relative stress to be placed upon them. 
 This is on the very reasonable ground which finds 
 expression in the query: "Why teach something that 
 is never, or at most, seldom to be used?" It is pre- 
 cisely here that the case for silent reading scores its 
 strongest point. How often does the average person 
 have occasion for oral exhibition ? The writer has been 
 present in several classes in which the students were 
 asked to recall one instance of oral reading outside of 
 the school in the past month. The instance was not 
 forthcoming. 
 
 The plain facts are that the occasions requiring 
 oral exhibition from either the adult or the pupil are 
 notoriously few in life. Practically all the reading 
 of the average person is done silently. He reads the 
 newspaper, the magazine, the scientific journal, the 
 novel, not for purposes of oral display, but for the sole 
 purpose of gathering the thoughts and ideas contained 
 therein. Consequently the training in oral reading 
 which lasts throughout all the grades of the elementary 
 school seldom, if ever, functions in actual life. But 
 for the silent type of reading which a person is obliged 
 to use almost daily during his school career as well as 
 subsequently, the school has made no attempt to train. 
 And this in spite of the fact that the school is supposed 
 to train for the subsequent activities of actual life, and 
 to have its own activities approximate, as closely as 
 possible, the real conditions of life.
 
 22 SILENT READING 
 
 STRESS ON ORAL READING DETRIMENTAL TO 
 EFFECTIVE SILENT READING 
 
 Failing to perceive the necessity of preparing the 
 pupil for this important activity in which men and 
 women in every vocation of life must engage, the school 
 not only fails to train the pupil for this kind of work, 
 but actually builds up fixed habits of word-pronuncia- 
 tion and articulation of a slow, plodding character 
 which many investigators allege are almost ruinous to 
 the formation of the opposite habits of rapid, effective, 
 silent reading, of the meaningful type. While it is true 
 that clear and distinct articulation and accurate pro- 
 nunciation are necessary for correct speech and are 
 acquirements of no doubtful value, for which the school 
 must train, yet it is necessary to recognize that correct 
 enunciation is not synonymous with reading, which 
 consists essentially in the gathering of thought from the 
 printed page. It is quite a distinct process, and should 
 be taught as such, in a period set aside for the specific 
 task of word-pronouncing, syllabication, articulation, 
 and the explanation of the phonic properties of the 
 various letters. 
 
 The distinction between these two types of work is 
 fundamental and cannot be emphasized too strongly. 
 The failure of many schools to recognize it has led 
 them to the confusing assumption that in teaching 
 articulation, the phonic properties of words, and accu- 
 rate word-pronouncing they are teaching reading.
 
 INTRODUCTION 23 
 
 The result of this stress on the oral side has been to 
 focus the attention largely upon the pronunciation, 
 with a corresponding loss to the comprehension. In- 
 deed, the number of pupils whose minds are found 
 to be practically blank when stopped suddenly 
 and asked to tell what they had just been reading 
 aloud, has seldom failed to astonish the teachers 
 who have occasionally employed this simple checking 
 device. It shows how detrimental alike to speed and to 
 effective comprehension is the exclusive teaching of the 
 oral aspect of reading. 
 
 While there is every reason to believe that the teach- 
 ing of reading has made many strides forward since the 
 days of Horace Mann, yet it would seem that some 
 points of similarity still persist, as is evident from the 
 following description of the results of the teaching of 
 reading as reported by him to the Massachusetts 
 Board of Education in 1838. The passage is quoted 
 by Judd, as follows : 
 
 " I have devoted especial pains to learn with some degree of 
 numerical accuracy how far the reading, in our schools, is an 
 exercise of the mind in thinking and feeling and how far it is 
 a barren action of the organs of speech upon the atmosphere. 
 My information is derived principally from the written 
 statements of the school committees of the respective towns, 
 gentlemen who are certainly exempt from all temptation 
 to disparage the schools they superintend. The result is, 
 that more than eleven-twelfths of all the children in the 
 reading classes in our schools do not understand the mean- 
 ing of the words they read ; that they do not master the sense
 
 24 SILENT READING 
 
 of the reading lesson, and that the ideas and feelings intended 
 by the author to be conveyed to, and excited in, the reader's 
 mind, still rest in the author's intention, never having yet 
 reached the place of their destination. And by this is not 
 meant that the scholars do not obtain such a full compre- 
 hension of the subject of the reading lessons, in its various 
 relations and bearings, as a scientific or erudite reader would 
 do, but that they do not acquire a reasonable and practical 
 understanding of them. It would hardly seem that the 
 combined efforts of all persons engaged could have accom- 
 plished more in defeating the true objects of reading." 
 
 Not very different in this respect from the days of 
 Horace Mann are the present results of the teaching 
 of reading according to the conclusion of Judd, reached 
 after an exhaustive study of the present status of read- 
 ing in our schools : 
 
 " Many a pupil leaves school equipped with the mechanical 
 ability to read words, but utterly unacquainted with the 
 possibility of interpretation. School reading has been a 
 formal ceremony for the pupil. He has formed the habit of 
 thinking that the words have been adequately dealt with 
 when they have been sounded. The fault is with the school's 
 selection of reading matter and with the school's emphasis 
 on mere mechanical perfection in oral reading. ... The 
 present practice of continuing drills in the mechanics of read- 
 ing through the elementary school undoubtedly retards 
 pupils rather than helps them." 
 
 Thorndike (31) comes to a similar conclusion after 
 an exhaustive analysis of errors made by children con- 
 cerning the matter they had just read :
 
 INTRODUCTION 25 
 
 " In school practice it appears likely that exercises in silent 
 reading to find the answers to given questions, or to give a 
 summary of the matter read, or to list the questions which 
 it answers should in large measure replace oral reading. The 
 vice of the poor reader is to say the words to himself without 
 actively making judgments concerning what they reveal. 
 Reading aloud or listening to one reading aloud may leave 
 this vice unaltered or even encouraged. Perhaps it is in their 
 outside reading of stories and in their study of geography, 
 history, and the like, that many children really learn to read." 
 
 Moreover, the habits of slow mechanical reading, 
 resulting from a grossly misplaced emphasis and from 
 defective methods of teaching reading, last not only 
 during the pupil's school career, but tend to persist 
 till his dying day. Undoubtedly many a reader need 
 go no farther than his own case to find a concrete illus- 
 tration of the following generalization made by Huey : 
 
 "Doubtless many of us dawdle along in our reading at a 
 plodding pace which was set and hardened in the days of 
 listless poring over uninteresting tasks or in imitation of the 
 slow reading aloud which was so usually going on either with 
 ourselves or with others in the school." 
 
 EXCLUSIVE TEACHING OF ORAL READING AN 
 ANACHRONISM 
 
 In view of the above findings, the question might 
 well be raised : Why then does the school devote the 
 entire reading period in practically all the grades to 
 the teaching of oral exhibition, as though it were the 
 sole instrument of value, to the complete exclusion and
 
 26 SILENT READING 
 
 absolute disregard of silent reading? In other words, 
 why does the school devote itself almost exclusively, 
 so far as the reading period is concerned, to the setting 
 up of habits of word-pronouncing and the vocaliza- 
 tion of the printed page, which habits are very seldom 
 if ever used outside of the formal reading period of the 
 school, and neglect almost entirely the formation of the 
 habits of far greater value habits of rapid, effective, 
 silent reading, the type that is quick to assimilate the 
 thought from the printed symbols, the kind that is 
 called into function every day of the pupil's school 
 career and during practically every day of his subse- 
 quent life? 
 
 The answer to the above question is twofold. First, 
 the majority of the investigations reported above are 
 of comparatively recent date. This is especially true 
 of the mass of investigations measuring the efficiency 
 of reading by means of the standardized tests. The 
 bulk of them have occurred only within the past decade. 
 The practical significance of the findings is only now be- 
 ginning to filter through into the vast army of practical 
 school men and women superintendents, principals, 
 and teachers. Educational practice must always lag 
 somewhat behind educational theory, as well as behind 
 the findings of scientific research. When the results 
 of scientific studies become disseminated sufficiently 
 broadcast throughout the ranks of practical school 
 people, the schools, especially those directed by pro- 
 gressive educators, will not fail to modify their present
 
 INTRODUCTION 27 
 
 practice and to incorporate into their mode of procedure 
 the new methods and the relative stress which spring 
 directly from the certain findings of scientific research. 
 
 The second reason for the almost exclusive teaching 
 of oral reading and the neglect of silent reading is the 
 fact that, as yet, no systematic methods for the teach- 
 ing of silent reading have been worked out. There are 
 as yet no types of training, in rapid silent reading, 
 which have been developed and adapted to the actual 
 working conditions of the classroom. The experiments 
 to train in effective silent reading thus far have been 
 largely concerned with one or two, or at most a small 
 group of subjects, and the experimenters have relied to 
 a large extent upon the technical apparatus of the 
 psychological laboratories, such as the tachistoscope, 
 to produce the desired results. Needless to say, the 
 general use of such apparatus by the schools, at the 
 present time at least, is both impractical and impos- 
 sible. Indeed, the bulk of the instruments, such as the 
 tachistoscope, the fall chronometer, and the various 
 kinds of perceptual apparatus, have been constructed 
 for the use of but a single subject, and extensive modi- 
 fications would be necessary before they would be suit- 
 able for group usage. 
 
 In sharp contrast to the paucity or rather complete 
 lack of classroom methods for the teaching of silent 
 reading, numerous methods and systems for the teach- 
 ing of the mechanics of oral reading have been most 
 carefully worked out. Occupying for centuries a domi-
 
 28 SILENT READING 
 
 nant place in the elementary curriculum, it has received 
 the thought and study of many minds. It is but nat- 
 ural that teachers will continue to use the present 
 methods of teaching reading until other methods have 
 been devised which not only shift the emphasis to that 
 phase of the teaching of reading where it will produce 
 the best results, but which are suitable for classroom 
 use as well. 
 
 Viewed in the light of its historical genesis, the 
 teaching of reading as an oral subject has come down 
 into the present curriculum as a heritage from the an- 
 cient days when oratory, speech, and vocal instructions 
 were not only the most important but almost the only 
 instruments for the dissemination of thought and 
 knowledge. Before the invention of printing, manu- 
 scripts, involving as they did so much and such tedious 
 labor, were necessarily extremely few in comparison 
 with the products of the present day press, and could 
 not constitute the important medium of instruction 
 which books do at the present day. While print- 
 ing has revolutionized the conditions for the spread 
 of ideas and the general means of education, yet the 
 school under the sway of custom and tradition has 
 preserved almost intact the curriculum of reading of the 
 centuries past, with its sole stress upon oral exhibition, 
 in spite of the fact that the objective conditions of life 
 for which that curriculum was constructed now no 
 longer exist. The subject of reading, like all other 
 branches of the curriculum, tends, with the lapse of
 
 INTRODUCTION 29 
 
 time, to become petrified a fossil of ancient prac- 
 tices unless it is subjected to constant revision, 
 modification, and adaptation to meet the correspond- 
 ing changes in the objective conditions of modern life. 
 The need of a readjustment, of a substantial shift of 
 emphasis in this portion of the school's curriculum, 
 to meet the changed external conditions of the modern 
 world is both obvious and urgently imperative. 
 
 SHIFT EMPHASIS FROM ORAL TO SILENT READING 
 
 This needed readjustment can be effected by shifting 
 the emphasis from oral to silent reading in other 
 words, by teaching silent reading. This is the conclu- 
 sion reached by practically every investigator in this 
 field. As a result of their investigations of the compara- 
 tive efficiency of the oral and silent reading of the pupils 
 in the elementary schools, in the high school, and in 
 the college, Pintner and Gilliland draw the following 
 significant conclusion : 
 
 " Thus it would appear that silent reading is undoubtedly 
 the more economical besides being the method best adapted 
 to the ordinary activities of life, since the vast majority of 
 our reading is silent. This being the case, we are forced to 
 raise the pedagogical question, and ask why it is that so much 
 more attention is given to oral reading than to silent reading 
 in our schools?" 
 
 Probably no one has brought out more strongly or 
 convincingly the pressing need for a radical shift of 
 emphasis from oral to silent reading than has Judd.
 
 30 SILENT READING 
 
 After pointing out the mass of evidence from the photo- 
 graphic records of eye-movements and from the applica- 
 tion of standardized reading tests, which show the un- 
 mistakable superiority of silent reading, Judd concludes : 
 
 "Enough (evidence) has been brought together to make 
 it certain that in the middle grades there is a change in rela- 
 tion between oral language and reading which ought to be 
 recognized by a radical change in methods of instruction. 
 Oral reading should give way to silent reading and phonic 
 analysis should give place to word analysis. Meanings should 
 be emphasized and not the mechanical pronunciation of words. 
 
 "Many schools have not recognized the demand for a new 
 type of instruction in reading from the fourth grade on and 
 as a result have seriously injured the development of pupils. 
 . . . The oral methods which are legitimate in the lower 
 grades become inappropriate with the growth in fluency and 
 range of recognition. ... It is a mistake to jeopardize 
 the child's independent, fluent, silent reading in the fourth 
 grade by insisting during this period on the usual oral exer- 
 cises. . . . The conscientious teacher supplied with a read- 
 ing book and a period in the program carries on the well- 
 known reading farce in the vain hope that the effects of un- 
 successful teaching will be overcome by a liberal application 
 of the same methods that produced the difficulties. 
 
 "Teachers ought to recognize with clearness the fact that 
 in the upper grades silent reading is the really useful type of 
 reading. They ought to understand that pupils outgrow 
 oral training just as infants outgrow creeping when they 
 learn to stand up and walk." 
 
 The above quotation from Judd may be said to 
 represent fairly well the consensus of opinion of all the 
 investigators and careful students of this subject.
 
 CHAPTER II 
 THE PROBLEM 
 
 EVEN after the need for the teaching of effective 
 silent reading has come to be generally recognized by 
 teachers, they will still be hampered and restrained by 
 the lack of a satisfactory technique, the lack of suitable 
 methods for the teaching of rapid silent reading. Up 
 to the present writing, there has been practically no 
 systematic attempt to utilize the wealth of data afforded 
 by the experimental investigations and the laboratory 
 researches into the nature and mode of functioning of 
 the various constituent processes of the reading com- 
 plex, for the formulation of methods of training in ef- 
 fective rapid silent reading which are adapted to a 
 whole class under the ordinary conditions of the 
 schoolroom. This appears to the writer to be the out- 
 standing problem, of urgent, practical importance, 
 in the entire field of the psychology and pedagogy of 
 reading. Unless the findings of the psychological 
 laboratory and of the experimental investigations can 
 be adapted to the development of practical methods 
 of classroom procedure, the teaching of reading will 
 continue to be in the future, as it has been in the past, 
 
 31
 
 32 SILENT READING 
 
 the slow, mechanical, oral type, unbenefited and un- 
 improved by all the discoveries of scientific research. 
 
 To construct, for the teacher in the classroom, 
 types of training in effective rapid silent reading, 
 based upon the findings of experimental science, con- 
 stitutes the problem in this investigation. Efficiency 
 in silent reading depends upon both speed and com- 
 prehension. While means have been taken to safe- 
 guard the comprehension, the chief stress has been 
 placed upon methods of increasing the rapidity of silent 
 reading. In an age in which the best thought is com- 
 mitted to writing, and in which the printed symbol has 
 supplanted the spoken word as the chief instrumental- 
 ity for the dissemination of knowledge, the ability to 
 read with speed and yet with understanding is of funda- 
 mental importance. 
 
 Since the literature in practically every field is now 
 so voluminous that none but the fastest can hope to 
 cover any considerable portion of its extent, it may be 
 safe to say that, other things being equal, a person's 
 efficiency in his profession will be largely proportional 
 to the rapidity with which he can gather the thought 
 from the printed page. An increase of speed in silent 
 reading will, therefore, not only effect an appreciable 
 economy of time and effort in the work of education, 
 but it will increase the efficiency of both the pupil and 
 the adult. It will, moreover, satisfy a keenly felt human 
 want. For there are probably few persons who have 
 not, at some time or other, felt themselves aglow with
 
 THE PROBLEM 33 
 
 the insatiable longing to penetrate deeper into their 
 chosen field of labor, to make themselves masters in their 
 field by assimilating the written thought and knowledge 
 of the masters who have gone before. A considerable 
 increase in speed of reading is the one effective instru- 
 ment which will aid them in the realization of their 
 desire. 
 
 The writer feels that the problem is, therefore, not 
 only of academic interest, but is one of real, intense, 
 practical importance and functional value. If success- 
 ful in the construction of types of training which will 
 actually increase the speed of reading to a considerable 
 extent, even doubling or tripling its rate, the writer 
 feels that the study will have served as a useful instru- 
 ment in bringing the results of over half a century of 
 scientific experimentation to bear directly upon the 
 actual teaching of reading, rendering the latter im- 
 mensely more fruitful and effective for the children of 
 America. 
 
 The investigation seeks first to determine the various 
 factors conditioning the development of speed in silent 
 reading, to ascertain the comparative influence of the 
 different factors in accelerating the reading rate, and 
 to combine the important factors into a formulation 
 of successful types of training in silent reading. While 
 the investigation seeks mainly to furnish a practical 
 answer to the question, implicit in the minds of 
 thousands, How can I read more rapidly and yet 
 with understanding? there are a number of minor
 
 34 SILENT READING 
 
 problems upon which it is expected to yield additional 
 data. 
 
 Some of the questions upon which this study fur- 
 nishes information for the development of at least 
 tentative answers are the following : How does an 
 increase in the speed of reading affect the compre- 
 hension? In other words, Is the comprehension 
 thereby decreased, unaffected, or increased? Do the 
 individual differences in rate tend to decrease or to 
 become more prominent as a result of training in silent 
 reading? What grades show the greatest susceptibil- 
 ity to improvement in rate of silent reading ? At what 
 stages in the progress through the grades should speed 
 of silent reading receive the greatest stress? Does 
 speed of reading seem to be conditioned chiefly by a 
 native neurological factor whose limits of possible 
 variation or modification are rather definitely fixed ? 
 If so, may the rate at which the neurological factor 
 functions in the eye movements or the visualization 
 of the printed page be determined by a reaction-time 
 test? Does the quality of comprehension tend to 
 persist in spite of changed mental attitudes, assumed 
 in reading for different purposes? Do the present 
 norms for rate of silent reading for the various grades 
 still possess validity after training has been given in 
 rapid silent reading? What are tentative standards 
 for rate for grades in which pupils have been trained 
 in rapid silent reading? 
 
 Data will also be presented concerning some of the
 
 THE PROBLEM 35 
 
 problems previously mentioned, such as the effect of 
 habits of increased speed upon the perceptual span, 
 upon the duration and number of fixation-pauses per 
 line, and upon the rate of the interfixation movements 
 and the return sweeps data which will be of interest 
 chiefly to students of the technical psychology of 
 reading. Some suggestions for the improvement of the 
 Courtis Silent Reading tests, employed in this investi- 
 gation, have also resulted from the study. 
 
 The succeeding chapters will accordingly treat of 
 (1) the factors affecting the development of speed in 
 silent reading, (2) their formulation into types of train- 
 ing, (3) the results of the application of such methods, 
 (4) the interpretation of the results, (5) the effect 
 upon the motor behavior of the eyes, and (6) the con- 
 clusions.
 
 CHAPTER III 
 
 FACTORS INFLUENCING THE RATE OF SILENT 
 READING 
 
 WHAT are the factors affecting the rate of silent 
 reading? With the single exception of Quantz's ex- 
 periment (26) during 1897 in the psychological labora- 
 tory at the University of Wisconsin, there has been no 
 other investigation of an experimental character, to 
 the writer's knowledge, which has made the answer 
 to the above query the primary object of research. 
 There have been, however, numerous investigations 
 which have ascertained the rate of silent reading of 
 various groups of pupils. These pupils have often 
 displayed different levels of achievement, as well as 
 different methods of procedure and different charac- 
 teristics in their reading, so that deductions have oc- 
 casionally been drawn as to the factors which condi- 
 tioned the resultant rates of achievement. These have 
 been made chiefly by the process of correlation, rather 
 than by the rigorous isolation of a single factor and 
 then, under controlled conditions, measuring its po- 
 tency. The latter method, while most fruitful in the 
 physical sciences, is not possible in the complex type 
 of mental activity required in a single act of reading. 
 
 36
 
 THE RATE OF SILENT READING 37 
 
 Correlations have been found sufficiently high, 
 however, to indicate in a general way the influence 
 of certain factors in accelerating or in retarding the 
 reading rate. The factors which will be enumerated 
 are the by-products, as it were, of numerous investiga- 
 tions of various phases of reading, conducted both in 
 the psychological laboratory and in the schoolroom. 
 It is thought that all the factors which investiga- 
 tions have indicated to be of appreciable consequence 
 in conditioning the rate have been included in the 
 subsequent enumeration. The present study thus en- 
 deavors to utilize all the data which previous investi- 
 gations have afforded on this particular phase of read- 
 ing. These factors have been incorporated into various 
 types of training to increase the rate. By applying 
 these types of training to hundreds of school children 
 in grades three to eight inclusive, the influence of 
 these factors in accelerating the rate will thus be 
 experimentally tested. 
 
 The following is a list of factors which previous in- 
 vestigations and the general literature on the subject 
 indicate to be of some importance in conditioning the 
 rate of reading. Each factor will be considered 
 separately. 
 
 1. Practice in Rapid Silent Reading 
 
 That practice in a certain line of activity is an im- 
 portant factor in conditioning the quantity and quality 
 of the resultant product is a fact which probably no
 
 38 SILENT READING 
 
 one will deny. This principle of the influence of prac- 
 tice applies to the reading process as it does to other 
 mental activities. Huey's (6) own case is an in- 
 teresting illustration of the result achieved through a 
 determination to increase his reading rate by per- 
 sistent practice. He says : 
 
 "I have considerably increased my own speed in reading 
 by waking up to the fact that my rate was unnecessarily slow, 
 and then persistently reading as fast as possible with well 
 concentrated attention, taking care to stop short of fatigue 
 until the new pace was somewhat established. I thus reached 
 a speed of a page per minute for such books as Ellis' 'The 
 Criminal' of the Contemporary Science Series, maintaining 
 this rate for a half hour or so at a time, and with very good 
 comprehension of what was read, although after such read- 
 ing a very hasty review of the leading points was the most 
 satisfactory procedure. My earlier speed was not half so 
 great." (6) 
 
 Peters (37) conducted an experiment in 1914-15 to 
 determine the influence of speed drills on silent read- 
 ing. The investigation was conducted in grades three 
 to six inclusive. The speed drills lasted "ordinarily 
 from five to ten minutes." The regular oral reading 
 work was not discontinued, these drills being given in 
 addition, without, however, giving a greater total period 
 of time to the experimental group than that received 
 by the control group. The experiment lasted seven 
 months. Peters reports a "gain of 18.7 per cent in 
 speed, but a trifling loss in quality as a result of the 
 speed drills." The following is a summary of the re-
 
 THE RATE OF SILENT READING 
 
 39 
 
 suits in speed attained by the "drill" and the "no- 
 drill" groups. 
 
 TABLE I. SPEED IN READING 
 
 
 Nov. 6 
 
 JAN. 20 
 
 APR. 20 
 
 JUNE 2 
 
 Drill Group .... 
 No-drill Group . . . 
 
 167.7 
 201.1 
 
 266.5 
 248.0 
 
 219.8 
 253.8 
 
 261.4 
 264.9 
 
 The experiment of Peters bears some points of simi- 
 larity to that part of the present investigation which 
 aims at determining the influence of Type I Train- 
 ing in Rapid Silent Reading. (See Chapter IV.) This 
 investigation differs very considerably from Peters' 
 study, however, in that this type of training con- 
 stitutes a method complete in itself, to which was 
 devoted the entire reading period. The ordinary 
 procedure in the reading period of conducting oral 
 reading recitations, phonetic exercises, drills in word- 
 pronouncing, etc., was entirely replaced by training 
 in speed and comprehension in silent reading. In 
 the experiment of Peters the regular oral reading work 
 was continued as usual, with the addition of a speed 
 drill lasting "ordinarily from five to ten minutes." 
 
 The length of time assigned for the drill in speed in 
 Peters' experiment is thus seen to be quite indefinite. 
 Classes securing the full ten minutes for this drill would 
 thus receive twice as much time for practice in rapid 
 reading as the classes using only five minutes. Such 
 a considerable variation in the possible amount of time
 
 40* SILENT READING 
 
 to be used in speed drills, with apparently no record 
 kept of the amount each class actually used would seem 
 to render difficult and uncertain any comparison be- 
 tween the gains made by the different classes. Further- 
 more, the continuance of the regular work in oral read- 
 ing, to which most of the period was devoted, along with 
 the speed drills of five to ten minutes' duration, would 
 seem to render the final result the tangled product of a 
 rather complicated and mixed course of training. 
 
 C. T. Gray (19) reports the results of training a 
 single fifth grade pupil in rapid silent reading. The 
 practice period varied from 15 to 20 minutes a day and 
 continued for twenty days. "Speed was the aim here, 
 at the expense of comprehension, if necessary. Never- 
 theless comprehension was also checked up by oral 
 reproduction. . . . The material for the training was 
 carefully selected, so as not to be too difficult and yet 
 to require considerable effort on the part of the reader." 
 Concerning the results of the practice, as determined 
 by the same test which was taken before and after the 
 training, Gray says : " In every test there is a decided 
 improvement in speed. In only two of the tests is 
 there any decrease in comprehension, and in each case 
 this loss is on the most difficult test of the series." 
 
 Unlike the comparatively small increase reported by 
 Peters the gain here varied from 84 per cent to 217 
 per cent. While no conclusion can be drawn from the 
 performance of a single pupil, the large gain reported 
 struck the writer as deserving of note. It raised in his
 
 THE RATE OF SILENT READING 41 
 
 mind the query, Would it be possible to secure a similar 
 percentage of gain when dealing with a much larger 
 group of subjects? The investigation of the influence 
 of Type I Training in Rapid Silent Reading repre- 
 sents the endeavor to secure an answer to that question. 
 
 Oberholtzer (23) conducted an investigation with 
 1800 pupils in grades one to eight in the Tulsa public, 
 schools to determine the rates of reading. Among 
 the conclusions he draws from this study are the 
 following : 
 
 "The rate of reading in all the grades is relatively 
 lower than it should be if the habit of reading had 
 from the beginning been acquired through conscious 
 effort to improve the rate." 
 
 "One room showed an increase of 50 per cent in the 
 rate after two weeks' practice in rapid reading. These 
 tests showed that oral expression and the power to grasp 
 the content were equally improved." 
 
 In an unpublished Master's Thesis (33) at the Uni- 
 versity of Chicago, Miss Bowden thus reports the re- 
 sults of practice upon the reading of a number of adults : 
 "The figures show that . . . the reading rate may 
 be markedly increased by practice." She also reports 
 the influence of a ten-minute exercise in rapid reading 
 upon the rate of a seven-year old boy. The practice 
 continued for 40 days ; the child's rate was increased 
 from 12 words per minute to 55 words per minute. 
 Miss Bowden is of the opinion that similar exercises 
 in rapid reading would prove beneficial in increas-
 
 42 SILENT READING 
 
 ing the rate of elementary-school children whose 
 motor habits are not definitely set. 
 
 That practice in rapid silent reading is therefore a 
 factor of prime importance in conditioning the rate, 
 the above-mentioned investigations abundantly show. 
 It is true that many of these studies dealt with but a 
 very small number of subjects, while in others the se- 
 curing of data on this particular point was but a 
 secondary consideration. In fact, the investigation 
 conducted by Peters seems to have been the only one 
 involving a fairly large number of subjects, which had 
 as its primary aim the discovery of the influence of 
 speed drills upon the rate of reading. 
 
 These studies are, however, interesting and signifi- 
 cant, at least in the writer's judgment, not so much for 
 the bulk or weight of their findings taken by themselves 
 as for the corroborative evidence of an experimental 
 character which they lend to a theory which has in- 
 trinsically very strong a priori grounds of credibility. 
 For, as will be pointed out at greater length in the 
 exposition of this type of training in Chapter IV, prac- 
 tice has a very marked influence on other types of ac- 
 tivity, and there appears to be no inherent reason why 
 reading should prove to be the solitary exception. Of 
 the three types of training developed in the present 
 study, and of any other embryonic ones touched upon 
 in the literature on this subject, this method is, in the 
 writer's judgment, the most important one, and the one 
 which is possessed of the strongest intrinsic grounds of
 
 THE RATE OF SILENT READING 43 
 
 plausibility. It has accordingly been made the basic 
 principle or factor upon which has been constructed 
 a distinct type of teaching method. This type is 
 called Training in Rapid Silent Reading. 
 
 2. The Decrease of Vocalization in Silent Reading 
 
 The attention of students who have endeavored to 
 analyze the reading complex into its constituent pro- 
 cesses has often been attracted by the concomitant 
 movements of the lips, tongue, inner palate, pharynx, 
 vocal cords, and the general mechanism of speech. The 
 reader apparently not only visualizes or perceives the 
 printed words, but says them to himself. This com- 
 bination of motor and auditory elements constitutes the 
 so-called "inner speech" of silent reading. A consid- 
 eration of the origin of this habit, of the psychological 
 role it plays in reading, and of the possibilities of eradi- 
 cating it, is reserved for a subsequent chapter. The 
 writer will present here merely the results of investi- 
 gations which indicate that the habit of vocalization, 
 involving movements of the lips, tongue, etc., is a 
 factor influencing the rate of silent reading. 
 
 One of the earliest attempts to discover the influence 
 of lip movement and of the general motor tendency to 
 vocalize was made by Quantz (26) in 1897. Concern- 
 ing the results of his study, Quantz concludes : 
 
 "This shows that the motor tendency (as manifested by 
 lip movement) in any degree has an influence detrimental 
 to rapidity of reading, and the stronger the tendency the
 
 44 SILENT READING 
 
 greater is the hindrance ; the numbers representing the lip 
 movement regularly decrease as reading rate increases, and 
 vice versa. . . . The ten slowest readers show almost 
 double the amount of lip movement that the ten most rapid 
 do." 
 
 Quantz further sought to ascertain the relationship 
 between lip movement and the rate of reading. His 
 conclusion is: "The ratio of lip movement to total 
 amount of reading" is inverse. The "extent of read- 
 ing works directly against movement of lips" as evi- 
 denced by the fact that " not one of those whose reading 
 is widest is a lip mover to any extent which can be ob- 
 served." 
 
 C. T. Gray (19) reports the results upon the rate of 
 silent reading of training given to decrease vocalization. 
 The two subjects chosen displayed a pronounced habit 
 of vocalizing. The training consisted in first pointing 
 out the disadvantages of the possession of this habit in 
 a marked degree. They were then requested to read 
 in their natural manner, but without vocalizing. Gray 
 writes : 
 
 "Comprehension was tested either with oral questions 
 or with oral reproductions. That the practice was effective 
 was shown by the fact that no vocalization was noted when 
 the subjects took the final tests. Their instructor reported 
 that at first there was considerable distraction caused by 
 the attempts to eliminate the vocalization. There was, 
 however, a gradual decrease in this distraction, and before 
 the end of the practice period the reactions of the children 
 seemed perfectly normal. The practice continued for twenty
 
 THE RATE OF SILENT READING 
 
 45 
 
 days. The length of time for each day was from twenty to 
 thirty minutes." 
 
 A summary of the results of the training as deter- 
 mined by having the two subjects take the same test 
 before and after the practice period is presented in 
 Table II. 
 
 TABLE II. 
 
 RECORD OP Two SUBJECTS BEFORE AND AFTER PRACTICE 
 IN THE DECREASE OF VOCALIZATION 
 
 
 READING TO ANSWER QUESTIONS 
 
 READING TO REPRODUCE THE 
 SUBSTANCE OF CONTENTS 
 
 js 
 1* 
 
 Rate 
 
 Comprehen- 
 sion 
 
 J 
 I* 
 
 Rate 
 
 Comprehen- 
 sion 
 
 6th Grade 
 Subject 
 4th Grade 
 Subject 
 
 A 
 
 D 
 
 E 
 
 Gain 177% 
 
 Gain 60% 
 Loss 13% 
 
 Loss 75% 
 
 Loss 11% 
 Loss 11% 
 
 B 
 C 
 D 
 
 Gain 160% 
 Gain 88% 
 Gain 39% 
 
 Loss 0.2% 
 Loss 100% 
 Gain 71% 
 
 Table II is to be read thus: On selection A, the sixth grade subject 
 made a gain of 177 % in rate, but lost 75 % in comprehension as 
 measured by the answering of questions on the matter read. In read- 
 ing selection B to reproduce the substance of the contents, the sixth 
 grade subject gained 160 % in rate, and lost 0.2 % hi comprehension. 
 
 Gray thus summarizes the "positive" results of the 
 above experiment : 
 
 1. "Training in reading without vocalization results in a 
 marked decrease in this motor accompaniment of reading. 
 
 2. " Training of the type mentioned above results in an in- 
 crease in rate. In considering these results for gain in 
 rate, it should be remembered that the practice work was 
 done at the normal rate."
 
 46 SILENT READING 
 
 The " negative " results may be summarized as 
 follows : 
 
 1. "In five of the seven tests taken by the two subjects 
 there is a loss in comprehension; in one of the tests no 
 change is shown in the results for comprehension; and in 
 one test there is a gain." 
 
 In the case of these two subjects the attempt to 
 decrease vocalization interfered rather considerably 
 with their comprehension. The conscious attempt at 
 inhibition was apparently somewhat disturbing, draw- 
 ing off part of the attention from the assimilation of 
 the content to the manner of reading it. This dis- 
 turbing element would seem to be no more than should 
 be expected during the initial stages of the breaking 
 down of an old habit and the substitution of a new 
 one. The experiment lasted such a short period 
 that it leaves undetermined the effect that would be 
 produced by a longer period of practice, when the de- 
 crease in vocalization would have become habitual or 
 automatic, and would in consequence have drawn no 
 conscious attention from the comprehension of the 
 content. Furthermore, the small number of subjects 
 in this experiment there were but two renders it 
 unsafe to draw any conclusion, save one of a most tenta- 
 tive character, concerning the effect this training would 
 produce upon a large number of pupils. 
 
 That part of the present investigation which deals 
 with Type II --Training to Decrease Vocalization 
 by extending this training to a much larger number of
 
 THE RATE OF SILENT READING 47 
 
 pupils, and by allotting to it the entire reading period 
 of thirty minutes daily for a period of two months, 
 endeavors to throw some light on questions left unan- 
 swered by C. T. Gray's simple, but suggestive experi- 
 ment. As will be pointed out in a subsequent chapter, 
 the method worked out in Type II, while in principle 
 similar to Gray's, differs from it in many important 
 respects. It introduces a number of new details, 
 devices in motivation, etc. Furthermore, it involves 
 reading not at the ordinary leisurely rate, as in Gray's 
 experiment, but at as fast a rate of speed as is con- 
 sistent with effective comprehension. 
 
 Miss Abell (34) reports the results of an early investi- 
 gation of the reading rates of 41 students at Wellesley. 
 Concerning the relation of vocalization to rapidity in 
 silent reading, Miss Abell draws the following conclusion : 
 
 "A characteristic correlate, in the case of our subjects, 
 of slow reading, is the actual pronunciation or the vivid 
 articulatory imagination of the words read. This device, 
 which may assist in the comprehension of a strange word or 
 of an obscure meaning, is certainly a hindrance when it 
 becomes habitual. The discouragement of a child's tendency 
 to accompany reading by articulation is an evident pedagogi- 
 cal requirement." 
 
 Dodge (35) reports that in his own silent reading, at 
 the ordinary rate, the vocalization of practically every 
 word occurs. In his most rapid reading of ordinary 
 material, merely the beginnings of words are vocalized, 
 while in his reading of very familiar matter, at top-
 
 48 SILENT READING 
 
 speed, only some words are pronounced. This further 
 confirms the conclusion that a decrease in vocalization 
 is correlated with an increase in speed. 
 
 Huey (6) reports a number of exceptionally fast 
 readers. Among the number cited is the case of a 
 mathematician who "read the whole of a standard 
 novel of 320 pages in two and one-fourth hours." Com- 
 menting upon the nature of the reading process at such 
 a rate, Huey says : 
 
 "I am inclined to think that at any such speed the mean- 
 ings suggested immediately by the visual forms suffice for 
 all but the more important parts, and that these meanings 
 are felt sufficiently, without inner utterance, to permit 
 selection of what is more important, the more important 
 places themselves having a fleeting inner utterance to vivify 
 their meaning. . . . The inner speech in such cases must at 
 any rate suffer a foreshortening and atrophy of articulatory 
 details which reduce it to little more than a slight motor 
 tallying as the meanings are felt or dwelt upon." 
 
 The results of the investigations above cited suffice 
 to show that vocalization as manifested by lip move- 
 ment, etc., in silent reading is undoubtedly a factor of 
 some consequence in affecting the rate. Its effect, 
 however, is to retard rather than to accelerate. Con- 
 sequently, its elimination or at least its decrease would 
 seem to be instrumental in increasing the rate. An 
 attempt to accomplish this result has been made by 
 devising Type II --Training to Decrease Vocaliza- 
 tion. Though the lessening of inner speech is the
 
 THE RATE OF SILENT READING 49 
 
 fundamental factor and the distinguishing character- 
 istic of this method, there are also, as will be seen later, 
 other factors of scarcely less importance which have 
 been incorporated into this type of training. 
 
 3. Training in Perception 
 
 Photographic records of the eye-movements and 
 pauses during reading have served to emphasize the 
 fact that most of the total reading time is consumed 
 by the pauses. In fact, the duration of the fixations 
 being equal, the total reading time will then bear a 
 direct relation to the number of fixations. The fewer 
 the fixations, the shorter will be the reading time, the 
 greater the number of fixations the longer the reading 
 time required. Consequently, if a type of training can 
 be devised which will effect a decrease in the number 
 of fixations, the reading time will be correspondingly 
 decreased. But the number of pauses per line is, gen- 
 erally speaking, conditioned by the width of the per- 
 ceptual span. Training, therefore, which will enlarge 
 the perceptual span will result in a reduction of the 
 number of fixations per line, with a corresponding in- 
 crease in the speed of reading. 
 
 Photographic records of the behavior of the eyes dur- 
 ing reading have focused attention upon the fixation 
 pauses, and have shown them to be the crucial points 
 wherein any appreciable economy in the length of the 
 reading time must be effected. A more detailed study 
 of the nature of the fixation pause, its duration, the
 
 50 
 
 SILENT READING 
 
 extent of its visual reach, and the psychic processes 
 occurring therein, is reserved for a subsequent chapter. 
 It will suffice to indicate here that evidence of an ex- 
 perimental character shows that training in perception 
 is a factor of importance in accelerating the rate in 
 silent reading. 
 
 The experimental evidence showing the influence of 
 this factor is the result of an investigation reported by 
 C. T. Gray (19). Two fourth grade subjects, a boy 
 and a girl, were given practice in quick perception by 
 means of short-exposure exercises. Various kinds of 
 material were exposed by a fall-tachistoscope for a 
 fraction of a second. As the period of exposure was 
 so brief that but a single fixation was possible, the 
 material would have to be grasped in a single span of 
 attention. The effort to perceive all the material ex- 
 posed would thus tend to widen the span of attention, 
 if the latter were at all susceptible to such improve- 
 ment. Tables III and IV give a summary of the results 
 achieved by each pupil as reported by Gray. 
 
 TABLE III. RECORD OF FOURTH GRADE BOY, BEFORE AND AFTER 
 SHORT-EXPOSURE PRACTICE 
 
 
 AVERAGE 
 ON 2 
 
 WORDS 
 
 AVERAGE 
 
 ON 3 
 
 WORDS 
 
 AVERAGE 
 
 ON 4 
 
 WORDS 
 
 AVERAGE 
 
 ON 5 
 
 WORDS 
 
 AVERAGE 
 
 ON 6 
 
 WORDS 
 
 AVERAGE 
 
 ON 7 
 
 WORDS 
 
 Before 
 
 
 
 
 
 
 
 Practice 
 
 1.8 
 
 2.5 
 
 2.5 
 
 3.0 
 
 3.3 
 
 Omitted 
 
 After 
 
 
 
 
 
 
 
 Practice 
 
 2.0 
 
 3.0 
 
 4.0 
 
 4.3 
 
 3.3 
 
 3.5
 
 THE RATE OF SILENT READING 
 
 51 
 
 TABLE IV. 
 
 RECORD OF FOURTH GRADE GIRL BEFORE AND AFTER 
 SHORT-EXPOSURE PRACTICE 
 
 
 AVERAGE 
 
 ON 2 
 
 WORDS 
 
 AVERAGE 
 
 ON 3 
 
 WORDS 
 
 AVERAGE 
 
 ON 4 
 
 WORDS 
 
 AVERAGE 
 
 ON 5 
 
 WORDS 
 
 AVERAGE 
 
 ON 6 
 
 WORDS 
 
 AVERAGE 
 
 ON 7 
 
 WORDS 
 
 Before 
 Practice 
 
 1.3 
 
 1.0 
 
 1.6 
 
 1.5 
 
 2.2 
 
 2.0 
 
 After 
 Practice 
 
 2.0 
 
 3.0 
 
 4.0 
 
 3.9 
 
 3.3 
 
 3.5 
 
 Table III shows a notable improvement in every 
 case, except in the span for the six-word series, where 
 the averages coincide. Table IV is even more striking. 
 It shows a widening of the perceptual span in every 
 series. While a fall-exposure apparatus was used 
 in the practice exercises, a different instrument was 
 employed in the original and final tests. 
 
 "This," says Gray, "eliminates to a large degree the ele- 
 ment of adjustment to the apparatus and makes the marked 
 increase in the perceptual span the more significant. The 
 importance of a wide span of attention . . . makes this re- 
 sult all the more valuable. It seems that positive results 
 can be obtained as early as the fourth year. In the light 
 of these facts it may be repeated here that without doubt 
 the type of training which is given in primary work by means 
 of the flash cards is a very important element in the early 
 training in reading. It is also clear that the foregoing 
 results indicate the desirability of a better grading and con- 
 trol of such work. Incidentally the result makes it clear 
 that training must come at the time in a child's development 
 when it can produce results. Late training is often defec- 
 tive."
 
 52 SILENT READING 
 
 Gray's conclusion concerning the necessity of giving 
 this training sufficiently early in the child's develop- 
 ment, i.e. before the sixth grade, is based upon results 
 secured from giving this type of training to two sixth 
 grade pupils. These results indicated that "no strik- 
 ing change was made in the span of attention." 
 Whipple (36) secured similar results with adults. 
 
 It is probable that a distinction has to be made here 
 between the absolute span of perception, as determined 
 by tachistoscopic exposures, and the relative span of 
 perception which actually functions in reading. Though 
 tachistoscopic exposures may show no widening of the 
 absolute span of attention as" a result of training in 
 perception, there may nevertheless result the function- 
 ing of a wider span of attention in reading than was the 
 case before the training. In other words, there results 
 the more effective use of the perceptual span in reading, 
 where it practically never covers the extent of printed 
 matter, which tachistoscopic exposures show is possible 
 for it. 
 
 Consequently, while training in perception may not 
 markedly increase the absolute span of attention for 
 pupils in the sixth grade and above, nor for adults, the 
 conclusion that the relative span of attention which 
 functions in the reading of such persons may not be 
 enlarged as a result of this type of training cannot be 
 drawn. That this widening of the perceptual span in 
 reading is not only a possible result of such training, 
 but that it is the result which actually occurred in the
 
 THE RATE OF SILENT READING 53 
 
 case of Gray's sixth grade subject B. R., the writer 
 thinks can be clearly demonstrated from Gray's own 
 figures. This will be shown in Chapter VI Train- 
 ing in Perception. It is sufficient tondicate here that 
 training to enlarge the perceptual span has been shown 
 to be no unimportant factor in accelerating the rate 
 of silent reading. It has accordingly been made the 
 basic principle upon which has been constructed the 
 third type of training. 
 
 4. The Character of Subject Matter 
 
 Simple material will naturally be read more rapidly 
 than difficult material. Under this head comes also 
 familiarity with the subject matter. Thus, the ordi- 
 nary reader would probably find a treatise on psychol- 
 ogy rather slow reading, while a psychologist would 
 read and assimilate it rapidly because of his greater 
 familiarity with the subject matter and greater ap- 
 perceptional capacity. 
 
 Dearborn found that by assigning to a mathemati- 
 cian, a teacher in a secondary school, and a psychologist, 
 subject matter unrelated to their respective lines of 
 study, marked difference in the reading rates resulted. 
 Dearborn reports that a reader saved one-third of the 
 total time required for the first reading upon pe- 
 rusing the same selection a second time. The results 
 secured in the present investigation results which 
 will be presented in a subsequent chapter prove 
 conclusively that, other things being equal, familiarity
 
 54 SILENT READING 
 
 with the subject matter notably increases the speed of 
 reading. 
 
 5. Habits of Eye-Movement 
 
 Records of the eye-movements of many readers as 
 shown in the monographs of Dearborn, C. T. Gray, 
 Schmidt, Huey, and Judd, indicate that as a general 
 rule rapid efficient reading is characterized by rather 
 uniform, rhythmical movements of the eyes in contrast 
 with the irregular movements and frequent regressions 
 of the stumbling, halting reader. Dearborn as quoted 
 by W. S. Gray (38), says : "It is in the writer's belief 
 clearly indicated by the above experiments that one of 
 the essentials of natural and rapid reading is that the 
 reader's eye should at once be able to acquire a regular 
 and uniform motor habit of reaction for each line." 
 
 Huey, in agreement with the above observation of 
 Dearborn, remarked in regard to the eye-movements 
 of the 20 graduate students whom he tested: "The 
 readers showed a rhythmic tendency. Each would 
 fall into a reading pace that seemed most natural to 
 him, and would then read page after page in almost 
 exactly the same time. Quite usually the differences 
 from page to page would not be over three or four 
 seconds. . . . Habits of eye-movement are doubtless 
 important factors in setting the pace." 
 
 It is doubtless true, however, that the character of 
 the eye-movements is affected very considerably by 
 other factors, such as the size of the visual span, the
 
 THE RATE OF SILENT READING 55 
 
 simple or difficult character of the subject matter, the 
 purpose for which it is read, and the ability to grasp 
 the meaning quickly. Yet over and above all these 
 factors it is probable that there is a habit of eye-move- 
 ment of a fairly definite kind, which functions in the 
 ordinary reading of the individual, and which tends 
 to persist to some extent, even when the character of 
 the subject matter and other factors are somewhat 
 changed. Of two individuals possessing equal assimila- 
 tive capacity, reading the same subject matter under 
 similar conditions, the habit of regular rhythmical eye- 
 movements would tip the scales in favor of its owner 
 over the unfortunate individual who has never rendered 
 an ocular motor reaction of this type habitual in his 
 reading. 
 
 Dean Fordyce (39) in a paper read at the Convention 
 of the National Education Association in 1917 seems to 
 regard this as the chief causal factor in accounting for 
 the differences in rate. "The difference in reading 
 rate," he says, "is largely a matter of the rhythmical 
 motor habits into which the eye is trained in the early 
 attempts to read. If the eye falls into the habit of 
 pausing on each word, the rate is slow and plodding ; 
 if into the habit of taking in the larger units of phrases 
 and clauses, the rate is correspondingly rapid and the 
 interpretation even easier. The trained eye grasps 
 the words of a phrase or sentence in a single unitary 
 act ; similarly the perceptive power grasps the ideas in 
 their combined form in the thought."
 
 56 SILENT READING 
 
 Dean Fordyce further relates the rather striking 
 acceleration of his own reading rate by gradually re- 
 placing his defective ocular motor reactions with regular 
 well-balanced eye-movements. He says : 
 
 "The writer discovered several years ago that his reading 
 rate was very much below the norm. He sought for months 
 a means of remedying the defect. Being convinced that his 
 difficulty was largely a matter of the defect in motor-eye 
 habits, he set about the ludicrous task of learning to read 
 as if he were a first-grader. The first reader was adopted for 
 the practice. A series of exercises consisting first of a column 
 of two-word phrases was used the first month. After his 
 eye had fallen into the habit of seizing such phrases as units, 
 another column of three-word phrases and clauses was 
 adopted for practice. Later a column of short sentences 
 was used. Through ten minutes of judicious daily practice 
 on these exercises, and on simple easily comprehended prose, 
 the author in a single year doubled his speed in reading, and 
 has been pleased to note a similar achievement among his 
 students as a result of such practice. By such exercises the 
 eye falls into new motor habits of a regular rhythmical nature, 
 enabling the student to read in large rather than small units. 
 The perceptual span increases with the ocular span." 
 
 In the exercises above described, however, there was 
 training not only in rhythmical eye-movements, but 
 also training to increase the perceptual span. The 
 widening of the visual span would naturally cause the 
 eye-movements to cover a greater extent of the printed 
 line, thereby necessitating fewer movements per line 
 and tending to inhibit the numerous regressive move- 
 ments which are the usual concomitants of a slow,
 
 THE RATE OF SILENT READING 57 
 
 halting reading pace. Thus the exercises with the col- 
 umn of two-word, and then three-word, phrases, etc., 
 would seem to be training which would have as its 
 immediate specific effect the enlarging of the perceptual 
 span to grasp these phrases as a unit in a single fixation. 
 That would be training solely in perception. The 
 work with the first reader and similar easily compre- 
 hended prose, however, would appear to be largely 
 training in the formation of rapid, regular, and rhythmi- 
 cal eye-movements. The rapid reading of such simple 
 familiar material which can be grasped instantly, seems 
 to be especially conducive to the setting up of regular 
 rhythmical eye-movements. These can gradually be 
 crystallized into a habit which functions in almost all 
 the individual's reading because it has now become for 
 him the natural manner of reading. 
 
 The entire increase in Fordyce's speed in reading 
 could not therefore be attributed solely to the forma- 
 tion of a habit of regular rhythmical eye-movements. 
 Apparently a considerable share of the increase must 
 be credited to an enlargement of the perceptual span, 
 or to a more effective use of it in reading. Some of it 
 is also doubtless due to the setting up of "higher order" 
 habits of eye-movements. 
 
 The case illustrates how closely and almost in- 
 separately are intertwined these two processes of the 
 reading complex eye-movements and perceptual 
 span. Thus the functioning of a wide perceptual span 
 in reading insures fewer movements per line, thereby
 
 58 SILENT READING 
 
 affording an extent of area sufficient to permit the eye 
 to fall into a certain regular sweep, or rhythm of move- 
 ment, which is practically impossible where the eye must 
 pause once or several times on practically every single 
 word. In the case of the training described by Fordyce, 
 the functioning of the wide perceptual span would 
 appear to be largely the causal factor and the regular 
 rhythmical motor reactions the resultant. There is 
 such an intimate interrelation, however, that they 
 probably interact upon one another. It is very prob- 
 able and the experiments of Fordyce, Huey, and 
 others lend great weight to the assumption that 
 the training in the regular rhythmical movements of 
 the eyes across the printed line tends to enlarge the 
 perceptual span, or at least to utilize it more effectively 
 by preventing less overlapping of the perceptual spans 
 in reading. As Fordyce expresses it "the perceptual 
 span increases with the ocular span." 
 
 This regular rhythmical character of the eye-move- 
 ments is pointed out as a general characteristic of 
 rapid, efficient reading. It is not always present, 
 however, in such reading. As Dearborn observes, 
 "rapidity of reading is not necessarily correlated with 
 regularity of movement," there being some fairly rapid 
 readers who do not seem to have acquired this habit of 
 eye-movement. They would seem to be fairly rapid, 
 however, not because of the absence of this habit but 
 rather in spite of it. An examination of the photo- 
 graphic records of the eye-movements published by
 
 THE RATE OF SILENT READING 59 
 
 C. T. Gray, Dearborn, Huey, and Judd, and of those 
 taken in the present investigation, shows clearly that 
 one of the important characteristics distinguishing the 
 rapid efficient reader from the slow plodding or stum- 
 bling one is the presence of a certain regular rhythmical 
 character of the eye-movements. This rhythm is 
 shown on the photographic records by a fair uniformity 
 in the number of pauses, which are usually an appre- 
 ciable distance apart, and by a somewhat "uniform 
 method of time distribution" in the pauses. The hy- 
 pothesis advanced by Dearborn (17) in explanation 
 of this regular rhythmical character of the eye-move- 
 ments in fast reading is that "the rapid reader dis- 
 tributes his attention more readily at the initial fixa- 
 tion of the line, and is enabled on this account to fall 
 more readily into a uniform habit of movement." 
 Whatever explanation of the fact is finally accepted, 
 there can be no doubt that the bulk of existing experi- 
 mental evidence points to the habit of regular rhythmi- 
 cal eye-movements as no insignificant factor in affecting 
 the rate in silent reading. W. S. Gray (20) thus aptly 
 epitomizes the evidence on this point : " Regular 
 rhythmical movements of the eyes are prerequisite to 
 rapid silent reading." 
 
 6. Purpose for Which the Subject Matter Is Read 
 
 The rate of reading is conditioned very largely by the 
 purpose or object for which one reads. Reading simply 
 to "get the gist" of the story or the selection induces
 
 60 SILENT READING 
 
 a mental attitude radically different from the attitude 
 assumed when one reads for the purpose of answering 
 detailed questions on the matter. The whole mental 
 "set," the entire mode of attack, the degree of con- 
 centration, the amount of reflective thought and logical 
 association, all of which influence the rate, will depend 
 largely upon the end for which the reading is done. 
 
 Whipple and Curtis (40), in their investigation of 
 skimming, found that a considerable "slowing down" 
 resulted when the subjects knew that they would be 
 required to reproduce. Besides ascertaining the dif- 
 ferences in rate for both oral and silent reading when 
 the reading was done for different purposes, C. T. Gray 
 (19) took photographs of the eye-movements for the 
 several different resultant types of reading. In the 
 first type, the reading was simply for the purpose of 
 securing "an intelligent understanding of the material. 
 Three other types of records were taken. One of these 
 was of reading which was followed by the answering 
 of questions on prose ; another was a passage of poetry, 
 with the answering of questions ; and a third was the 
 reading of prose for reproduction. . . . These results 
 (summarized below) indicate clearly that the reader 
 does differentiate between different types of reading and 
 evidently approaches different reading problems with 
 a different mental 'set.' The results here reported 
 are typical of all the subjects. It is also interesting to 
 know that the length of the pauses does not vary in any 
 large degree, while the number of pauses varies greatly."
 
 THE RATE OF STLENT READING 
 
 61 
 
 TABLE V. EYE-MOVEMENT RECORD OF ONE SUBJECT SHOWING 
 VARIATIONS ACCORDING TO PURPOSE FOR WHICH THE MATTER 
 WAS READ. 
 
 
 No. OF PAUSES PER LINE 
 
 AVERAGE LENGTH OF PAUSES 
 
 Prose 
 (Simply 
 to Under- 
 stand) 
 
 Prose 
 (Answer- 
 ing Ques- 
 tions) 
 
 Poetry 
 (Answer- 
 ing Ques- 
 tions) 
 
 Prose 
 (Repro- 
 duction) 
 
 Prose 
 (Simply 
 to Un- 
 derstand) 
 
 Prose 
 (Answer- 
 ing Ques- 
 tions) 
 
 Poetry 
 (Answer- 
 ing Ques- 
 tions) 
 
 Prose 
 (Repro- 
 duction) 
 
 Average 
 
 8.1 
 
 9.6 
 
 11.0 
 
 8.7 
 
 13.3 
 
 13.9 
 
 14.0 
 
 14.6 
 
 Table V is to be read thus : The average number of fixation-pauses 
 per line in reading prose (simply to understand) is 8.1, in reading prose 
 (to answer questions), 9.6, etc. The average length of the fixation- 
 pause in reading prose (simply to understand) is 13.3 fiftieths of a 
 second, etc. 
 
 TABLE VI. VARIATION IN NUMBER OF REGRESSIVE MOVEMENTS 
 ACCORDING TO PURPOSE FOR WHICH THE MATTER WAS READ 
 
 
 PROSE 
 
 PROBE 
 
 POETRY 
 
 POETRY 
 
 
 (Simply to 
 
 (Answering 
 
 (Answering 
 
 (Reproduc- 
 
 
 Understand) 
 
 Questions) 
 
 Questions) 
 
 tion) 
 
 Average per line . . 
 
 1.8 
 
 2.4 
 
 3.3 
 
 1.8 
 
 Table VI is to be read thus: The average number of regressive 
 movements per line in reading prose (simply to understand) is 1.8, in 
 reading prose (to answer questions) it is 2.4, etc. 
 
 The figures in the above table tell the inside story of 
 the differences in rate resulting from the different ends 
 for which the passage was read. They show that the 
 physiological processes are perceptibly different in each 
 type of reading. There is a variation in the number of 
 fixation-pauses, in the average duration of the pauses, 
 as well as in the number of regressive movements.
 
 62 SILENT READING 
 
 These are the physiological correlates which reflect 
 the differences in the conscious functioning in the 
 changed mental "sets" or attitudes. 
 
 Besides the various aims in reading which have al- 
 ready been mentioned, there are numerous others 
 which shade off gradually from one another. The 
 following are illustrations of purposes for which a selec- 
 tion may be read, each of which would affect the reading 
 rate in an appreciably different manner : 
 
 1. Simply to understand a passage, to "get the gist" 
 of it. 
 
 2. To find a few specific facts. 
 
 3. To determine the logical consistency of a line of 
 argumentation. 
 
 4. To appreciate the author's diction and general 
 rhetorical constructions. 
 
 5. To criticize the thought or language of the selec- 
 tion. 
 
 6. To remember so as to retell to others. 
 
 7. To be able to follow directions of procedure. 
 
 8. To select only the main points. 
 
 These are but a few of the numerous purposes for 
 which one actually reads at various times. They could 
 be drawn out at greater length. This list, however, 
 will suffice to illustrate the different attitudes assumed 
 in reading, with consequent differences in both rate and 
 quality of comprehension. That reading in the grades 
 for one or two simple purposes does not develop the 
 ability to read effectively for other ends is evident from
 
 THE RATE OF SILENT BEADING 63 
 
 the results reported in 1917 by W. S. Gray (41) in the 
 Survey of the St. Louis Schools : 
 
 "The facts presented above point to the conclusion that 
 the various phases of silent reading ability do not always 
 develop in the same proportion. Each phase needs special 
 attention, and the teacher must plan instruction so that 
 pupils receive effective training in all phases of silent reading 
 ability. This includes not only ability to reproduce and 
 ability to answer questions, but in addition, ability to select 
 pivotal ideas, ability to organize, ability to determine the 
 relative importance of facts, ability to associate new ideas 
 with one's store of knowledge, etc. To the extent that a 
 teacher gives specialized training along one or two of these 
 lines, just to that extent will her pupils be more likely to fail 
 in a situation which calls for silent reading ability of a broad 
 character." 
 
 The character of the purpose for which the reading 
 is done must accordingly be ranked among the factors 
 of primary importance in conditioning not only the 
 quality of comprehension, but also the rate of reading. 
 
 7. Concentration of Attention 
 
 Much of the time consumed in apparent reading is, 
 in reality, lost through unconscious wanderings of the 
 attention and fruitless daydreaming. The eyes often 
 remain fixed upon the page while the mind is visiting 
 distant climes and is busy " building castles in Spain." 
 How much of the time seemingly spent in the reading of 
 a book is actually frittered away in listless poring and 
 idle woolgathering, especially if the subject matter is
 
 64 - SILENT READING 
 
 rather heavy and somewhat uninteresting, is difficult 
 to determine ; but that it is considerable, the experience 
 of most readers leaves no room for doubt. The intro- 
 spection of a number of graduate students in education 
 and psychology and two university professors, compos- 
 ing a seminar group at which this study was first dis- 
 cussed, strongly confirmed the writer's own conviction 
 that this was one of the strategic places in the reading 
 process wherein could be effected a very considerable 
 improvement in both rate of reading and quality of 
 comprehension. If this tendency of the mind to wan- 
 der exists even in the case of adults, it must be much 
 stronger in the case of children whose imaginations are 
 usually very strong and active, but whose powers of 
 concentration are much less developed. 
 
 Quantz's study, showing that the rapid readers not 
 only averaged approximately 37 per cent superiority 
 in the quality of their work, but also introduced less 
 extraneous matter in their reproduction, offers corrobo^ 
 rative experimental evidence of the above conclusion. 
 Concerning the rapid type of reader, Quantz (26) says : 
 "He introduces only two-thirds as many thoughts not 
 found in the original selection." This fact doubtless 
 finds its explanation in the greater concentration of at- 
 tention necessitated by rapid reading, and the conse- 
 quent avoidance of daydreaming and woolgathering 
 which is undoubtedly the prolific source of the ex- 
 traneous material found in the reproduction of a 
 selection.
 
 THE RATE OF SILENT READING 65 
 
 Concentration of attention may be promoted by 
 various means. Chief among these is the judicious 
 selection of interesting material which makes a strong 
 appeal to the mind of the child. The matter must 
 grip the pupil's attention, especially in the lower 
 grades, instead of requiring his volatile, fluctuating 
 attention to try to grip it. Much of the slow listless 
 reading observable in many schools is largely traceable 
 to the unsuitable character of the subject matter, 
 which is not only not closely related to the child's 
 immediate interests, but is also not infrequently be- 
 yond the pale of his actual experience. The conse- 
 quence is that the concepts and imagery are lacking 
 in vitality ; the reading is verbal and hopelessly mechan- 
 ical - - "a sounding brass and a tinkling cymbal." 
 
 Besides the selection of subject matter that touches 
 off the springs of the child's immediate interests, there 
 are other devices chiefly in the line of motivation which 
 assist in overcoming die Wanderlust of the youthful 
 imagination and enlist the prompt and vigorous con- 
 centration of the mind. The pressure of a time control, 
 the individual graph, and the class chart, which will 
 be described later, are effective in accelerating the 
 reading rate chiefly through their immediate effect in 
 enlisting the strong concentration of the attention. 
 The continued concentration of the youthful reader is 
 scarcely susceptible to direct coercion ; it must be ef- 
 fected largely by indirect means, such as those above 
 described.
 
 66 SILENT READING 
 
 The significance of strong concentration of attention 
 in increasing the speed of silent reading seems to have 
 been too little recognized in the past. Some of the 
 means utilized in the present investigation owe their 
 incorporation into the types of training mainly to 
 their immediate influence in enlisting the active at- 
 tention of the reader. For even with the other factors 
 present, and a fair degree of concentration lacking, the 
 reading will be necessarily inferior in both quality and 
 rate. In the present study, then, this factor is strongly 
 stressed. 
 
 8. Ability to Grasp the Meaning of Contents 
 
 Since reading is in the last analysis a mental process, 
 namely, the assimilation of the thought from the 
 printed page, it is clear that the perception of the 
 printed symbols- is affected to some degree by the speed 
 with which those symbols can be understood. Con- 
 sequently in the reading of heavy difficult material, 
 the part of the reading process which demands the 
 bulk of the time and effort is probably not the op- 
 eration of "getting the material to the brain, but of 
 assimilating it after it is there." The necessity of 
 making the proper logical associations, of examining 
 the validity of the thought in the light of the principles 
 already acquired, involves reasoning and studying of 
 rather a high character. Such a process, which would 
 seem to be more appropriately termed "studying" than 
 simple "reading," naturally demands additional time.
 
 THE RATE OF SILENT READING 67 
 
 But in the case of reading of the ordinary type of 
 easily comprehended material, rather than in the 
 "studying" process above described, the assimilation 
 of the matter seems to occur as soon as the printed 
 symbols are perceived. There appears to be no ap- 
 preciable retardation of the rate due to the demands of 
 comprehension. The reverse of this process would 
 seem to be true at least in the majority of cases. The 
 rate of comprehension is slowed up, waiting upon the 
 tardier process of visualization, just as the latter is 
 itself impeded in the case of oral reading by the more 
 cumbersome process of vocalization. The introspec- 
 tion of many workers in this field, such as Huey, For- 
 dyce, and of a number of persons interested in the 
 present study, showed that their rate of reading was 
 slow, not because they could not assimilate the matter 
 rapidly, but because their perception of it was proceed- 
 ing at a slow plodding pace. 
 
 Their subsequent determination to increase their 
 reading rate resulted in quicker perception without 
 impairment of the comprehension. These instances, 
 which are typical of hundreds of similar cases, would 
 seem to indicate that the reading rate at least in regard 
 to the ordinary type of material is conditioned by 
 rate of perception as well as by rate of assimila- 
 tion. Most readers, in other words, proceed at a pace 
 that is well within the limits of their possible rate of 
 assimilation, which is seldom given the opportunity of 
 functioning at as rapid a rate as is really possible for it.
 
 68 SILENT READING 
 
 Ruediger (42), however, apparently considers the 
 assimilative factor the determining or "essential" one. 
 As a result of some experimentation he concludes : 
 "After having eliminated the physiological qualities 
 pertaining to the mechanism of vision we saw that 
 neither visual acuity, retinal sensitivity, nor the hori- 
 zontal extent of acute vision had any significant corre- 
 lation with reading rate. The essential factors that 
 determine reading must be looked upon as central 
 rather than peripheral." 
 
 This conclusion is undoubtedly true in regard to the 
 rate at which difficult material can be intelligently read, 
 and possibly it is true in regard to the upper limit of 
 speed at which more easily comprehended material can 
 be perused. In other words an individual's speed of 
 reading can never actually exceed the rate at which it 
 is assimilated. Otherwise it would be simple visualiza- 
 tion, not reading in the intelligent sense. But are there 
 many individuals who have attained a rate of reading 
 which is the upper limit at which the process of assimila- 
 tion can occur? The bulk of available evidence de- 
 mands a strong negative answer. On the contrary it is 
 probable that there are extremely few who have reached 
 the upper limit in their rate of reading, i.e. the highest 
 rate of reading at which the matter can be compre- 
 hended. There has been so much dead-level plodding, 
 and so little recognition of the possibility of increasing 
 the rate, and so few systematic attempts to do so, that 
 it is safe to say that the rate at which most individuals
 
 THE RATE OF SILENT READING 69 
 
 read can be considerably increased by the proper type 
 of training without decreasing the comprehension of 
 the matter read. The results of previous studies as 
 well as those to be presented in the present investiga- 
 tion, showing marked improvement in rate of reading 
 with no impairment of comprehension, constitute the 
 basis for the above statement. 
 
 While admitting that the correlation that has been 
 found to exist between the number of pauses and the 
 reading rate does not apparently square with his con- 
 clusion as to the determining r61e played by the assimi- 
 lative factor, Ruediger maintains that the number of 
 printed symbols perceived during a fixation may be 
 attributed to the central factor of comprehension just 
 as well as to effective perception. But the fact that 
 simple exercises, such as those which Dean Fordyce 
 describes with a first grade reader and with columns 
 of two-word and three-word phrases, and such as 
 have been worked out in the present study, have 
 notably increased the speed, not infrequently dou- 
 bling it, proves conclusively that there are many factors 
 besides the ability to assimilate which condition the 
 rate of reading. 
 
 Such simple exercises cannot presumably have 
 doubled the ability to comprehend, but they have 
 increased the size of the visual span functioning in 
 reading and have improved the character of the 
 habits of eye-movements and other peripheral fac- 
 tors, thus producing a marked increase in rate of
 
 70 SILENT READING 
 
 reading. This conclusion is further corroborated by 
 the results of Quantz's investigation of the relation of 
 visual perception to rate of reading. After pointing 
 out the marked correlation existing between these, 
 Quantz draws the following conclusion : " This shows 
 that mere quickness of perception with its large physi- 
 ological element is an important factor in deciding one's 
 rate of reading." Consequently, while the ability to 
 assimilate is not unimportant, there are other factors, 
 peripheral in character, whose influence in condition- 
 ing the rate of reading is of unquestionable significance. 
 
 9. Recognition of the Value of the Habit of Rapid Silent 
 
 Reading 
 
 Probably the chief reason why comparatively few 
 readers have acquired the habit of rapid silent reading 
 is because of the general failure to recognize the possi- 
 bility of greatly increasing one's rate, and consciously 
 to advert to the value of such a habit. The discovery of 
 the possibility of appreciably accelerating the rate is 
 the result of investigations of comparatively recent 
 date. This knowledge is only beginning to filter 
 through into educational practice and to reach the 
 popular mind. The recognition of the possibility of 
 establishing immeasurably more effective and economi- 
 cal habits of rapid silent reading, and the appreciation 
 of their value, are the first and necessary steps in their 
 acquisition. Once these facts have been clearly per- 
 ceived, the effort necessary to substitute " higher
 
 THE RATE OF SILENT READING 71 
 
 order" habits of reading for slow dawdling ones will be 
 readily forthcoming. 
 
 Accordingly, in the present investigation considerable 
 care was taken to point out, not only to the teacher the 
 possibility and the value of establishing such habits, 
 but also to get this lodged clearly in he minds of the 
 pupils. Unless they perceive the value of such habits, 
 and are made to feel the need of acquiring them, the 
 work is apt to be mechanical and lacking that earnest- 
 ness which is the sine qua non of effective habit forma- 
 tion, and, in fact, of all successful work. On the other 
 hand, if the value of such habits is perceived and the 
 need felt, the successful outcome is already half as- 
 sured. For from such an attitude there springs con- 
 centration of mind, enthusiasm, and sustained interest. 
 Consequently this must be considered no unimportant 
 factor in the acquirement of habits of rapid silent 
 reading. 
 
 10. The Will to Read Rapidly 
 
 Closely related to the preceding factor, and growing 
 logically out of it, is the will or determination to read 
 rapidly. The writer believes it important enough, how- 
 ever, to deserve separate enumeration. The general 
 literature on the subject of reading records several in- 
 stances of persons suddenly awakening to the realiza- 
 tion that their rate was unnecessarily slow. Accord- 
 ingly they resolved rather simply and naturally to 
 bring it up to a higher level. Marked success was
 
 72 SILENT READING 
 
 achieved. Thus Huey (6) owed the doubling of his 
 rate largely to his simple determination to "speed up." 
 The results of this study and the observations of 
 teachers show that as a general rule those pupils who 
 really displayed a determination to improve their rate 
 did so to a notable degree. The pupils who secured the 
 slightest increases were largely those who failed to 
 "warm up" to the experiment and who were lacking 
 in the strong determination to succeed. Indifference 
 to improvement in their habits of reading on the part 
 of individuals may be justly expected to produce in- 
 different results. The observations of the teachers 
 in this experiment offer sufficient corrob oration of this 
 conclusion. 
 
 When first reading Huey's "Psychology and Peda- 
 gogy of Reading" the writer became convinced of the 
 fundamental importance of this factor. The writer's 
 rate in reading Huey at that time was about 35 pages 
 per hour. Realizing that it was unnecessarily slow, he 
 determined to bring it up to a higher level, and before 
 finishing Huey, succeeded in reaching a speed of 50 
 pages per hour without any noticeable impairment 
 of comprehension. It is thought that the incident is 
 but typical of an improvement which can be effected 
 by practically every reader who becomes determined 
 to bring it about. Much of the ordinary reading is of 
 the lackadaisical, dead-level, plodding type broken by 
 many gaps of attention and much unconscious day- 
 dreaming. A sudden determination to "speed up"
 
 THE RATE OF SILENT READING 73 
 
 effects a change in the whole mental attitude ; it shoots 
 the attention rapidly over the subject matter, and 
 replaces listlessness with spirit and dynamic energy. 
 In fact, the statement can probably be made and the 
 results of the present investigation largely bear it out 
 that practically every reader who strongly wills to 
 increase his rate, can actually do so provided, of 
 course, he employs some suitable means. The cases of 
 Huey, of a number of graduate students, of Fordyce, 
 the last of whom also employed supplementary exer- 
 cises, are so many instances in point. 
 
 While the powerful influence exerted by a strong 
 determination in accelerating the reading rate is per- 
 haps rather obvious, the writer believes that this factor 
 hag not been sufficiently stressed, at least explicitly, 
 in most of the literature on the subject. It cannot 
 always be presumed to be present in the pupils. It 
 must be aroused and enlisted in the enterprise. Other- 
 wise, the undertaking in spite of endless technique 
 and elaborately worked-out methods is foredoomed to 
 failure. 
 
 11. The Pressure of a Time Control 
 
 This is a factor which the present investigation found 
 to be of value in stimulating and maintaining a high 
 rate of speed. When the pupil realizes that there is no 
 check upon his performance, no means .of determining 
 his rate, there is the tendency for daydreaming and 
 unconscious wanderings of the attention to occur. Not
 
 74 SILENT READING 
 
 that these disturbing elements are willfully invited, 
 but the mental attitude is such as to offer no strong 
 barrier to their entrance. The consciousness that a 
 clock or watch is accurately measuring the pupil's 
 rate of reading induces a mental "set," which focuses 
 the attention directly upon the task at hand, and which 
 is incompatible both with lackadaisical poring and 
 leisurely dawdling. 
 
 In other words the pressure of a time check serves 
 to "key up" the mind to a superior level of attention, 
 which enables it to devote its whole conscious energy 
 to the gathering of the thought from the printed page 
 and saves it from the necessity of struggling against 
 irrelevant ideas which tend to draw the attention away 
 from the context. This is a device which, as will be 
 pointed out later, is applicable to the three types of 
 training which have been devised to accelerate the 
 reading rate. An ordinary clock will suffice, though a 
 stop watch is more convenient. In any type of train- 
 ing where speed is one of the qualities aimed at, it 
 would seem that this factor cannot well be omitted. 
 
 12. Individual Graph 
 
 An effective presentation of the story told by the 
 clock in measuring the pupil's speed may be secured 
 by drawing the result on a chart. The single line 
 ascending or descending tells the pupil the story of his 
 success or failure. It is simple, impressive, and compre- 
 hended at a glance. There is no need for the comparing
 
 THE RATE OF SILENT READING 75 
 
 of arithmetical figures. The direction or slant of the 
 line tells the whole story. The ascent of the line be- 
 comes a source of joy; its decline a source of grief. 
 The latter tells, moreover, in a convincing manner the 
 necessity for greater effort on the next occasion. The 
 zest of a game, the spirit of a combat is aroused. The 
 pupil becomes determined to "make that line go up." 
 As will be indicated later on, in the exposition of the 
 effectiveness of the various devices employed in the 
 present investigation, the individual graph made one 
 of the strongest appeals to the pupils and proved one 
 of the most effective instruments in stimulating their 
 speed in reading. 
 
 13. The Class Chart 
 
 This served as a supplement to the individual graph. 
 It consisted essentially in portraying the average or 
 median of the class. Besides this the highest and 
 lowest individual record for each day was traced on 
 each chart. This showed the amount of variation 
 and the room for possible improvement open for many 
 members of the class. It had a stimulating effect upon 
 the slowest members of the class, causing each one of 
 them to strive not to be the last, which meant repre- 
 sentation at the bottom of the class chart for that day. 
 A similar effect was produced upon the faster ones, 
 causing each one of them to strive for the honor of 
 achieving the best record with the consequent repre- 
 sentation at the top of the chart for that day.
 
 \ 
 
 76 SILENT READING 
 
 The class chart did not admit, however, of the possi- 
 bilities of the extensive comparison between individual 
 members of the class which were afforded by the in- 
 dividual graph. The consequence was that it did not 
 secure the rather intense motivation effected by the 
 latter. It enabled comparisons to be made, however, 
 between the different classes in the schools. The 
 character of the motive which it engendered differed 
 from that prompted by the individual graph. It 
 fostered interest and loyalty to the class as a whole. 
 It created the desire to see the class as a whole excel, 
 instead of centering the attention solely upon the 
 individual's own performance It was a type of 
 motivation worth developing. It served, furthermore, 
 to hold the interest of the class as a unit in the progress 
 of the experiment. 
 
 To secure the best effect the class chart should be 
 large and should be placed conspicuously in the class- 
 room where pupils can easily see it. They should also 
 be encouraged to look at the chart of the other grade 
 or grades, in order to see where their class stands in 
 comparison. 
 
 In this study most of the class charts were kept 
 by the teacher, who computed the class average or 
 median. She also ascertained the highest and lowest 
 individual rates each day and recorded them on the class 
 chart. In the upper grades the keeping of the class 
 chart may be assigned to one of the more capable pupils 
 if the teacher so desires. It entails but a minimum of
 
 THE RATE OF SILENT READING .77 
 
 labor and exercises a stimulating influence in promoting 
 a commendable esprit de corps which in turn helps to 
 maintain the continued interest of the class as a unit 
 in the progress of the experiment. It is one of those 
 factors which strengthen directly the morale of the 
 group as a whole and exercise a reflex reenforcing in- 
 fluence upon the individual attitude as well. 
 
 14. Reaction Time 
 
 The perceptual span, the fixation-pause, the inter- 
 fixation movement, the return sweep, have all a large 
 physiological core. The general capacity of quickness of 
 perception is doubtless one of the elements conditioning 
 the speed of reading. The sensory data must be trans- 
 mitted over the afferent optic nerves to the visual and 
 thought centers in the brain; the complex response trans- 
 mitted over the efferent neural paths involves as a rule 
 not only the motor aspects of perception but usually 
 those of vocalization as well. This is evidenced by the 
 usual presence of inner speech in silent reading. While 
 it is true that the interfixation movements and the re- 
 turn sweeps consume but a small percentage of the total 
 reading time, it is probable that the neural factor in- 
 fluences the duration of the fixation-pauses as well as 
 the subsequent movement. 
 
 Here, however, as in the case of the central factor, it 
 is doubtless true that the speed of the reading habits of 
 most individuals is well within the upper limit of their 
 neural reaction time. Huey reports the visual reaction
 
 78 SILENT READING 
 
 time to be 206. 90o- (average deviation 20.7(r); while 
 Dodge quotes averages ranging from 151.00<r (mean 
 variation 9.9<r) to 181. 00<r. These appear to be shorter 
 than the average duration of a fixation-pause in reading. 
 After a study of the comparative length of the fixation- 
 pause and the visual reaction time, Schmidt concludes : 
 "It is not probable that the average of the pauses in a 
 given selection would be shorter in duration than the 
 reaction time of the eyes." 
 
 Furthermore, Schmidt, whose technique was well 
 adapted to secure accurate measurement, reports that 
 the rate of the interfixation movements and of "the 
 return sweep is not absolutely constant in the case of 
 any one individual, variations from time to time being 
 by no means uncommon." Here again there is an upper 
 physiological limit to speed of reading, but the rate of 
 an individual's ordinary reading seldom approaches 
 within even striking distance of this maximal neural 
 limit fixed by the rate of the ocular neural conductivity. 
 Whether this neural reaction time is itself subject to 
 improvement through training is an interesting question 
 upon which the present study seeks to throw some light. 
 
 15. The Visual Type of Imagery 
 
 In the literature of the psychology of reading there 
 are found some writers who think that the speed of 
 reading is largely determined by the subject's dominant 
 type of imagery. Quantz (26) interprets the results 
 of some of his experiments to indicate the superior ra-
 
 THE RATE OF SILENT READING 79 
 
 pidity of readers employing the visual type of imagery 
 over those using the auditory type. Quantz further 
 concludes that u the motor tendency in any degree has 
 an influence detrimental to the rapidity of reading, and 
 the stronger the tendency, the greater the hindrance." 
 Schmidt states quite definitely that the speed of read- 
 ing is conditioned by the dominant mode of imaging. 
 The visual type constitutes the rapid readers, the motor 
 type the slowest, while the "auditory type ranges be- 
 tween the two." As the whole question of imagery in 
 reading is closely tied up with the problem of speed 
 in reading, a more detailed consideration of the influence 
 of the various types of imagery upon the reading rate 
 and an examination of the evidence on this point will 
 be found in Chapter VI - - Training in Perception. 
 
 Miscellaneous 
 
 Besides the factors just enumerated there are others 
 whose influence* upon the rate of reading has engaged 
 the attention of various investigators. Beer, Messmer 
 (43), and Dearborn investigated the influence of the 
 length of words upon the rate. Both Beer and Mess- 
 mer found that a passage in which monosyllabic words 
 predominated required more time. This fact was as- 
 cribed to the relatively greater number of ideas con- 
 tained therein. Dearborn, however, maintains: "It is 
 not the short words as such, but the words which cannot 
 easily be grouped with others which necessitate separate 
 fixation."
 
 80 SILENT READING 
 
 The investigation conducted by Huey to determine 
 the effect of changes in the size of the type upon the 
 rate shows that the unit of recognition of reading is 
 but very slightly affected by changes in the sensory 
 content, e.g. changes in the size of the type do not 
 cause proportionate changes in the number and length 
 of the pauses. Consequently, the rate of reading is but 
 very slightly affected. These last mentioned investi- 
 gations and others involving relative legibility of letters 
 have not, in the judgment of the writer, disclosed 
 factors of any appreciable influence upon the rate which 
 could be incorporated into types of training to increase 
 the speed in silent reading. 
 
 It is thought that every factor which experimental in- 
 vestigation or strong a priori reasons have indicated to 
 be of consequence in conditioning the rate of reading has 
 been enumerated above. The attempt has been made 
 to blend the factors into various methods of training to 
 accelerate the rate in silent reading. ' Three distinct 
 methods have been formulated. They possess many 
 factors in common. They differ from one another 
 chiefly because each method is based upon a distinct 
 principle of training. 
 
 By applying these methods of training to hundreds 
 of pupils in various grades of the elementary schools, 
 three to eight inclusive, evidence of an empirical nature 
 will be secured to ascertain how speed can be developed 
 and to determine some of the factors conditioning its 
 development. The application of these types of train-
 
 THE RATE OF SILENT READING 81 
 
 ing to pupils in the schoolroom serves therefore as an 
 actual test of the efficacy of the previously mentioned 
 factors and renders possible the valuation of their effec- 
 tiveness. The formulation of the three types of 
 training which have been developed will accordingly be 
 presented in the following chapters.
 
 CHAPTER IV 
 TYPE I. TRAINING IN RAPID SILENT READING 
 
 ON the general principle of psychology that one 
 learns to do a thing by actually doing it, it would seem 
 logical to assume that one could learn to read rapidly 
 by practice in rapid reading. In contrast with methods 
 which analyze the reading complex into its main pro- 
 cesses, and then devise types of training adapted to 
 improve each of these specific processes separately, 
 this may be said to be the direct synthetic mode of 
 attack training in the whole organized process of 
 reading itself. This would seem to be the simplest 
 and, at the same time, the most indispensable type of 
 training. 
 
 In the case of the specialized type of training adapted 
 to secure the improvement of a single specific process, 
 such as visual span, or character of eye-movement, in 
 the reading complex, there is always the possibility that 
 the improvement effected under controlled circum- 
 stances may not be carried over into the actual reading 
 situation, where the circumstances affecting the func- 
 tioning of a single isolated process can no longer be con- 
 trolled. This may be due, among other causes, to the 
 
 82
 
 TRAINING IN RAPID SILENT READING 83 
 
 change in the reading situation which produces a corre- 
 sponding modification in the response ; or it may be due 
 to the simultaneous functioning of other processes in the 
 complex which tend more or less to inhibit or otherwise 
 to affect the mode of operation of the specific factor sub- 
 jected to the specialized training. Thus, for example, 
 training to increase the perceptual span by means of 
 tachistoscopic exposures of single disconnected phrases 
 may actually at least in the case of children suc- 
 ceed in enlarging the visual span for such a type of read- 
 ing. But whether the visual span enlarged for that 
 specific type of perception would carry over into a 
 different situation, such as obtains in the reading of a 
 selection of continuous closely related subject matter, 
 is uncertain. 
 
 For, in the latter case, the situation differs consider- 
 ably from the former. The subject matter is no longer 
 divided into phrases adapted for perception in a single 
 fixation, but has all the words closely following one an- 
 other in the printed line. There is, furthermore, a cen- 
 tral thread of thought running through the various 
 phrases and sentences of the context, which furnishes 
 meaning premonitions of the coming words and phrases, 
 thus enabling them to be grasped in much less than the 
 normal perception time. Their meanings are more 
 or less anticipated and peripheral vision seems sufficient 
 to confirm the correctness of the conjecture, without 
 requiring direct fixation of the foveal area. The well- 
 nigh complete absence of these meaning premonitions
 
 84 SILENT READING 
 
 in the reading of discrete words, or phrases, exposed 
 serially by means of the tachistoscope, shows how 
 radically one reading situation may differ from another. 
 It illustrates, moreover, the difference in the factors 
 which are called into play, as well as the difference in 
 their mode of functioning in the interpretation of the 
 printed symbols, according as the reading situation 
 varies. 
 
 Considerations such as the above serve to emphasize 
 the value of the direct method of attacking the problem 
 of accelerating the rate in silent reading. They show 
 the need of training in various reading situations by 
 supplying different types of material and presenting 
 different aims or purposes for which the reading is to 
 be done. They indicate, likewise, the necessity of em- 
 ploying certain precautionary devices to effect the 
 transfer of the improvement in the functioning of a 
 single specific process in the reading complex which 
 was secured under controlled conditions, over into the 
 ordinary type of reading. In the present investigation 
 the specialized types of training have, in every case, 
 been supplemented with practice in actual rapid read- 
 ing in order that the improvement effected in a single 
 factor may be rendered more likely to persist in ordinary 
 reading. 
 
 The general question of the transfer of specialized 
 training over into totally distinct and different mental 
 operations is not raised here. All the specific processes 
 that have been singled out for training in the present
 
 TRAINING IN RAPID SILENT READING 85 
 
 investigation have not only a close kinship but many 
 elements identical with the ordinary reading complex. 
 The manner of effecting a persistence of the improved 
 functioning of a single process becomes largely a 
 matter of adapting the supplementary reading situa- 
 tion so as to evoke the desired type of functioning of 
 the factor subjected to the previous specialized train- 
 ing. This will be exemplified in the subsequent exposi- 
 tion of the specialized types of training No. II and 
 III. These considerations are mentioned here to 
 differentiate clearly the types of training developed in 
 this study, and to point out the characteristic of this 
 first type of training, namely, its direct synthetic mode 
 of attack. 
 
 The present habits of reading are largely the result 
 of indifferent practices. Up until the last few years 
 there has been little or no recognition of the value of 
 speed in reading and consequently but little or no 
 systematic attempt to acquire it. The school's stress 
 upon oral exhibition has excluded such an attempt. 
 The first step in the development of habits of speed is 
 the breaking up of the present slow, plodding habits of 
 reading. Habit, as James has pointed out in his classic 
 treatment of this subject, is best overcome by a con- 
 trary habit. So, in reading, slow, dawdling habits are 
 best destroyed by practice hi the opposite type of 
 activity rapid, alert reading until that type be- 
 comes habitual. 
 
 To hasten the establishment of such habits and the
 
 86 SILENT READING 
 
 breaking up of the defective ones, the reading should 
 be done at top speed. This will probably cause some 
 difficulty at first and be somewhat trying. The latter 
 seems to be the inevitable accompaniment of the sub- 
 stitution of one type of habit for its opposite. Read- 
 ing at maximum speed, however, seems to be more 
 effective and more prompt in the establishment of the 
 desired type of ocular motor reactions than the 
 "gradual tapering off" of the slow dead-level plodding 
 pace. 
 
 To secure the highest speed and to prevent a relapse 
 into the slower eye-movement habits, certain devices 
 are essential. First, the reading stretch should be brief. 
 Preliminary experimentation showed that it was very 
 difficult to maintain the highest pitch of speed for more 
 than three or four minutes. The element of fatigue 
 quickly sets in where the effort is of the most intensive 
 sort and causes a decline in the rate of speed. Ac- 
 cordingly, in the present investigation, the reading 
 period was divided into stretches lasting usually two or 
 three minutes with an occasional variation of a four- 
 minute period. The instructions stated that a reading 
 stretch should seldom, if ever, exceed four minutes. 
 The pressure of a time-control, the individual graph, 
 and the class chart were likewise helpful in maintain- 
 ing the speed at the highest rate of which the readers 
 were capable at the time. 
 
 All the factors enumerated in Chapter III, with the 
 exception of No. 2 Training to Decrease Vocaliza-
 
 TRAINING IN RAPID SILENT READING 87 
 
 tion, and No. 3, Training in Perception, have been 
 incorporated into this type of training in rapid, silent 
 reading. The inclusion of all these factors, and the 
 presentation in brief outline of the resultant method 
 proved a rather difficult task. While the adequate 
 exposition of many of the factors involved, and of 
 their relative importance, would have required con- 
 siderable space, it was felt that the statement of the 
 method, which was to serve the teachers also as a set of 
 "directions for procedure," should be brief. It is, at 
 best, a somewhat difficult and delicate undertaking 
 to translate a set of written directions into the correct 
 mode of procedure. If the directions are involved and 
 complicated, the difficulty is enhanced, and the chances 
 for confusion and divergences of procedure are corre- 
 spondingly increased. Accordingly, an attempt was 
 made to present the essentials of the method in the 
 following brief outline or statement of Type I - 
 Training in Rapid Silent Reading. A copy of these 
 directions was sent to each teacher using this method. 
 
 STATEMENT OF 
 TYPE I. TRAINING IN RAPID SILENT READING 
 
 For the Teacher. The purpose of this investigation is to 
 determine the extent to which speed in silent reading can be 
 increased by training the pupils in this practice for a specified 
 length of time. The practical value of a type of training 
 which will accelerate the rate of silent reading is obvious. 
 Your cooperation in this study will aid in definitely ascer-
 
 88 SILENT READING 
 
 taining means of accomplishing this end. The results of this 
 investigation will be sent to every teacher participating in 
 the work. 
 
 The period of training in rapid silent reading should last 
 rom April 8 until May 29. It should replace the regular 
 work in reading and no other classroom time should be de- 
 voted to reading. Thirty minutes per day should be allowed 
 for the work. Since the time element enters into all the 
 work, a clock should be placed in the front of the classroom so 
 that it can be seen by all the pupils. The method should 
 consist essentially of alternate reading and reproduction. 
 The reading should be timed and a measure of the amount 
 read in the time allowed should be regularly taken. Pupils 
 should be kept informed of their speed of reading. Repro- 
 duction should consist both of free paraphrase orally, 
 or in writing and of answers to specific questions based 
 on the text. The length of the reading period and of the 
 reproduction period should vary with the grade of the pupils 
 and with the subject matter. In general, however, the 
 reproduction should not occupy more than one-quarter of the 
 total time allowed for the exercise. 
 
 Only interesting material should be selected. It should 
 also be easily within the understanding of the pupils. Since 
 the object is to set up habits of rapid reading, emphasis upon 
 the simplicity of the selection is necessary. 
 
 In order to assist the pupils in covering as much reading 
 matter as possible, a definite preparation may be made con- 
 sisting : (a) of thought preparation, or (6) of word prepara- 
 tion, or (c) of both. The thought preparation should con- 
 sist of such an introduction by commentary or by the ques- 
 tion-answer process as will arouse interest and enlist the 
 attention of the pupils. This preparation should in no case 
 be long. For the word preparation, the teacher should 
 select such words as in her judgment would be unfamiliar to
 
 TRAINING IN RAPID SILENT READING 89 
 
 the pupils, and should present them briefly, explaining their 
 meaning. The preparation both thought and word 
 may be abridged or even omitted when the material is such 
 as to give no difficulty. In no case should the preparation 
 occupy more than five minutes of the 30 assigned to the exer- 
 cise. 
 
 Each pupil should keep a chart of his daily performance, 
 and a complete chart of the daily class performance should 
 be conspicuously displayed in the classroom. 
 
 It is desirable that the directions to pupils given by each 
 teacher concerned in this investigation be substantially the 
 same. The following points are suggested indicating the 
 nature of the direction. Literal adherence to them is not 
 requested. Their spirit, however, should be maintained. 
 
 Point out the advantage of a rapid rate of reading. Try 
 to get the children to see this in terms of an addition to their 
 own interests and pleasure. Carefully direct them where 
 to begin the assignment of the day. See that they all begin 
 at the same tune and at the same place in the text. Say in 
 substance : " Read it as fast as you can. I want to see how 
 much you can read in two (or three) minutes. But remem- 
 ber that I am going to ask you to tell me about what you 
 have read so do not skip anything. Try to read faster to-day 
 than you did yesterday." See that the pupils have a pencil 
 at hand and direct them to stop reading at once as soon as 
 you say "stop." Direct them then to mark the end of the 
 line which they are reading when told to stop. Pupils may 
 now reproduce what they have read as indicated above. In 
 a similar manner, reading and reproduction are to be con- 
 tinued till the end of the 30 minutes assigned. Have the 
 pupils then report the number of pages and lines beyond the 
 last full page which they read. 
 
 As part of your preparation for the exercise you will be 
 expected to know the average number of words per line in
 
 90 SILENT READING 
 
 the matter which is being read and the number of lines per 
 page (if pages are broken by illustrations, or for other reasons, 
 special account of these pages will have to be taken). From 
 the pupils' reports as to pages and lines read, on this and 
 subsequent assignments for the day, estimate the number 
 of words read that day and divide by the total number of 
 minutes used in reading. This will give the number of words 
 read per minute. Each pupil should figure out his own score, 
 and should immediately enter it upon the chart, which will 
 thus serve as a record of his daily progress in silent reading. 
 
 On May 1, and May 28, besides the record of the aver- 
 age number of words read per minute by each pupil during 
 the total reading time, a record of the average number of 
 words read per minute by each pupil during each of the 
 various reading periods constituting the total time devoted 
 to reading on each of the above mentioned days should also 
 be kept. 
 
 On May 2, and May 29, the pupils should be tested 
 with the Courtis Silent Reading Tests, Forms II and III, 
 respectively. As soon as possible after May 29, a copy 
 (or the originals) of the individual charts, the class chart, 
 and a record of results and observations should be mailed 
 to the Bureau of Educational Research, University of 
 Illinois, Urbana, Illinois. 
 
 To clear away any misconceptions in regard to the 
 various points in the statement of the method, and to 
 insure uniformity in all the important steps in the 
 procedure, a representative from each of the school 
 systems participating in the experiment was invited 
 to a conference at the University of Illinois. Here the 
 application of the method was demonstrated to the 
 visiting delegates by a teacher and her pupils from
 
 TRAINING IN RAPID SILENT READING 91 
 
 one of the Champaign public schools. Miss Davis, the 
 teacher demonstrating the method, had previously 
 been instructed in its application by Dr. Buckingham 
 and the writer. Besides the ocular demonstration, 
 queries from the delegates concerning various details 
 of the method were answered, and its adjustment to 
 particular classroom conditions was explained. 
 
 As a result of the questions which arose at this con- 
 ference, the following topics seem to require additional 
 discussion : (l) Length of Reading Period, (2) Alter- 
 nate Reading and Reproduction, (3) Thought Prepara- 
 tion, (4) Word Preparation, (5) Directions to Pupils. 
 
 1. Length of Reading Period 
 
 Thirty minutes was specified as the length of the 
 reading period for all the grades participating in the 
 experiment. This amount was decided upon after 
 investigations showed a range from approximately 
 120 minutes devoted to reading in some third grades to 
 only about 15 minutes allotted to reading in the seventh 
 and eighth grades in some schools. Furthermore, 
 considerable variation in the amount of time devoted to 
 reading in even the same grades was found to exist 
 among school systems in different cities. Since the 
 investigation extended to all the grades from the third 
 to the eighth, a reading period that was to obtain 
 throughout these six grades would necessarily have to 
 be somewhat of a compromise between the extremes. 
 The heavy curriculum of an eighth grade precluded the
 
 92 SILENT READING 
 
 possibility of devoting an hour and a half daily to 
 reading. On the other hand, the teachers in the third 
 grade felt that the abbreviation of the reading time 
 to a single 15-minute period could not be done without 
 causing a rather serious impairment of the progress of 
 the pupils. Accordingly, a 30-minute period was de- 
 cided upon as representing an approximate average of 
 the amount of time devoted to reading by the six upper 
 grades in the elementary schools. 
 
 The allotment of this specific amount of time carries 
 with it, therefore, no implication of a normative char- 
 acter. It does not pretend to represent the proper 
 length of the reading period for all the six grades men- 
 tioned. The amount of time assigned to reading will 
 naturally vary from grade to grade, less time being 
 allotted to the formal reading period in the upper than 
 in the lower grades. A reading period of 30 minutes 
 was decided upon in the present investigation solely 
 for the sake of uniformity. For, obviously, the amount 
 of time devoted to the training must be uniform if 
 comparisons are to be made between the amounts of 
 progress made by the different grades. This uniformity 
 in the allotment of time renders possible a comparison 
 of the extent of gain registered by the third grade 
 pupils with that made by, say, the fifth or eighth 
 grade pupils. Without this uniformity in the length 
 of the reading period throughout all the grades a com- 
 mon basis of comparison would be wanting. 
 
 Comparisons, such as the above, may be expected to
 
 TRAINING IN RAPID SILENT READING 93 
 
 throw some light on the interesting question : In 
 what grades do the pupils show the greatest susceptibil- 
 ity to improvement in reading rate ? From the answer 
 to the above question there flows a corollary of practical 
 value to the school concerning the relative stress 
 to be placed upon the formation of habits of rapid 
 silent reading in the different grades of the elementary 
 school. 
 
 2. Alternate Reading and Reproduction 
 
 While this study is concerned chiefly with the accelera- 
 tion of the rate, it is recognized that the development of 
 speed at the expense of comprehension does not make 
 for efficiency in reading. Accordingly, means were 
 adopted not only to safeguard the comprehension and 
 to prevent its impairment, but also to effect, if possible, 
 a development in assimilative capacity that should ad- 
 vance pari passu with the growth in speed. For this 
 purpose after each reading stretch of two or three 
 minutes the pupils were called upon to reproduce. 
 Recognizing that reading in actual life situations is 
 done not for one constant adamantine purpose, but for 
 a great variety of ends, such as to "get the gist," to 
 answer certain specific questions, to get the central 
 thought, to secure a certain point of information, the 
 reproduction in the present study was made to assume 
 a like variety of forms. Sometimes a pupil would 
 be called upon to reproduce the main thought of the 
 selection ; at other times specific items of information,
 
 94 SILENT READING 
 
 important facts, names, and causes, would be requested. 
 Thus the development of a single, narrow type of 
 reading ability was guarded against, and a rather well- 
 rounded, many-sided reading activity was brought into 
 play. 
 
 The teacher was advised to have her questions and 
 her other methods of testing the comprehension formu- 
 lated, at least to some degree, in advance. Preliminary 
 experimentation showed that no small amount of time 
 was lost by the teacher in deciding what to ask about 
 the passage just read by the pupils. This hesitancy 
 and indecision slowed up the whole procedure, consumed 
 a considerable share of the time, and reacted unfavor- 
 ably upon the "keyed-up" mental "set" of the pupils. 
 By having the questions, or at least the cues to them, 
 written in the margin of the book, this source of wastage 
 was eliminated. 
 
 The teachers were advised to scatter the questions 
 promiscuously among the class, so that every pupil 
 would feel himself constantly exposed to a call. This 
 would tend to prevent skipping or skimming and insure 
 perusal. If a pupil were found unable to answer a 
 simple question on the matter he purported to have 
 read, the teacher was encouraged to test him in the 
 subsequent reproductions to show him the need for 
 greater comprehension in his reading as well as to 
 prevent cheating. 
 
 Because of the large element of guesswork which they 
 invite, questions of a simple "Yes" or "No" variety
 
 TRAINING IN RAPID SILENT READING 95 
 
 were not encouraged. Sometimes the reproduction 
 was in writing. This insured the testing of each pupil. 
 Usually, however, oral reproduction was called for. 
 Since the development of habits of speed was especially 
 aimed at, the suggestion was offered that the reproduc- 
 tion should not, as a rule, be too detailed demanding 
 nonessential particulars and points which a reader in 
 an actual life situation would seldom be expected to 
 remember. Such a procedure would naturally militate 
 against the formation of habits of rapid assimilation 
 by its disproportionate and unnatural stress upon un- 
 important details. It would tend to eliminate the 
 functioning of extra-foveal vision, by necessitating 
 the careful and even scrupulous fixation of each 
 word. A slow, plodding type of reading, devoting as 
 much time to the fixation of a mere preposition or 
 article as to the subject of the sentence would be 
 the not unlikely result of such a procedure. 
 
 Preliminary experimentation disclosed the lurking 
 of another danger in the application of the method. 
 When the method was first tested out, a tendency was 
 observed for the teacher to consume the bulk of the 
 period with the mechanical details, getting the class 
 ready to read, reproducing the material read, and 
 computing how much was read. Under these cir- 
 cumstances, a comparatively small portion of the 
 total time was devoted to actual practice in rapid 
 reading. Thus, in the preliminary experimentation, 
 a teacher was found to consume four minutes in pre-
 
 96 SILENT READING 
 
 paring the class to begin reading, and ten minutes in 
 each of the two reproductions which consisted of 
 halting questions and slower answers. The result 
 was that she secured but two reading stretches of three 
 minutes each. Thus, out of a 30-minute period, but 
 six minutes were devoted to actual reading. This was 
 distinctly contrary to what was intended. Accordingly, 
 the instructions stressed the necessity of prompt, rapid 
 reproduction. 
 
 The reproduction should be short and brisk just 
 sufficient to convince the teacher that the pupils have 
 grasped what they have read. While no ironclad rule 
 was laid down specifying the amount of time to be used 
 for each reproduction, the general direction was given 
 that the reproductions all together should not consume 
 more than one-quarter of the total reading period. The 
 short periods for reproduction following each reading 
 stretch afforded the pupils an opportunity for tempo- 
 rary relaxation, thus preventing the onset of fatigue 
 and rendering it possible for them to maintain the 
 reading rate at its highest pitch. 
 
 3. Thought Preparation 
 
 The introspection of trained students as well as 
 casual observation show that most readers are slow in 
 getting started. There usually occurs what psycholo- 
 gists call the "warming up" (Anregung) period, before 
 the individual gets into "the swing" of it (Gewohnung). 
 The length of this warming up period varies with
 
 TRAINING IN RAPID SILENT READING 97 
 
 different individuals and with different materials. That 
 such a warming up process is not necessary, how- 
 ever, was made evident by preliminary experimentation 
 on this point. The pressure of a time-control proved 
 instrumental in replacing the listless attitude with one 
 of prompt, concentrated attention. At the sound of 
 the word "Go," the reader started at a high rate of 
 speed. A record of the number of words read during 
 each of the minutes in a five-minute period shows that it 
 is possible for a reader to strike practically his highest 
 level of speed in the initial minute. The reason why 
 the reading often starts at a slow dawdling pace is 
 because the interest is not yet aroused. In the present 
 study, the pressure of a time-control was chiefly instru- 
 mental in securing prompt concentration of attention. 
 As an auxiliary to this device, however, the thought 
 preparation was utilized to enlist the immediate interest 
 of the reader in the selection. The thought preparation 
 was of different kinds, varying largely with the char- 
 acter of the subject matter. Thus, in a narrative selec- 
 tion, sometimes it was found advisable to trace in a few 
 deft sentences the outline of the story leading up to the 
 climax and then suddenly stopping. The children in 
 many such cases could hardly wait for the signal to 
 begin, so anxious were they to "find how the story 
 came out." Sometimes the end could be attained by 
 saying a few words about one of the main characters, 
 by sketching the background of the story, or by briefly 
 outlining the life of the author. The character of
 
 98 SILENT READING 
 
 the thought preparation was found, furthermore, to 
 depend upon the general interest manifested by the 
 pupils in the training to increase speed. In many 
 classes, where the pupils entered whole-heartedly into 
 the experiment, no such device was needed to enlist 
 the already keen interest of the pupils. The method 
 was left rather flexible to enable the teacher to adapt 
 it to suit the different types of subject matter and the 
 varying needs and interests of the class. 
 
 4. Word Preparation 
 
 In the lower grades especially, it was found advisable 
 to devote a few minutes to a consideration of the mean- 
 ing and the pronunciation of some of the more difficult 
 or unusual words which the pupils would encounter 
 in the selection. By writing them on the blackboard 
 along with their meanings, before the pupils com- 
 menced their reading, the obstacles which might have 
 retarded their progress were eliminated in advance. 
 In the upper grades but little of this word preparation 
 was found necessary. As a rule, it occupied but a few 
 minutes. In no case was the preparation both 
 thought and word permitted to consume more than 
 five minutes of the thirty assigned to the exercise. 
 
 5. Directions to Pupils 
 
 Considerable care was taken to point out to the pupils 
 the advantages of rapid, efficient reading. They were 
 made to feel the need for such training by awakening
 
 TRAINING IN RAPID SILENT READING 99 
 
 in them the consciousness of their present slowness and 
 inefficiency. From this need there sprang the earnest 
 desire to secure such training. They were given 
 plainly to understand that in the measure in which they 
 entered earnestly and enthusiastically into the under- 
 taking, just in that measure would they reap the results. 
 
 In an endeavor to establish regular, rhythmical habits 
 of eye-movements and to utilize to some extent periph- 
 eral vision the pupils were advised to "run their eyes 
 along the line" as fast as they could grasp its meaning. 
 The pupils were urged to put forth their best efforts to 
 read as fast as possible. The Board of Education of 
 New York advises that pupils be urged to read rapidly, 
 but not to hurry. This distinction without a difference 
 found no application in the present set of instructions. 
 The pupils were frankly urged to "speed up" their 
 reading. The reservation adopted here, however, was 
 that they understood what they read. They were free 
 to hurry as much as they could, provided they grasped 
 the meaning. The highest possible rate of speed con- 
 sistent with an assimilation of the subject matter was 
 the ideal placed before each pupil. This direction had 
 its basis in the conviction that such a procedure would 
 be the most prompt and efficacious in breaking up the 
 existing slow, plodding habits of reading and in estab- 
 lishing in their stead the opposite type of reading habits. 
 
 It is thought that the above statement contains the 
 essentials of Type I. The results of the application of 
 this method will be presented in a subsequent chapter.
 
 CHAPTER V 
 
 TYPE II. TRAINING TO DECREASE 
 VOCALIZATION 
 
 WHAT is the nature of the so-called "inner speech" 
 of silent reading? Is it always present in such read- 
 ing? Why is there such a process as vocalization in 
 silent reading at all? What is its origin? What role 
 does it play in the gathering of thought from the 
 printed page? Is it necessary in reading, or may it 
 be eliminated, or at least abbreviated? If so, how? 
 While these questions are of interest, especially to the 
 psychologist, a brief treatment of them here will enable 
 the general reader to appreciate the significance of this 
 factor on silent reading, and to understand the raison 
 d'etre of this second type of Training in a study that 
 aims at the acceleration of the reading rate. Many 
 of the above questions cannot, it is true, be answered 
 as yet with certainty; but recent investigations have 
 served to throw some interesting light upon this rather 
 strange and mysterious accompaniment of silent reading. 
 
 DEFINITION 
 
 Huey's definition of inner speech in silent reading as 
 "a combination of auditory and motor elements, with 
 
 100
 
 TRAINING TO DECREASE VOCALIZATION 101 
 
 one or the other predominating according to the 
 reader's habitual mode of imaging" may be said to 
 reflect fairly well the general view of students who have 
 investigated this process. In other words, reading is 
 not confined to the visualization of the printed symbols. 
 Concomitant with this visualization there occur move- 
 ments, more or less incipient in character, of the tongue, 
 lips, vocal chords, larynx, inner palate, throat, and the 
 general physiological mechanism that functions in oral 
 speech. The reader goes through the form of saying 
 the words to himself. The difference between the 
 inner speech of silent reading and the oral speech of 
 conversation is thus seen to be one, not of kind, but of 
 degree degree of movement of the vocal organs, and 
 degree of sound produced. 
 
 HISTORY OF THE STUDY OF "INNER SPEECH" 
 
 While this habit of vocalization in silent reading 
 and even in thinking seems to have been always with 
 us, its existence does not seem to have been consciously 
 adverted to and commented upon until the second half 
 of the last century. Ribot (44) seems to have been 
 one of the first to have called attention to this phenom- 
 enon. Writing in the Revue Philosophique in 1879, 
 Ribot observes " L'homme fait, qui lit silencieusement, 
 accompagner chaque perception visuelle d'un mouve- 
 ment secret d'articulation." 1 This observation occurs 
 
 1 " In silent reading, every visual perception is accompanied by a 
 concealed articulatory movement."
 
 102 SILENT READING 
 
 as an obiter dictum, as Ribot was primarily interested in 
 demonstrating the general importance of movement 
 in connection with psychical processes. The case of 
 inner speech in reading is cited merely as a specific 
 illustration of the general law of psycho-physics, by 
 which every sensory stimulus terminates in a motor 
 reaction. In the literature in English the first clear 
 reference to the existence of this habit appears to be the 
 statement of Bain (45) in 1868 : "A suppressed articu- 
 lation is, in fact, the material of our recollection, the 
 intellectual manifestation, the idea of speech." In the 
 statements of these psychologists one can see the fore- 
 shadowing of the present pragmatic theory of con- 
 sciousness, in which the incipient motor tendencies of 
 the type just mentioned are made to play stellar roles 
 in explaining the functioning of the various types of 
 thought and ideation. 
 
 The first systematic treatment accorded this process 
 seems to have come from the pens of the French psychol- 
 ogists, Egger (46) and Ballet (47). Relying on intro- 
 spection, Egger notes the constant persistence of this 
 inner speech in both his thinking and reading. The 
 latter he thus aptly describes, as quoted by Pintner (48) : 
 "Lire, en effet, c'est traduire Tecriture en parole:" 1 
 and of thinking, he says, "A tout instant, 1'ame parle 
 interieurement sa pensee." Ballet calls attention to 
 the additional factor of audition in reading, pointing 
 
 1 " To read is, in reality, to translate writing into speech." 
 
 2 " In every instance, a person expresses his thought in inner speech."
 
 TRAINING TO DECREASE VOCALIZATION 103 
 
 out the intimate connection between the articulation 
 of words and the hearing of them. Whether or not 
 audition always follows on the articulation of words, 
 Ballet does not state. 
 
 Somewhat more positive and dogmatic was the con- 
 clusion reached by the German psychologist Strieker 
 (49) that it was impossible to have an idea of a word 
 without experiencing the sensations of innervation 
 arising from the stimulation of the articulately muscles 
 in inner speech. For example, it is impossible to have 
 an idea of the sound of the letter B without feeling an 
 incipient muscular movement in the lips. As he him- 
 self expressed it : "Die Vorstellung des Lautes B und 
 des Gefiihl in der Lippen sind also in meinem Bewusst- 
 sein unzertrennlich assoziert. . . . Diese Gefuhle 
 sitzen in den Muskeln." Indeed, not only are sensa- 
 tions of movements in the articulatory muscles in- 
 separately connected with the letter or word, but the 
 consciousness of these muscular movements really con- 
 stitutes the idea of the word. Since Strieker supple- 
 mented the results of his own introspection by ques- 
 tioning a hundred other observers, and finding similar 
 results, his conclusions assumed the nature of general- 
 izations of a rather universal character. To the ques- 
 tion then, Is inner speech necessary in silent reading 
 and even in thinking ? Strieker answers unequivocably 
 in the affirmative. 
 
 1 "The mental image of the sound B and the (corresponding") sensation 
 in the lips are inseparably associated in my consciousness. . . . These 
 sensations arise from the muscles."
 
 104 SILENT READING 
 
 Stumpf, Paulhan, and Baldwin agree with Strieker 
 that abridged articulately movements are usually 
 present ; but that they are necessary, or even, de facto, 
 always present, they deny. Pushing Strieker's theory 
 concerning the necessity of articulatory movements to 
 secure the idea of a word to its logical conclusion, that 
 the idea of a tone must likewise be impossible without 
 the corresponding articulatory movements, Stumpf (50) 
 endeavors to refute it by citing his own ability and that 
 of other musicians to recall a tone without movement 
 of the articulatory muscles. "Ohne lautes, leises oder 
 stilles Singen kann ich verschiedene Tone vorstellen." 1 
 Paulhan instances his ability to have an auditory image 
 of one vowel while pronouncing a different one aloud 
 an impossible performance according to Strieker's 
 conclusion. Paulhan points out that Strieker's con- 
 scious advertence to the articulatory movement was 
 probably instrumental in causing the movement. 
 While previous writers had called attention to the close 
 connection between the motor and the auditory ele- 
 ments in inner speech, but had given no definite answer 
 as to whether the one could exist without the other, 
 we find Paulhan (51) stating that the auditory may be 
 present even though the motor be lacking. Whether 
 the motor element may exist without arousing the 
 auditory, Paulhan does not decide. Differing some- 
 what from Egger, he maintains that thought is an inner 
 
 1 " I can recall different tones without singing loudly, softly, or 
 silently."
 
 TRAINING TO DECREASE VOCALIZATION 105 
 
 language which need not necessarily be converted into 
 words or verbal imagery "la pens6e est un langage, 
 non une parole, et, si la representation des mots lui 
 est utile, elle parait, de son c6te, faciliter beaucoup 
 cette representation." l 
 
 In general consonance with the conclusions of 
 Paulhan, Baldwin (52) cites his ability to image a note 
 while in the very act of uttering another vocal sound in 
 a different pitch. He does not find the movement of the 
 articulatory muscles a necessary condition for the 
 imaging of the word the latter being produced even 
 when the articulatory muscles are held rigidly motion- 
 less. Baldwin states that when his attention is with- 
 drawn from the larynx and directed to the ear, the 
 movement of the former disappears. In view of the 
 faint vestigial character of the movement, however, the 
 question might well be raised here, whether the move- 
 ment really disappeared, or whether it was merely not 
 perceived because the attention was withdrawn from it. 
 In the writer's judgment, the latter is only too probably 
 the case. Bastian (53) and Collins (54) oppose the 
 theory of Strieker and Bain on more purely anatomical 
 evidence. The latter instances a case of cortical motor 
 aphasia in which articulatory movement was entirely 
 lacking though the patient was nevertheless able to 
 read. 
 
 In an introspective study of his own silent thinking, 
 
 1 " Thought is a language, not a speech, and if the verbal imagery is 
 useful, the idea seems in its turn to facilitate greatly the appearance in 
 consciousness of such imagery."
 
 106 SILENT READING 
 
 Dodge (55) found inner speech was clearly present. 
 Reproductions of movement sensations from the vari- 
 ous organs that function in actual speech the tongue, 
 lips, throat, thorax, seem to constitute the essen- 
 tial elements of the words. An interesting distinction 
 is drawn by Dodge between the sensations coming 
 from the actual movements of the organs of speech and 
 the ideas of those movements. The latter Dodge 
 maintains are indispensable; the former are not. 
 Bawden (56) supports this conclusion, also maintaining 
 that kinsesthetic or motor ideas suffice to constitute 
 the meaning of words, even when actual articulation 
 or movement of the muscles is absent. 
 
 Thus far the psychologists had relied solely upon 
 introspection to detect the presenee of articulatory 
 movement. An effort was made by Curtis (57) to re- 
 move this matter from its complete dependence upon 
 the subjective factor of the subject's own introspection 
 and place it upon an objective basis. Accordingly, 
 he placed a large tambour on the larynx of the subject. 
 A record of the movements made while the subject was 
 thinking or reading silently was compared with the 
 movements made when the subject was relaxed, think- 
 ing of nothing in particular. The curves of the former 
 were much larger than those of the latter. Following 
 this same mode of attack, Courten (58) at the Uni- 
 versity of Yale employed a Rousselot exploratory bulb 
 which rested upon the tongue, and was connected with 
 a Marey tambour. Since the tongue is one of the
 
 TRAINING TO DECREASE VOCALIZATION 107 
 
 important organs called into play by articulation, a 
 record of the movement or non-movement of the tongue 
 would throw some light on the question : Does a sup- 
 pressed vestigial articulation always accompany silent 
 reading? Courten found that the curve of movement 
 varied both with the individual and with the degree of 
 concentration in thinking and in reading. In every 
 case, however, movement of some sort was clearly 
 recorded. 
 
 From the above mentioned objective findings two 
 inferences would seem to follow. 
 
 1. Too much weight should not be placed upon the 
 report of a subject whose introspection is unable to 
 detect a slight movement of the articulatory muscles. 
 
 2. The absence of articulatory movement in some 
 degree or other does not seem to be conclusively demon- 
 strated in any individual case. 
 
 While the articulatory movements that occur in 
 thinking and in silent reading are usually of a faint 
 vestigial character, there are occasions when the articu- 
 lation becomes very pronounced. Hansen and Leh- 
 mann (59) have shown that when the subject is thinking 
 very intently of some name or number an unconscious 
 whispering usually occurs. Though accompanied by 
 no perceptible movement of the lips, the sound can 
 nevertheless be distinctly heard by observers when the 
 subject is placed in especially favorable acoustic condi- 
 tions. These experimental findings of Hansen and 
 Lehmann offer corroborative experimental evidence to
 
 108 SILENT READING 
 
 the observation made by Egger, in 1881, that there 
 are certain mental states during which the inner speech 
 is especially vigorous despite one's best efforts to check 
 it occasions when it is impossible to "faire taire 
 notre pense'e." 
 
 After reviewing the literature on this subject rather 
 exhaustively, Pintner concludes: "The general result 
 from all these experiments can be summed up by saying 
 that silent reading is accompanied by articulation in 
 some degree or other. This activity of articulation is, 
 so far as we know, a universal habit. Whether it is a 
 necessary habit is another question." 
 
 ORIGIN OF HABIT OF INNER SPEECH 
 
 Before investigating the latter question it would 
 seem advisable to consider : What is the origin of this 
 habit ? The answer to this query may throw some light 
 upon the former. Is articulation in silent reading the 
 result of a habit of associating the sound of words with 
 their visual appearance, which habit has been acquired 
 through the process of education ? Or is it more largely 
 an inherited tendency ? To this question Quantz (26) 
 gives a very definite answer : 
 
 " Lip movement in silent reading is not an acquired habit, 
 but a reflex action, the physiological tendency to which is 
 inherited. It is not 'second nature' but essentially first 
 nature ; not something to be unlearned but to be outgrown. 
 It is a specific manifestation of the general psycho-physical 
 law of ' dynamogenesis ' by which every mental state tends to 
 express itself in muscular movement."
 
 TRAINING TO DECREASE VOCALIZATION 109 
 
 Among the considerations which Quantz adduces to 
 support the above statement are the following : 
 
 1. "The child in learning to read does not learn to move 
 his lips. The lip-movement is most decided at the very 
 beginning, and grows less so, as he becomes conscious of it 
 and controls it voluntarily. In learning to write, also, the 
 child moves his tongue and the muscles of his face ; some- 
 times even his feet or his whole body. But he afterwards 
 finds that this expenditure of energy is unnecessary. It is 
 brought under control when the writing itself becomes 
 partially reflex, leaving the higher conscious processes more 
 free to attend to the inhibition of these lower useless ones. 
 The same act is then performed more exactly, more quickly, 
 and with less exhaustion. Similarly, lip-movement is an 
 unnecessary expense of energy not only useless but detri- 
 mental. 
 
 2. "In the answers to the 'personal sheet' many persons 
 say that though not habitual lip-movers they do move their 
 lips when giving very close attention, or when reading matter 
 which is very difficult, absorbingly interesting, or highly 
 emotional. This means simply that we regularly inhibit 
 these vocal reflexes, but that when our whole attention is 
 given to the thought under consideration the watchfulness 
 over these motor tendencies is relaxed, and they find expres- 
 sion. We tend to 'think aloud' when preoccupied not 
 only lip-movement but actual speech is unintentional. It 
 is true also that these impulses to vocal movement are 
 stronger, and hence more likely to find an outlet, when the 
 corresponding mental processes are more vivid. This 
 follows from the general principle, experimentally demon- 
 strated by Fe>6, that 'the energy of a movement is propor- 
 tional to the intensity of the mental representation of that 
 movement.'
 
 110 SILENT READING 
 
 3. "The statement that lip-movement is 'natural/ and 
 reading without lip-movement an acquired habit, is entirely 
 in accord with mental laws. The development of mind is not 
 only in handing over processes once conscious to the control 
 of the reflex mechanism as in walking or playing a familiar 
 tune on the piano thus leaving consciousness free for the 
 acquisition of higher powers and the performance of tasks 
 more difficult; but the interchange is also in the opposite 
 direction originally reflex processes are frequently brought 
 under the control of the higher consciousness, and inhibited 
 if they are considered useless or detrimental." 
 
 The explanation of Quantz and the considerations ad- 
 vanced in its support seem to possess a certain amount 
 of plausibility. On the general psycho-physical theory 
 of dynamogenesis, one would naturally be led to expect 
 some form of motor reaction to the mental processes 
 involved in silent reading. The question might be 
 raised here why this reaction should always take the 
 specific form of movement of the musculature of articu- 
 lation. The nexus between the mental processes in- 
 volved in the assimilation of thought from the printed 
 page and the movement of the larynx, tongue, vocal 
 cords, lips, and the general physiological mechanism of 
 articulation does not seem to be per se an obviously 
 natural or necessary one. Why does not the motor 
 reaction take some other form of outlet the move- 
 ment of musculature other than the articulatory ? 
 The theory of dynamogenesis might explain the 
 presence of some form of motor reaction to the mental 
 processes involved in the interpretation of printed sym-
 
 TRAINING TO DECREASE VOCALIZATION 111 
 
 bols, but taken by itself, it does not seem sufficient to 
 explain why the reaction should always take the form 
 of articulatory movement. 
 
 In the writer's judgment, it would seem that to ex- 
 plain this latter connection between the interpretation 
 of printed symbols and the inner-vocalization of the 
 symbols, recourse must be had to the acquired habits 
 of speech and oral reading. The pronunciation of words 
 is learned by the child either through unconscious 
 imitation, or through express formal instruction. In 
 either event it must be acquired through practice. 
 Though the physiological capacity or tendency to speech 
 is inherited, the actual pronunciation of words must 
 be learned. Since speech is acquired before reading, 
 words have meaning to the child at first only as sounds. 
 The school ties on to these sounds certain visual sym- 
 bols in the form of printed letters or words. The latter 
 gradually become enriched with meaning through their 
 evoking the proper sounds which arouse the corre- 
 sponding meaning. The association between the visual 
 form of a word and its sound is stressed by the school 
 in the primary grades until the association becomes 
 very intimate and, apparently, inseparable. Reading 
 then consists of the stimulation of the visual imagery 
 which in turn arouses the auditory and articulatory 
 elements, whence only the meaning is reached. 
 
 The point made here is that the association of the 
 printed word with its corresponding sound and articu- 
 lation is consciously and purposely built up by the school.
 
 112 SILENT READING 
 
 The synaptical connections between the visual, the 
 auditory, and the articulatory centers in the nervous 
 system are thus made closer and closer. The mental 
 associations and the synaptical connections are not 
 inherited, however, but are acquired as the result of 
 practice and training be it conscious or unconscious 
 in character. 
 
 There is no intrinsic reason why the visual form of 
 words must necessarily be associated with their sound 
 in order to convey meaning. It is simply a matter of 
 economy of effort. Since ordinarily the average person 
 deals with words first as sounds, and continues later 
 to use them, to a great degree, in actual speech, the 
 school simply capitalizes the stock of meanings 
 already attached to the sounds by associating the 
 sound of words with their visual form. It would still 
 be possible, however, to link the meaning of words to 
 their printed form through the direct medium of the 
 visual imagery, or through the tactile channel, without 
 employing the intercessory assistance of the auditory 
 element. In teaching the deaf and dumb to read by 
 the manual method, this appears to be precisely what 
 is done. The auditory is naturally entirely lacking, 
 the visual imagery together with the kinsesthetic being 
 employed to convey directly the meaning of the visual 
 symbols. 
 
 Moreover Quantz's statement that "lip-movement 
 is * natural' and reading without lip-movement is an 
 acquired habit, is entirely in accord with mental laws"
 
 TRAINING TO DECREASE VOCALIZATION 113 
 
 needs to be examined rather carefully. It is undoubt- 
 edly true that the musculature of the lips as well as all 
 the other muscles of the body is gradually called into 
 play as the physical organism develops from infancy 
 to maturity. But the movement of the lips and other 
 articulatory organs that accompanies silent reading 
 is not simply movement of a generic character. It is 
 the definite specific form of movement that occurs when 
 the words that are being read are actually pronounced. 
 As Huey well observes, "while the inner speech is but 
 an abbreviated and reduced form of the speech of every- 
 day life, a shadowy copy as it were, it nevertheless 
 retains the essential characteristics of the original." 
 Now, while it is to be admitted that the movement of 
 the lips and of the motor organs is natural as the or- 
 ganism develops, and while some form of motor reac- 
 tion to the various mental processes is to be expected 
 on the psycho-physical theory of dynamogenesis, the 
 peculiar specific form of the lip and other articulatory 
 movement such as occurs in actual speech is 
 still left unexplained. 
 
 The explanation is to be found not in the list of 
 inherited tendencies or connate reflexes, but in the 
 acquired habits of the individual. The mastery of the 
 phonetic properties of words, and their correct pronun- 
 ciation requiring, as they do, particular types of lip 
 maneuvers, the careful manipulation of the tongue, 
 and other delicate and "unnatural" articulatory move- 
 ments have not alas been inherited, but are the re-
 
 114 SILENT READING 
 
 suit of conscious training and much practice, as every 
 primary teacher realizes only too painfully. The kind 
 of articulatory movement that accompanies silent 
 reading cannot therefore be said to be natural in the 
 sense that it has not been learned or acquired. It has 
 been acquired. But the constant association between 
 the visual symbol of the word and its phonetic proper- 
 ties has become so deeply ingrained in the synaptical 
 fibers of the neurones, through both the conscious 
 linkage of these two by the school, and through the 
 constant usage of a word in both its visual and auditory 
 form in daily life, that the sight of a word comes to 
 arouse its appropriate sound and its corresponding 
 articulatory movements. Thus inner speech becomes 
 an automatic reflex accompaniment of silent reading. 
 But the point never to be forgotten is that it has been 
 acquired not inherited " ready made." 
 
 Is INNER SPEECH NECESSARY IN SILENT READING? 
 
 The answer to this question is really implicit in the 
 conclusion reached in the discussion of its origin ; 
 namely, that it is an acquired habit, not an inborn 
 characteristic. Consequently it would seem theoreti- 
 cally possible to learn at the start to read through the 
 sole medium of the visual imagery, or if the habit of 
 inner speech has been already acquired, to inhibit 
 the habit. By inner speech is here meant that definite 
 specific type of suppressed articulatory movement 
 which is similar in contour and general outline to the
 
 TRAINING TO DECREASE VOCALIZATION 115 
 
 larger movements of the gross musculature of articu- 
 lation which functions in oral speech. The mere 
 unconscious quiverings of an organ which have been 
 detected by delicate instruments during silent reading, 
 when the tongue, lips, etc., are held, for all practical 
 purposes, rigidly motionless, would not therefore con- 
 stitute inner speech as the term is used in this para- 
 graph. 
 
 As a result of his experiment at the University of 
 Chicago, in inhibiting articulatory movements in silent 
 reading, Pintner (48) draws the following conclusions : 
 
 1. " That articulation during the reading process is a habit 
 which is not necessary for that process. 
 
 2. "That practice in reading without articulation can 
 make such reading as good as the ordinary reading of the 
 same subject. 
 
 3. "That practice in reading without articulation tends 
 to aid ordinary reading, most probably by shortening the 
 habitual practice of articulation." 
 
 Similar to the above is the conclusion reached by 
 Secor (60) after his experiment on this subject at 
 Cornell, namely that "articulation and audition" are 
 not to be regarded as "absolutely necessary elements." 
 
 The conclusion reached by Pintner and Secor is fur- 
 ther confirmed by the observation of W. A. Schmidt (7) : 
 
 " It is quite possible, too, that if training in oral reading 
 were discontinued at an early stage, and training in rapid 
 silent reading were stressed, the tendencies toward inner 
 speech might be greatly reduced and visualization cultivated,
 
 116 SILENT READING 
 
 at least in part. As a matter of fact, we appear to have the 
 ability to take in all kinds of situations visually without 
 speech accompaniments. This is true even in the case of the 
 interpretation of many printed symbols." 
 
 MEANS OF INHIBITING INNER SPEECH 
 
 From the conclusion that articulately movements 
 are not necessary in silent reading to the devising of 
 means of inhibiting them is rather a far cry. Their 
 inhibition is another matter. And the validity of the 
 former conclusion does not depend upon the successful 
 execution of the latter. Various devices have been 
 employed to inhibit articulation. Secor reports that 
 whistling and the uttering of the alphabet aloud "com- 
 pletely removed all traces of articulatory movement, 
 so far as the possibility of discovering this by introspec- 
 tion is concerned." Though the saying aloud of a letter 
 or word would apparently engage the musculature of 
 articulation during the moment of actual utterance, 
 there would still remain the possibility of inner move- 
 ment during the intervals between the words which 
 would be likely to escape introspection. To guard 
 against such interstitial movements, the subjects were 
 requested to read while prolonging a letter as much as 
 possible. This served but to verify the results of the 
 previous introspection that no movement occurred. 
 In the writer's judgment, however, even this last 
 method does not necessarily remove the possibility 
 of a faint, subtle, vestigial movement of some of the
 
 TRAINING TO DECREASE VOCALIZATION 117 
 
 smaller articulatory muscles which are obviously not 
 employed in the simple prolongation of a letter. 
 
 Pintner had his subjects count aloud the series 13, 
 14, 15, 16 ; 13, 14, 15, 16 ; etc., while reading silently. 
 The pronunciation of these numbers requires a more 
 elaborate musculature of articulation than the simple 
 letters of the alphabet used in Secor's experiment. The 
 introspection of Pintner's two subjects reported a com- 
 plete absence of articulatory movement, though audi- 
 tion still remained in the case of one of the subjects. 
 Incidentally the introspection of the latter observer 
 casts an interesting ray of light upon the relationship 
 between the motor and the auditory elements hi inner 
 speech. If the introspection reflects his actual mental 
 experience, it shows that the union between the motor 
 or articulatory element and the auditory is not nearly 
 so inseparable as many writers have supposed. It 
 lends no corroboration to the generalization made by 
 Huey (6) : "The fact is that what we say is always 
 heard as well, and there comes to be an indissoluble 
 union of the auditory and motor elements." 
 
 As a result of this phase of his experiment, Pintner 
 concludes: "I think we are justified in saying that 
 reading without articulation can take place, . . . and 
 that practice in reading without articulation increases 
 the ordinary rate of reading, no doubt due to the fact 
 that after such practice the amount of articulation 
 made use of is not so great as formerly." 
 
 No satisfactory method has yet been devised of ex-
 
 118 SILENT READING 
 
 eluding audition in silent reading. Secor found that 
 the playing of a xylophone near the observers quite 
 as often as not left the inner hearing undisturbed. 
 Similarly, loud noises seemed ineffective in suppressing 
 the inner auditory accompaniment in the silent reading 
 of most people. 
 
 In concluding this brief discussion of the history, 
 origin, and nature of inner speech, it is only fair 
 to say that when due allowance has been made 
 for the findings of all the investigations on this subject, 
 there still remains too much speculation and too few 
 scientifically established facts; too much conjecture 
 and too little certainty. The problem of inner speech 
 why we persist in vocalizing in silent reading is 
 still with us, shrouded in some of the pristine mystery 
 which it presented to its first systematic investigators 
 - Egger and Ballet. 
 
 It is hoped that this review of the investigations of 
 inner speech has served to show at least the significance 
 and importance of this factor in silent reading. It 
 will give the reader a better and more intelligent in- 
 sight into Type II - - Training to Decrease Vocaliza- 
 tion. The validity of this method does not depend, 
 however, upon any of the tentative conclusions as to its 
 origin, the relationship between the motor and auditory 
 elements previously presented. Neither does it hinge 
 upon the correctness of the answer to the theoretical 
 question : Can every vestige of articulatory movement 
 be inhibited ? Prescinding from such theoretical con-
 
 TRAINING TO DECREASE VOCALIZATION 119 
 
 siderations, this method endeavors to decrease lip-move- 
 ment and that form of articulation which notably hin- 
 ders speed in silent reading. Whether or not a slight 
 vestigial quivering of the larynx, barely detectable by 
 a tambour, exercises any effect upon the rate is a 
 theoretical question which does not concern us here. 
 Certainly its effect, at most, is but negligible. But 
 the more or less elaborate movement of the gross mus- 
 culature of articulation, the tongue, lips, and throat 
 muscles, has a very pronounced effect upon the rate, 
 as every investigation of this subject has clearly shown. 
 
 A summary of the findings on this point showing the 
 retarding influence of a marked habit of vocalization 
 was presented in Chapter III. It was this evidence 
 that showed the advisability of decreasing vocalization 
 to increase speed, and was thus directly responsi- 
 ble for the formulation of this type of training. The 
 evidence presented there is well epitomized in the con- 
 clusion of Huey : " The direct linking of visual form to 
 ideas, cutting out of the circuit the somewhat cumbrous 
 and doubtless fatiguing audito-motorizing mechanism, 
 would seem to be a consummation to be wished for." 
 Similar in effect is the conclusion of Dearborn : "The 
 effect of articulating is to decrease ordinarily the span 
 of attention" and consequently to retard the rate of 
 reading. 
 
 The following is a statement of the second method, 
 a copy of which was sent to every teacher using this 
 method.
 
 120 SILENT READING 
 
 STATEMENT OF 
 
 TYPE II. TRAINING, TO DECREASE VOCALIZA- 
 TION IN SILENT READING 
 
 Vocalization may show itself in rather extreme form by 
 elaborate lip-movement, or in its usual form, by incipient 
 movements of the lips, tongue, pharynx, vocal cords, and the 
 general mechanism of the throat. The reader feels or hears 
 himself pronouncing the words. This constitutes the so- 
 called "inner speech" of silent reading. Inner speech is 
 present in some form in the reading of most pupils. 
 
 Inner speech has the effect of slowing up the rate of read- 
 ing, causing the individual to read no faster than he can 
 actually pronounce the words to himself. Hence, perception 
 must wait upon pronunciation. The rate of reading, in 
 other words, is made dependent upon the rate of inner 
 speech. If this process of vocalization be gradually lessened 
 and finally eliminated, the rate of silent reading may be 
 greatly accelerated. The period of training to decrease 
 vocalization should last from April 8 until May 29. It should 
 replace the regular work in reading and no other classroom 
 time should be devoted to reading. Thirty minutes per 
 day should be allowed for the work. The method should 
 consist essentially of alternate reading and reproduction. 
 The reading should be timed and the place in the selection 
 reached at the end of each of the various periods should be 
 regularly marked. Since the time element enters into all the 
 work, a clock should be placed in the front of the classroom 
 so that it can be seen by all the pupils. Pupils should be 
 kept informed of their speed of reading. 
 
 Reproduction should consist both of free paraphrase 
 orally or in writing and of answers to specific questions 
 based on the text. The length of the reading period and of 
 the reproduction period should vary with the grade of the
 
 TRAINING TO DECREASE VOCALIZATION 121 
 
 pupils and with the subject matter. In general, however, 
 the reproduction should not occupy more than one-quarter 
 of the total time allowed for the exercise. 
 
 Only interesting material should be selected. It should 
 also be easily within the understanding of the pupils. Since 
 the object is to set up habits of rapid reading, emphasis upon 
 the simplicity of the selection is necessary. 
 
 In order to assist the pupils in covering as much reading 
 matter as possible, a definite preparation may be made, 
 consisting : (a) of thought preparation, or (6) of word prepa- 
 ration, or (c) of both. The thought preparation should 
 consist of such an introduction by commentary or by the 
 question-answer process as will arouse interest and enlist 
 the attention of the pupils. This preparation should in no 
 case be long. For the word preparation the teacher should 
 select such words as in her judgment would be familiar 
 to the pupils, and should present them briefly, explaining 
 their meaning. The preparation both thought and word 
 may be abridged or even omitted when the material is 
 such as to give no difficulty. In no case should the prepara- 
 tion occupy more than five minutes of the 30 assigned to 
 the exercise. 
 
 Each pupil should keep a chart of his daily performance, 
 and a complete chart of the daily class performance should 
 be conspicuously displayed in the classroom. 
 
 It is desirable that the directions to pupils given by each 
 teacher concerned in this investigation be substantially the 
 same. The following suggestions are offered, indicating the 
 nature of the directions to be given to the pupils. Literal 
 adherence to them is not required. Their spirit, however, 
 should be maintained. 
 
 Point out the advantage of a rapid rate of reading. Tell 
 them that their effort to read rapidly will be more successful 
 if they avoid moving their lips and tongue and do not at-
 
 122 SILENT READING 
 
 tempt to pronounce each word to themselves. For it is 
 precisely this attempt to pronounce each word which slows 
 up the rate of reading. Articulation of words in silent read- 
 ing, therefore, is a serious hindrance to rapid silent reading, 
 which should be eliminated in the interests of efficiency. 
 Try to get the children to see that their speed and efficiency 
 in silent reading will depend largely upon their elimination 
 of inner speech. 
 
 Say to them in substance: "Read this selection as fast 
 as you can. While reading do not move your lips or tongue. 
 Do not pronounce the words to yourself, as that will cause 
 you to read more slowly than you otherwise would. I want 
 
 to see how much you can read in 1 minutes. But do 
 
 not skip anything, as I am going to ask you to tell me about 
 the story you have read. Try to read faster to-day than you 
 did yesterday." 
 
 See that the pupils have a pencil at hand and direct them 
 to stop reading at once as soon as you say "stop." Direct 
 them then to mark the end of the line which they are reading 
 when told to stop. Pupils may now reproduce what they 
 have read as indicated above. In a similar manner, reading 
 and reproduction are to be continued until the end of the 
 30 minutes assigned. Have the pupils then report the num- 
 ber of pages and lines beyond the last full page which they 
 read. 
 
 As part of your preparation for the exercise you will be 
 expected to know the average number of words per line in the 
 matter which is being read and the number of lines per page 
 (if pages are broken by illustrations or for other reasons, 
 special account of these pages will have to be taken). From 
 the pupils' reports as to pages and lines read each day, esti- 
 mate the number of words read that day and divide by the 
 
 1 Number of minutes to be varied by the teacher, as two minutes for 
 one reading stretch, three minutes for another, etc.
 
 TRAINING TO DECREASE VOCALIZATION 123 
 
 total number of minutes used in reading. This will give the 
 number of words read per minute by each pupil. Each stu- 
 dent should figure out his score, and should immediately 
 enter it upon the chart, which will thus serve as a record of 
 the pupil's daily progress in silent reading. The teacher 
 is also requested to keep a diary in which she will daily 
 record notes and observations concerning the progress of the 
 experiment. 
 
 On May 1 and May 28, besides the record of the aver- 
 age number of words read per minute by each pupil, during 
 the total reading time, a record of the average number of 
 words read per minute by each pupil during each of the vari- 
 ous reading periods constituting the total time devoted to 
 reading on each of the above mentioned days should also 
 be kept. 
 
 On May 2 and May 29, the pupils should be tested with 
 the Courtis Silent Reading Tests. As soon as possible after 
 the last Friday in May a copy (or the originals) of the indi- 
 vidual charts, the class chart, and a record of results and 
 observations should be mailed to the Bureau of Educational 
 Research, University of Illinois, Urbana, Dlinois. 
 
 It will be noticed that this method employs all the 
 auxiliary devices, the time control, and the technique 
 of Type I. The difference lies in the inclusion in the 
 second method of the additional factor or principle - 
 decreasing the vocalization. The two principles, prac- 
 tice in rapid silent reading and decrease of vocalization, 
 are both fundamental in this latter Type. The stress 
 is now placed upon the lessening of the inner speech, 
 though emphasis upon rapid reading still remains. 
 In order to determine the comparative efficacy of these
 
 124 SILENT READING 
 
 two factors in accelerating the rate, it was originally 
 planned to exclude the latter principle and to construct 
 this method directly upon the decreasing of the vocali- 
 zation as the sole basic principle. Preliminary investi- 
 gation showed, however, that the effort to lessen inner 
 articulation was far more successful when the individual 
 not only consciously endeavored to inhibit such move- 
 ments but also read rapidly at the same time. The 
 comfortable, leisurely rate of reading to which the 
 average individual is accustomed seems particularly 
 favorable to the arousal of the usual articulatory 
 movements. 
 
 It was found that when to the conscious attempt at 
 inhibition was joined rapid reading, the decrease of 
 inner articulation was effected more promptly and more 
 thoroughly. Apparently, these two factors are so 
 closely interrelated that a mutual causal relationship 
 exists between them. The exclusion of rapid reading 
 from the training to decrease vocalization would have 
 deprived the latter of one of its most effective aids. 
 Accordingly, the basic position in the second method is 
 shared coordinately by these two principles, which 
 work hand-in-hand. The coordinate stress thus placed 
 upon rapid reading while the subject is trying to check 
 the articulatory movement is one of the important 
 particulars which differentiates this type of training 
 from that received by the two subjects in Gray's experi- 
 ment which was discussed in Chapter III. 
 
 It will still be possible, moreover, to determine to
 
 TRAINING TO DECREASE VOCALIZATION 125 
 
 some extent the amount of influence exerted upon the 
 acceleration of the rate by the decrease of vocalization. 
 Since this latter factor is the only one possessed by Type 
 II which is not also included in Type I, the difference in 
 the amount of gain effected by these two Types can be 
 largely attributed to the only factor not common to 
 both the conscious decrease of vocalization in silent 
 reading.
 
 CHAPTER VI 
 TYPE III. TRAINING IN PERCEPTION 
 
 THE core of the reading complex is the process of 
 perception. The movement of the eyes from pause to 
 pause, their convergence and divergence in fixation, 
 the numerous and varied play of the neural-muscular 
 mechanism involved in the ocular adjustment to the 
 printed symbols, are all subsidiary to the main process 
 of perception. The latter constitutes the actual seeing 
 or visualization of the printed words. The results of 
 numerous investigations, as stated in the earlier chap- 
 ters, have shown that the overwhelming bulk of the 
 reading time, in fact from twelve-thirteenths to twenty- 
 three twenty-fourths, is consumed by the fixation- 
 pauses. The total reading time, therefore, may be 
 said to be practically the number of the pauses 
 multiplied by their average duration. But the number 
 of pauses per line is dependent upon the size of the per- 
 ceptual span the wider the span the fewer the pauses, 
 and vice versa. Hence a ty*pe of training that is to effect 
 any appreciable economy in the reading time must, in 
 some way or other, influence the perceptual process that 
 occurs in the fixation-pause. The improvement may be 
 
 126
 
 TRAINING IN PERCEPTION 127 
 
 secured either by widening the visual span, causing 
 fewer fixations, or by accelerating the perceptual process, 
 thus lessening the duration of the fixation-pauses, or 
 by a combination of both these results. 
 
 PERIPHERAL VISION 
 
 Upon the fixation-pause., during which all vision or 
 perception occurs, has been focused the attention of 
 many investigators. The perception that occurs 
 therein is of two kinds, foveal and peripheral. The area 
 that can be grasped by foveal vision in any one fixa- 
 tion is relatively small. Only about five letters (61) 
 will be "unequivocally clear" when the eyes are held 
 quite stationary. The adjoining letters will not be so 
 sharply defined, but will shade off gradually into a hazy 
 outline, growing fainter and fainter as they recede 
 from the foveal area. These letters are grasped only 
 by peripheral vision. The area of peripheral vision 
 is consequently much larger than that of clear vision. 
 The number and duration of pauses per line and the 
 speed of reading hi general are thus seen to be condi- 
 tioned to a considerable degree by the effective utiliza- 
 tion of extra-foveal vision. The importance of this 
 phase of the reading complex seems to have been 
 rather generally overlooked. Practically no conscious 
 attempt is made in the present teaching of reading in 
 the schools to develop peripheral vision or to enlarge 
 its areas. Reading which is wholly or chiefly dependent 
 upon foveal vision will be slow and halting, marred
 
 128 SILENT READING 
 
 by too frequent pauses, which inhibit the development 
 of a regular rhythmical swing of the eyes in traversing 
 the printed line. 
 
 Dodge (61), who has analyzed the functioning of the 
 various elements in the reading complex with marked 
 acuteness, thus describes the work of extra-foveal 
 vision : 
 
 "Sometimes the peripheral vision of words, when they 
 are indistinctly seen in the hazy part of a line, is sufficient 
 for reading. It is always useful, furnishing an important 
 premonition of coming words and phrases, as well as a con- 
 sciousness of the relation of the immediately fixated symbols 
 to the larger groups of phrase and sentence. Without this 
 premonition of coming words and the outlines of larger 
 groups, the process of reading would be slow and difficult. 
 
 "In normal reading there is abundant evidence that the 
 word forms, indistinctly seen in peripheral vision, begin the 
 reading process well in advance of direct fixation and the 
 consequent clearing up of the letters. 
 
 "It is this premonition of coming words and phrases, 
 sentence breaks, and paragraphs, that determines the posi- 
 tion of future fixations, and reduces the duration of individual 
 fixations sometimes to pauses one-quarter the normal percep- 
 tion time. They are often less than the simple reaction time 
 of the eye." 
 
 Not only is peripheral vision effective in decreasing 
 the number and duration of the fixation-pauses per line 
 and consequently accelerating the rate, but it is also 
 instrumental in fostering and stimulating meaning 
 premonitions, thereby improving comprehension as 
 well. Hand in hand with the reaching out of percep-
 
 TRAINING IN PERCEPTION 129 
 
 tion beyond the immediate field of clear vision seems 
 to go the reaching ahead of the assimilative factor 
 beyond the meaning grasped in direct fixation to those 
 dimly felt in the extra-foveal area. Meaning premoni- 
 tions, made more numerous and more vigorous by the 
 functioning of a wide peripheral vision, play an important 
 role in the rapid interpretation of the printed page. 
 Just as extra-foveal vision frequently suffices for the 
 reading of a word without the agency of direct fixation, 
 so meaning premonitions not infrequently carry the 
 reader over many a word without the immediate ap- 
 prehension of the word in direct fixation. This is 
 evidenced by the fact that a person in reading a selec- 
 tion aloud in which the meaning premonitions stream 
 thick and fast into his consciousness will unwittingly 
 substitute a synonym for the word actually occurring 
 in the passage. 
 
 Concerning the importance of these meaning pre- 
 monitions, Dodge (61) says : 
 
 "Adequate premonitions are as conspicuously lacking 
 in the stumbling reading of childhood as in our reading of a 
 foreign language. Their development is a most important 
 part of the training in rapid reading. The very rapid 
 reader who makes two or three steps in a line can see no 
 more distinctly than the plodder with ten. His main ad- 
 vantage is in the way he grasps what he sees only indistinctly 
 in the extra-fixational vision." (Italics inserted.) 
 
 The Type of Training which will be outlined in this 
 chapter has aimed to develop the more generous utiliza-
 
 130 SILENT READING 
 
 tion of peripheral vision, and of its mental correlate 
 meaning premonitions. The direction to the pupils 
 in the exposure exercises, requesting them to read the 
 sentence "as a whole," allowing time sufficient for 
 but a single fixation, was calculated to call the peripheral 
 vision into play, while the suggestion to look up oc- 
 casionally and tell " how much they saw ahead " and how 
 much of the remaining meaning of the sentence they 
 could give, was designed to develop meaning premoni- 
 tions along with the extra-foveal vision. 
 
 FOVEAL VISION 
 
 However important a factor peripheral vision may 
 be in developing rapid, effective reading, direct or 
 f oveal vision still constitutes the core of the perceptual 
 process perception per se. In the field of clear vision 
 it is not a matter of conjectural interpretation or in- 
 ferences from previously perceived premises, but a 
 matter of direct fixation and immediate apprehension. 
 The functioning of direct fixation hi connection with 
 extra-foveal vision, Dodge (61) thus describes : 
 
 "In adult reading the moment of actual fixation seems 
 to be an incident somewhere in the middle of the reading 
 process. Coming between the premonition and the after 
 echo, its effect is to correct, to confirm, and to intensify the 
 premonition. Psychologically its function is selective and 
 definitive. It emphasizes the excitation of suitable residua 
 and inhibits the misfits. The pedagogical importance of 
 word forms is clear. It is possible that special training in 
 peripheral vision would be worth while, but it is equally
 
 TRAINING IN PERCEPTION 
 
 131 
 
 evident that no training is adequate which does not provide 
 for the corrective cooperation of direct fixation." 
 
 The Type of Training to be outlined in this chapter 
 aims definitely at the enlarging of the perceptual span 
 functioning in reading. The work in exposure exer- 
 cises attacks this problem directly. It endeavors to 
 widen the span of attention so as to grasp three, four, 
 five, six, and even more words if possible, in a single 
 fixation. C. T. Gray (19) showed that a marked 
 increase in the visual span resulted in the case of two 
 fourth grade subjects after training in perception by 
 means of tachistoscopic exposures. In the case of the 
 two sixth grade subjects, however, similar short ex- 
 posure exercises did not seem to effect any appreciable 
 widening of the perceptual span as determined by the 
 number of words grasped in a tachistoscopic exposure. 
 However, an examination of the data afforded by the 
 photographic record of the eye-movements in the read- 
 ing of one of the two subjects, B. R., taken before and 
 after the training in perception, yields a rather striking 
 result. The following is a statement of the record as 
 reported by Gray (19). 
 
 TABLE VII. EYE-MOVEMENT RECORD OP B. R. BEFORE AND AFTER 
 
 PRACTICE 
 
 
 AVERAGE 
 
 AVERAGE 
 
 AVERAGE No. 
 
 
 No. OF 
 
 LENGTH OF 
 
 REGRESSIVE 
 
 
 PAUSES 
 
 PAUSES 
 
 MOVEMENTS 
 
 Before practice . . . 
 
 15.5 
 
 15.4 
 
 4.5 
 
 After practice . . . 
 
 6.1 
 
 12.6 
 
 1.2 
 
 After 3 months' interval 
 
 8.4 
 
 
 1.2
 
 132 SILENT READING 
 
 An examination of these records clearly demonstrates 
 a striking enlargement of the perceptual span which 
 functions in actual reading. This is evidenced by the 
 reduction in the average number of pauses per line 
 from 15.5 in the first reading to 6.1 in the reading after 
 the practice in perception. This indicates that the 
 size of the perceptual span was more than doubled, since 
 its width bears an inverse proportion to the number of 
 fixation-pauses per line. Even the average duration 
 of the perceptual process in the fixation-pauses was 
 decreased from 15.4 to 12.6, showing improvement in 
 this phase of the process also. A notable decrease in 
 the average number of regressive movements per line 
 is likewise noticeable. B. R/s rate of reading as de- 
 termined by careful tests had increased from 1.0 word 
 per second before the practice to 5.0 words per second 
 after the training an improvement of 400 per cent. 
 
 Consequently while B. R.'s span of attention showed 
 no notable increase in grasping numbers of words ex- 
 posed by a tachistoscope, it showed a striking enlarge- 
 ment in reading an actual context. In alluding to this 
 point in Chapter III, the writer termed the latter the 
 relative span of attention, the span which functions in 
 reading a context material, in contradistinction to the 
 absolute span which functions in tachistoscopic exposure 
 work. Exercises of this latter character have shown 
 that the perceptual span which they call into play, 
 grasping, as it does, five and six words in a single fixa- 
 tion, is seldom fully utilized in ordinary reading where
 
 TRAINING IN PERCEPTION 133 
 
 the span rarely exceeds two or three words. Conse- 
 quently, while the absolute span of perception may 
 show no increase as a result of training in perception, 
 the relative span which functions in ordinary reading 
 may reveal a striking increase, as in the present instance. 
 In other words, the perceptual span is utilized more 
 effectively in reading. The point made here is that this 
 is true not only of fourth-grade subjects, but also of 
 older subjects, as B. R.'s case well illustrates. The 
 eye-movement records of B. R. certainly lend no sup- 
 port to the tentative conclusion of Gray that training 
 in perception, if it is to produce results, should come 
 before the fourth grade. For the perceptual span as 
 it functions, not in the grasping of isolated words or 
 phrases, but as it is used in the reading of connected 
 material, is the important factor. That this latter is 
 susceptible to improvement has been clearly shown by 
 the record of Gray's subject, B. R. 
 
 RELATION OF SPEED OF READING TO TYPES OF IMAGERY 
 
 What is the effect of the different types of mental 
 imagery upon the rate of reading? In other words, 
 Is the reading rate largely conditioned by the mode 
 of imaging ? If so, what types are conducive to rapid- 
 ity and what ones tend to retard? These are some 
 of the interesting questions which have been touched 
 upon in recent studies in the psychology of reading. 
 
 To the above queries W. A. Schmidt (7) gives no un- 
 certain answer :
 
 134 SILENT READING 
 
 "Individuals of this (visual) type" says he, "are by nature 
 rapid readers, other conditions being equal. Their speed is 
 not due to scanning, however, for they have no occasion to 
 resort to this, since there is almost no limit to the rate of 
 visualization. The motor type, on the other hand, tends to 
 represent the slowest readers, the dependence upon the 
 physiological mechanism being in this case quite marked. 
 The auditory type ranks between the two, the hearing of 
 the words being in this case often quite vestigial. This 
 type appears to be much more common than either of the 
 other two types. Most of the evidence which the writer has 
 been able to gather through interviews with a large number 
 of individuals seems to support the conclusion that the 
 auditory-motor type of reader can compete with the visual 
 only when he is able to resort to scanning. There can be 
 little doubt that rapid readers fall almost exclusively into 
 these two classes those on the one hand who are good 
 visualizers and those on the other hand who have acquired 
 the ability to gather meaning from the printed page without 
 definitely reading all the words and sentences." 
 
 The above conclusion of Schmidt's appears to the 
 writer to possess a certain amount of plausibility. It 
 seems at least a priori logical to assume that reading 
 which involves merely, or at least chiefly, visualization, 
 requires less functioning of the physiological mechanism 
 than the auditory or motor types with their more or less 
 elaborate movement of the musculature of articulation. 
 Involving a less elaborate functioning, visualization 
 would seem to be a more direct and more rapid mode 
 of reading. Schmidt appears to go rather far, however, 
 in maintaining that " good readers fall almost exclusively
 
 TRAINING IN PERCEPTION 135 
 
 t 
 
 into two classes" - good visualizers and those able to 
 scan ; that the auditory or motor readers can compete 
 with visualizers only when they resort to scanning. 
 
 In the first place, the data which throw any light 
 upon the influence of the various types of imagery upon 
 the reading rate are exceedingly meager. The scanty 
 evidence that is available, however, does not seem 
 to warrant so positive a statement as to the condition- 
 ing, determining influence of the mode of imaging 
 upon reading rate. There are numerous instances 
 of very rapid readers who are decidedly "auditaires" 
 and "motaires" but who can compete very suc- 
 cessfully with the visuals Moreover, it is to 
 be remembered that the audition may be of the 
 faint, vestigial psychical character involving, accord- 
 ing to the introspection of some psychologists, no 
 movement whatsoever of the musculature of articula- 
 tion. In the writer's judgment, it is in the latter fac- 
 tor the elaborate functioning of the physiological 
 mechanism of vocalization that the retarding in- 
 fluence is largely to be located. If the latter be elim- 
 inated, a high rate of speed may be reached regardless 
 of the type of imagery, as the results of the present in- 
 vestigation indicate. 
 
 " Most of the evidence which the writer has been able 
 to gather through interviews" is mentioned by Schmidt 
 as the data l from which he generalizes concerning the 
 
 1 After the above had been written, the writer received a communi- 
 cation from Dr. Schmidt, in which the latter mentions Quantz's experi- 
 ment as corroborative of his generalization. As the writer had already
 
 136 SILENT READING 
 
 I 
 
 conditioning influence of the various types of imagery 
 upon the reading rate. A grave difficulty which con- 
 fronted the writer in his endeavor to secure data on 
 this point by questioning readers as to their mode of 
 imaging and their reading rate was their inability in 
 the overwhelming majority of cases to determine with 
 anything approaching certainty what is their dominant 
 type of imagery. Nor is that strange. For, practi- 
 cally all the recent investigations (62-67) of imagery 
 types have shown that, as a general rule, individuals 
 employ not a single, constant type, but many types 
 of imagery. Netschajeff (68) has conclusively demon- 
 strated the existence of such combinations of imagery 
 types, for example, as the visual-motor, and the audi- 
 tory-visual-motor. Indeed, even in a single type of 
 mental work, individuals are found using now one, now 
 another mode of imaging. There is a confusing com- 
 plexity of modes of imaging in a single individual. 
 
 In view of this fact it is not surprising that an in- 
 dividual when interviewed can scarcely state offhand 
 what is his precise type of imagery. Even when sub- 
 analyzed the results of Quantz's experiment in this chapter, it is felt 
 that all of the essential data upon which Dr. Schmidt's conclusion rests 
 have been considered. Dr. Schmidt also states that the adults whom 
 he interviewed "were, for the most part, individuals who had studied 
 psychology rather extensively. In consequence they were, as a rule, 
 able to attempt (whether they succeeded is quite another question) to 
 classify themselves as auditory, motor, or visual in type." In the 
 writer's judgment, the attempt to classify oneself in regard to imagery 
 type, without using any psychological test, is rather precarious. Dr. 
 Schmidt has the gratitude of the writer for his additional information.
 
 TRAINING IN PERCEPTION 137 
 
 jected to psychological tests in the laboratory, the 
 dominant type of imagery is not always established with 
 certainty. Indeed, there is no single, psychological 
 test that can be relied upon to discover with certainty 
 the dominant mode of imaging in the case of a "mixed" 
 type. It is in the latter case, moreover, that most 
 individuals are found to belong. 
 
 Meumann (69) epitomizes the results of his own ex- 
 tensive investigations and those of other investigators 
 when he states : " Investigation has shown, however, 
 that the great majority of people belong not to pure, 
 but to mixed or balanced ideational types. In these 
 cases, we have a compromise between ideational type 
 and sense-modality." Illustrating the complexity of 
 modes of imaging in a single individual, he says : 
 "For instance, the concrete visual type of ideation 
 may coexist in any individual with verbal ideation 
 of another source, with vocal-motor or with verbal 
 auditory imagery." 
 
 Consequently the question might well be raised as to 
 the accuracy and the general value to be attached to 
 information derived from interviews upon a question 
 that is notoriously difficult to answer, even when 
 fortified with the results of a psychological test, to 
 say nothing about the case when it is answered off- 
 hand without such a basis for a reply. 
 
 The only express attempt, to the writer's knowledge, 
 to attack the problem of the influence of imagery types 
 upon reading rate was the experiment of Quantz (26)
 
 138 SILENT READING 
 
 at Wisconsin. Quantz sought to determine the relative 
 influence of the visual and the aural bias upon the 
 reading rate in three ways : 
 
 " (1) by testing the visual and auditory span ; that is, the 
 limit of power to repeat correctly words read or heard once ; 
 (2) by detection of differences between two variant readings 
 of the same passage ; (3) by the ability to reproduce the 
 thoughts of two selections, one of which was read to the sub- 
 ject, the other read silently by him, at the same time." 
 
 Whether any of these three tests actually determined 
 the dominant type, or even the relative strength of the 
 visual and auditory imagery in the individual, is very 
 doubtful. Underlying the three methods employed by 
 Quantz is the assumption that the capacity to grasp 
 and remember or reproduce words presented orally 
 or visually constitutes an index of the relative strength 
 of the auditory or visual type of imagery. That words 
 presented orally and reproduced in the same manner 
 do not necessarily tap solely the auditory type of 
 imagery is evident from the fact that there is always 
 the possibility of transforming the auditory impression 
 into another more favored type of imagery. The same 
 is true of the possibility of transforming the visual 
 impression into an auditory image. The existence of 
 such a substitute mode of imaging, a "surrogate," or 
 sort of vicarious functioning of imagery types, which 
 has been experimentally demonstrated by Meumann, 
 serves to render the fundamental assumption under- 
 lying Quantz's three tests very doubtful, and the con-
 
 TRAINING IN PERCEPTION 139 
 
 elusions based upon such tests, of similar questionable 
 validity. 
 
 Indeed, Meumann (69) cites as an illustration of a 
 process wherein substitutive imagery may be employed, 
 practically the same task as Quantz 's test of the span of 
 prehension "the limit of power to repeat correctly 
 words read or heard once." 
 
 "Let us take," says Meumann, "the case of an individual 
 who has read a list of words or to whom such a list has been 
 dictated, with a view to having them reproduced immediately 
 in vocal form. The mixed type is always able to adapt itself 
 to this task. ... In recalling auditory words immediately 
 after hearing them, the visualizer may employ auditory 
 images chiefly, but may make a secondary use of the visual 
 imagery into which he has transformed the auditory material ; 
 but in recalling a conversation with a friend after a long inter- 
 val, ... he has recourse chiefly to visual images of expres- 
 sions employed by his friend during the conversation. The 
 individual who belongs to a pure type, on the other hand, 
 endeavors in immediate retention to transform into his own 
 favored imagery the sensory impressions which he receives." 
 
 The results obtained from the three tests applied 
 by Quantz are by no means uniform. The evidence 
 from the first test (span of prehension) Quantz con- 
 strues to indicate "a gradual increase in the rate of 
 reading as the subject moves away from the auditory 
 type and toward the visual." The results of the second 
 test (detection of differences) are just the reverse, show- 
 ing that the auditory type is decidedly more conducive 
 to rapidity in reading than the visual type ; while the
 
 140 SILENT READING 
 
 third test (comparison by simultaneous processes) 
 seems to show a correlation between eye-mindedness 
 and rapid reading. The conflicting character of the 
 results obtained tends strongly to corroborate the 
 writer's conclusion that not one of Quantz's three tests 
 is really capable of determining whether the subject 
 was dominantly of the visual or of the auditory type. 
 Not having determined that first essential fact for a 
 comparison of the relative advantage of the visual or 
 auditory bias upon the reading rate, conclusions based 
 upon correlations of reading rate with prehension span 
 or similar exercises are quite lacking in validity. But 
 even if it were assumed that the imagery types were 
 established with precision, a conclusion as to the de- 
 cided advantage of the visual over the auditory would 
 scarcely be warranted in view of the conflicting char- 
 acter of the results obtained. 
 
 Thus the results meager and conflicting as they 
 are of the only direct experimental investigation of 
 this problem would hardly seem to support so positive 
 a conclusion as Schmidt's that "the auditory-motor 
 type of reader can compete with the visual only when 
 he is able to resort to scanning." 
 
 Meumann does not treat specifically of the effect 
 of the imagery types upon rate of reading, but he does 
 discuss their influence upon rate of associative learning. 
 In this connection, he states that "the auditory-motor 
 individual always seems to possess a less reliable, but 
 a more rapid memory" than the individual of the visual
 
 TRAINING IN PERCEPTION 141 
 
 type. The auditory-motor type seems to function 
 more rapidly in associative learning. 
 
 To the question then : Is the visual or the auditory 
 bias more conducive to rapidity in reading, no positive, 
 definite answer, in the writer's judgment, can as yet 
 be given. The paucity of experimental evidence on 
 this point is such as to render any conclusion but ten- 
 tative and highly conjectural. The conclusion of 
 Schmidt's ranking the visual type as the most advan- 
 tageous, the motor type the least, and the auditory 
 as intermediate, while theoretically very plausible, 
 is as yet without sufficient experimental corroboration. 
 But the meager evidence that does exist warrants no 
 such fatalistic attitude as Schmidt's concerning the all- 
 powerful, conditioning influence of the imagery type 
 upon reading rate. Rapid readers have been found 
 among all types of imagery. In the present study the 
 vast increase in speed reported of all the experimental 
 pupils in certain grades, in which there were beyond 
 doubt different types of imagery indicates that 
 marked improvement in rate may be secured regardless 
 of the pupil's mode of imaging. 
 
 Training in Perception 
 
 The third type of training has been based directly 
 upon the findings of the psychological laboratory as to 
 the functioning of the eyes during reading, showing 
 the crucial importance of developing and widening the 
 perceptual span, or at least of making more effective
 
 142 SILENT READING 
 
 use of it in reading, of decreasing the duration of the 
 fixation-pauses, and of cultivating a regular, rhythmical 
 swing of the eyes in the inter-fixation movements. 
 The method was worked out carefully in its entirety. 
 It was found, however, that none of the present types 
 of tachistoscope could be used satisfactorily for groups 
 of pupils. In class exercises the word or phrase exposed 
 should be legible equally (or approximately) to every 
 pupil. An apparatus which would expose cards, con- 
 taining words and phrases of varying length, suitably 
 to a whole class, for intervals ranging from about one- 
 twelfth of a second to two seconds, and which would, 
 moreover, be easily manipulated by the teacher was 
 found necessary if the work was to be done with the 
 scientific accuracy necessary for the purposes of this 
 experiment. 
 
 Such an instrument could not be devised in the time 
 at the writer's disposal before the launching of the ex- 
 periment on the scheduled date. C. H. Stoelting of 
 Chicago, with whom the plans were carefully discussed, 
 has given the writer the assurance that an instrument 
 of this type suitable for classroom purposes can be 
 devised. Superintendent Taylor of Oregon has also 
 worked out plans for its construction. The proper 
 instrument may be available shortly. Though this 
 type of training was not actually applied in the present 
 investigation because of the lack of proper technical 
 apparatus, it is thought that the presentation of an 
 outline of the method may be worth while. This would
 
 TRAINING IN PERCEPTION 143 
 
 seem to be so in view of the probable appearance of the 
 necessary mechanical device in the near future and be- 
 cause there is considerable evidence which points to a 
 method of this kind as capable of greajbly accelerating 
 the reading rate. It is the hope of the promoters of 
 this experiment to submit this type of training to an 
 actual test when the necessary mechanical facilities are 
 at hand. 
 
 STATEMENT OF 
 
 TYPE III. TRAINING IN RAPIDITY OF READING 
 BY MEANS OF PERCEPTION CARDS 
 
 The cards used should be four inches wide and of varying 
 lengths to contain words, phrases, or sentences. 
 
 The use of these cards is analogous to that of the exposure 
 of material in laboratory work by means of an exposure ap- 
 paratus or tachistoscope. In both cases, the subject is re- 
 quired to perceive the material exposed in as short a time as 
 possible. There is experimental evidence tending to show 
 that the amount of verbal material which a person can per- 
 ceive in a given time can be increased in other words, that 
 perception may be improved. Probably it is not true that 
 perception as a native endowment may be improved, but the 
 method of using the native endowment i.e. the perform- 
 ance, is susceptible of extensive improvement. 
 
 The rate at which children can perceive words, phrases, 
 and sentences has a rather obvious bearing on their speed of 
 reading. Reading consists of a successive series of percep- 
 tions. The eye does not move regularly along the printed 
 line but jumps from one point to another, pausing at each 
 point. It is during the pause that perception takes place. 
 In a very real sense, therefore, the printed line consists of a 
 series of flashes or exposures.
 
 144 SILENT READING 
 
 To improve the rate of reading, therefore, we need to 
 reduce the number of exposures per line by increasing the 
 amount of material presented at each exposure and to shorten 
 the length of each exposure. This suggests that we may 
 directly attack the problem of increasing the speed of reading 
 (a) by training the pupil to "see more at a glance," and 
 (6) by training him to see the material more quickly. 
 
 Obviously, we cannot control either the amount seen at a 
 glance or the time of the exposure by using the printed page. 
 If, however, we can present, by means of perception cards, 
 portions of the text in the form of words, phrases, and sen- 
 tences, we may gradually lengthen the amount of material 
 on the cards as the pupil's ability to perceive it increases, 
 and we may also reduce the time during which it is placed 
 in view. 
 
 This is the fundamental idea in the method we are now 
 proposing. 
 
 The reading period should be 30 minutes long. Half the 
 period should be devoted to the presentation of material by 
 means of the perception cards. The remainder of the period 
 should be devoted to reading from the book in which, as 
 far as possible, the same words, phrases, and sentences are 
 met. 
 
 During the first half of the period the attempt should 
 always be to make the period of exposure of the cards as short 
 as possible. Day by day the length of the material should 
 be increased. In fact, either one of two methods may be 
 used and both should be used during the course of the ex- 
 periment: (a) the time of exposure may be kept constant 
 and the length of the material may be increased; (6) the 
 length of the material may be kept constant and the time de- 
 creased. 
 
 In order to obviate as far as possible any interference with 
 the perceptive processes due to the presence of unfamiliar
 
 TRAINING IN PERCEPTION 145 
 
 words, the teacher should drill on single words which she has 
 'reason to believe are unknown to a number of the children. 
 These words may be placed on the board or presented in the 
 usual manner. 
 
 It is desirable that the material be printed on the percep- 
 tion cards rather than written. For this purpose a stamping 
 outfit should be provided. It is recognized also that a great 
 deal of work will be involved in preparation for the class 
 period. The promoters of the experiment are willing to 
 cooperate with the teachers in securing help for this work. 
 
 The second part of the period should be devoted to a con- 
 scious attempt on the part of the teacher to secure a transfer 
 of the perception processes from the cards to the printed page. 
 When a phrase or sentence is encountered, children should 
 be told that the phrase or sentence is to be "seen all together." 
 Such directions as "do not look at every word," "see it as 
 a whole," "read it all together," may serve to carry over 
 from the work with the cards to the work in the book 
 something of the same attitude. 
 
 This method should continue from the first Monday in 
 April to the last Friday in May. An accurate record of the 
 progress of the pupils should be kept so that the increase in 
 perceptual span and in rapidity of perception may be noted. 
 A simple form will be provided for this purpose. The 
 teacher is also requested to keep a diary in which she will 
 daily record notes and observations concerning the progress 
 of the experiment. 
 
 The teacher should select easy reading material for use 
 during this experiment in order that the difficulty due to 
 unfamiliar words may be reduced. In all the reading other 
 than that done in the half hours set aside for the experiment, 
 the teacher should seek by admonition, encouragement, and 
 example to inculcate the idea of increasing the span of per- 
 ception e.g. reading by phrases and to decrease the time
 
 146 SILENT READING 
 
 required to "look at" a group of words. An interesting 
 exercise may be devised in which pupils may be requested to 
 glance at the first words of a paragraph and immediately 
 raise their eyes, whereupon they may tell "how much they 
 saw." 
 
 The following suggestions are offered indicating the nature 
 of the directions to be given to the pupils. Literal adherence 
 to them is not requested. Point out the advantage of a rapid 
 rate of reading. Try to get them to see this in terms of an 
 addition to their own interests and pleasure. Say to them in 
 substance : "I shall show you some cards containing words, 
 phrases, and sentences. They will be shown for only a frac- 
 tion of a second. So you will have to read them very quickly, 
 with a single glance of the eyes. Try to read all that is 
 printed on each card. You will have to pay very close at- 
 tention when I show you the card, as otherwise you will not 
 be able to read all that is on the card. This practice in read- 
 ing a number of words at a glance will help you to read more 
 rapidly. Try to do better than you did yesterday." 
 
 Spend about 15 minutes in perception card training. 
 After each exposure ask some of the pupils what was printed 
 on the card. Toward the end of the training in perception, 
 have all the pupils reproduce in writing what was on each of 
 the five last cards exposed. The percentage of words cor- 
 rectly reproduced will constitute the pupil's score in percep- 
 tion. State the total number of words exposed on the five 
 cards and it will then be easy for each pupil to figure out 
 his own score. Direct each pupil to enter his score imme- 
 diately upon the chart that each pupil will keep for that pur- 
 pose. 
 
 The second half of the period should now be devoted to 
 rapid silent reading. The pupils should be encouraged to 
 carry over the habit of "seeing many words at a glance" 
 gained by their training in perception, into the actual work
 
 TRAINING IN PERCEPTION 147 
 
 of silent reading. They should be directed to " run their eye 
 across a line of words" as rapidly as they can, consistently 
 of course, with an understanding of what they see. Through- 
 out this reading period there should be a conscious effort to 
 utilize the habit acquired through training in perception, by 
 perceiving several words at each fixation, instead of but a 
 single word. This should result in an increased rate of speed 
 in silent reading. 
 
 Say to the pupils in substance : " Read this selection as 
 fast as you can. I want to see how much you can read in 
 
 l minutes. Try to see as much as you can read in a single 
 
 glance, running your eyes rapidly across the lines. A fast, 
 regular, rhythmical movement of the eyes will help you to 
 read rapidly. But remember that I am going to ask you to 
 tell me about the matter you have read, so do not skip any- 
 thing. Try to read faster to-day than you did yesterday." 
 
 See that the pupils have a pencil at hand and direct them 
 to stop reading at once as soon as you say "stop." Direct 
 them then to mark the end of the line which they are reading 
 when told to stop. Pupils may now reproduce what they 
 have read. Reproduction should consist both of free para- 
 phrase orally, or in writing and of answers to specific 
 questions based on the text. The length of the reading period 
 and of the reproduction period should vary with the grade of 
 the pupils and with the subject matter. In general, however, 
 the reproduction should not occupy more than one-quarter 
 of the total time allowed for the exercise. Reading and 
 reproduction are to be continued till the end of the 15 min- 
 utes assigned. Have the pupils then report the number of 
 pages and lines beyond the last full page which they read. 
 
 As part of your preparation for the exercise you will be 
 expected to know the average number of words per line in the 
 
 1 Number of minutes, say two, three, or four, to be determined by 
 the teacher.
 
 148 SILENT READING 
 
 matter which is being read and the number of lines per page 
 (if pages are broken by illustrations or for other reasons, 
 special account of these pages will have to be 'taken). From 
 the pupils' reports as to pages and lines read on this and sub- 
 sequent assignments for the day, estimate the number of 
 words read that day and divide by the total number of min- 
 utes used in reading. This will give the number of words 
 read per minute by each pupil. Each student should figure 
 out his own score and should enter it immediately on the 
 chart on which he has already written his perception score. 
 This chart will thus serve as a record of the pupil's daily 
 progress. It is also suggested that a complete chart of the 
 daily class performance might be conspicuously displayed 
 in the classroom. 
 
 On May 1 and May 28, besides the record of the aver- 
 age number of words read per minute by each pupil during 
 the total reading time, a record of the average number of 
 words read per minute by each pupil during each of the vari- 
 ous reading periods, constituting the total time devoted to 
 reading on each of the above mentioned days, should also be 
 kept. 
 
 On May 2 and May 29, the pupils should be tested with 
 the Courtis Silent Reading Tests Forms II and III. As soon 
 as possible after the last Friday in May, a copy (or the origi- 
 nals) of the individual charts, the class chart, and a record of 
 results and observations should be mailed to the Bureau of 
 Educational Research, University of Illinois, Urbana, Illinois.
 
 CHAPTER VII 
 STATEMENT OF PROCEDURE 
 
 AFTER having devised the preceding types of train- 
 ing, the next step was to apply them to a sufficiently 
 large number of pupils in the different grades of the 
 elementary school. This would serve to test the 
 efficacy of the factors included therein. Although these 
 factors had been selected because psychological experi- 
 mentation, pedagogical investigations, and a priori 
 evidence had suggested their special significance in 
 accelerating the reading rate, their value as constituents 
 of an organized and usable method of training to in- 
 crease speed could be determined only by subjecting 
 these methods to an actual test under ordinary school- 
 room conditions. 
 
 Fortunately, at this time the newly established Bureau 
 of Educational Research at the University of Illinois, 
 as one phase of its activities in stimulating the interest 
 of the schools of the state in the scientific study of edu- 
 cation, had secured the names of the superintendents 
 and teachers who were interested in the problem of 
 reading. From this number and from others who 
 graciously offered to participate in this study, the 
 following selection was made. It comprises 40 grades 
 in 20 schools located in nine cities in Illinois. 
 
 149
 
 150 
 
 SILENT READING 
 
 ClTT 
 
 SCHOOL 
 
 GRADE 
 
 TEACHER 
 
 Bloomington, 111. 
 
 St. Mary's 
 
 5 
 
 Sister Bernadine 
 
 Bloomington, 
 
 St. Mary's 
 
 6 
 
 Sister Mary 
 
 Chicago, . . . 
 
 Goodrich 
 
 4 
 
 Clara E. Klass 
 
 Chicago, . . . 
 
 Goodrich 
 
 5 
 
 Virginia Lee 
 
 Chicago, . . . 
 
 Hendricks 
 
 5 
 
 Rose McComskey 
 
 Chicago, . . . 
 
 Hendricks 
 
 7 
 
 Bridget McElherne 
 
 Chicago, . 
 
 Hendricks 
 
 8 
 
 Ida M. Stodder 
 
 Chicago, . . . 
 
 Hendricks 
 
 4 
 
 Gertrude Tighe 
 
 Chicago, . . . 
 
 Hendricks 
 
 5 
 
 Ruth Wimmer 
 
 Chicago, . . . 
 
 Manierre 
 
 8 
 
 Margaret Dougherty 
 
 Chicago, . . . 
 
 Manierre 
 
 5 
 
 Mrs. Nellie Hahn 
 
 LaSalle . . . 
 
 Lincoln 
 
 7 
 
 Mary Y. Duncan 
 
 LaSalle, . . . 
 
 Lincoln 
 
 4 
 
 Margaret Flanagan 
 
 LaSalle, . . . 
 
 Jefferson 
 
 7 
 
 Elizabeth Hanley 
 
 LaSalle, . . . 
 
 Jefferson 
 
 4 
 
 Marcella Mullin 
 
 Oregon, . . . 
 
 Oregon 
 
 7 
 
 Elizabeth Joiner 
 
 Oregon, . . . 
 
 Oregon 
 
 4 
 
 Marion Metzger 
 
 Rushville, . . . 
 
 Webster 
 
 6 
 
 Minnie B. Thomas 
 
 Rushville, . . . 
 
 Webster 
 
 3 
 
 Dena Tutt 
 
 Streator, . . . 
 
 Garfield 
 
 3 
 
 Minnie Shinn 
 
 Streator, . . . 
 
 Grant 
 
 6 
 
 Lulu Leland 
 
 Streator, . . . 
 
 Grant 
 
 3 
 
 Frances C. Linskey 
 
 Streator, . . . 
 
 Grant 
 
 8 
 
 J. E. Mather 
 
 Streator, . . . 
 
 Greeley 
 
 6 
 
 Rose M. Boyd 
 
 Streator, . . . 
 
 Greeley 
 
 7 
 
 Gertrude Scanlan 
 
 Streator, . . . 
 
 Lincoln 
 
 4 
 
 Lalla Roberts 
 
 Streator, . . . 
 
 Plumb 
 
 5 
 
 Lillian SeCora 
 
 Streator, . . . 
 
 Sherman 
 
 6 
 
 Ruth Kelley 
 
 Argo, .... 
 
 Argo 
 
 4 
 
 Marie McCarthy 
 
 Argo, .... 
 
 Argo 
 
 5 
 
 Mary L. McGuire 
 
 Sycamore, . . . 
 
 Central 
 
 4 
 
 Hazel B. Mason 
 
 Sycamore, . . . 
 
 Central 
 
 7 
 
 Florence Wolhensack 
 
 Taylorville, . . 
 
 East 
 
 4 
 
 Mrs. Agnes Abrams 
 
 Taylorville, . . 
 
 East 
 
 7 
 
 Leonora Drennen 
 
 Taylorville, . . 
 
 Hewitt 
 
 7 
 
 Maude Law 
 
 Taylorville, . . 
 
 Hewitt 
 
 4 
 
 Bessie Leigh 
 
 Taylorville, . . 
 
 Oak St. 
 
 4 
 
 Leona Peek 
 
 Taylorville, . . 
 
 Oak St. 
 
 7 
 
 Florence Layzell 
 
 Taylorville, . . 
 
 West 
 
 4 
 
 Mrs. Laura Huff 
 
 Taylorville, . . 
 
 West 
 
 7 
 
 Mrs. L. P. Anderson
 
 STATEMENT OF PROCEDURE 151 
 
 Among the number is a parochial school, St. Mary's 
 in Bloomington ; the rest are public schools. The 
 cities are fairly well scattered throughout the state. 
 They range from the metropolis Chicago, down to the 
 smaller cities of the state, such as Sycamore. The 
 constituency of these 20 different schools largely re- 
 flects the cosmopolitan character of our school popu- 
 lation. The pupils of the Goodrich School in Chicago 
 are almost exclusively of Italian parentage; those in 
 the Argo School are dominantly of Polish extraction. 
 The offspring of Bohemians, Slavs, Austrians, and 
 Hungarians are found in goodly numbers in the 
 Streator schools ; while children of native-born Ameri- 
 cans predominate in the remaining schools. The 
 subjects of this experiment were drawn, moreover, from 
 practically every social strata of our heterogeneous 
 population. This wide range in the character of the 
 subjects would seem to render any appreciable gains 
 in reading rate all the more interesting and significant. 
 It would give to them a broader basis, and would illus- 
 trate the wide applicability of the methods of training 
 to different types of subjects. 
 
 The training was extended throughout grades three 
 to eight, inclusive. Since reading in the primary grades 
 is so largely devoted to the mastery of the mechanics 
 of reading the distinguishing of the different letter- 
 forms, their phonetic properties taken singly and in 
 various combinations, and the recognition of simple 
 word-forms, it was not practicable to apply the
 
 152 SILENT READING 
 
 methods to the first and second grades. Indeed, in 
 the writer's judgment, it is not possible to devise a 
 type of training which is suitable alike to pupils in the 
 > first grade and to those in the eighth. The former 
 are taken up almost entirely with the mechanics of 
 reading, the latter have passed that stage completely. 
 From the third grade on, however, the method is appli- 
 cable since the formation of habits of rapid efficient 
 silent reading becomes in these grades a matter of 
 paramount importance. 
 
 LENGTH OF TRAINING 
 
 The total period of training extended from April 8, 
 1919, to May 29, 1919 a period of 39 school days. 
 On April 8, May 2, and May 29, the reading period was 
 devoted to the administration of the Courtis Silent 
 Reading Tests. This left 36 school days in which the 
 entire reading period of 30 minutes was devoted to the 
 types of training. After 18 days of training, Form II 
 of the Courtis Silent Reading Test was administered 
 to ascertain how much improvement in speed and com- 
 prehension was effected in the first half of the training 
 period, and how much in the latter half. The tests were 
 also given to the control groups at the same time in 
 order to secure a check for each half of the training 
 period. In many of the schools, the period of 39 school 
 days from April 8 to May 29 was interrupted by an 
 occasional free day, or vacation periods of one kind 
 or another. In every such case the date set for the
 
 STATEMENT OF PROCEDURE 153 
 
 administration of the final Courtis Test was extended 
 beyond May 29, so that the uniform number of reading 
 periods might be devoted to the training. 
 
 METHOD OP CONTROL 
 
 In order to determine how much of the increase in 
 speed and comprehension could justly be attributed 
 to the particular type of training which the pupils 
 received and not to their regular school work, each of 
 the 40 classes was divided into two groups, approxi- 
 mately equal in rate of reading. The obtaining of a 
 group whose aggregate ability in the particular activity 
 that is being investigated is approximately equal 
 to the aggregate ability of the experimental group is 
 the ordinary method of "control" in experimentation 
 of this character. An attempt was made in the present 
 investigation to secure in addition a more refined type 
 of control. Besides dividing the class into two groups, 
 the sums of whose aggregate scores were approximately 
 equal, the class was further subdivided into pairs of 
 pupils of approximately equal speed in reading. One 
 member of each of the pairs was placed in the experi- 
 mental group, the other in the control group. This 
 afforded a control not only for the experimental group 
 as a whole, but also for each individual member in the 
 group. 
 
 The advantages of the method of individual control 
 are unmistakable. Chief among them is the securing 
 of a basis of comparison which rests directly and inti-
 
 154 SILENT READING 
 
 mately upon the detailed specific facts rather than upon 
 specious statistical group totals and averages. The 
 equality between the constituent members of the two 
 groups apparently reflected in the equality of the group 
 totals is oftentimes more apparent than real. Though 
 the totals of two groups may be approximately equal, 
 the individual members in the two groups may not be 
 at all comparable. 
 
 The following case serves to illustrate the application 
 of this point in connection with the method of control 
 in the present study. Before receiving instructions 
 as to the method of procedure, a principal had con- 
 templated dividing his class into four sections, ranging 
 from the highest to the lowest in reading rate. The 
 sum of the scores of the upper and lower quartiles would 
 be approximately equal to the sum of the scores of the 
 two middle sections. The former were intended to 
 serve as the control group for the latter. Though the 
 total number of points scored by each group would be 
 approximately equal, the absolute lack of equality in the 
 scores of all the individual members of the experimental 
 group with the scores of the individual pupils in the 
 control group, would be not less real. Consequently, 
 a common basis of comparison between the individual 
 members of the two groups would be utterly lacking. 
 The equality between the individual members which 
 the group totals would apparently reflect would be 
 a statistical fiction, sine fundamento in re. 
 
 The method of individual control serves, moreover,
 
 STATEMENT OF PROCEDURE 155 
 
 to eliminate the vexing question, How are improve- 
 ments in different levels of ability to be compared ? 
 a question that is always with us in the loose, slipshod 
 method of group control. Take the case already cited 
 of the upper and lower quartiles serving as the control 
 group for the experimental section consisting of the two 
 middle quartiles. The members of the lower quartiles 
 have open to them a much greater room for possible 
 improvement before reaching either the standard rates 
 or the highest physiological limit, than say the pupils 
 of the second quartile. Is an increase of 50 words per 
 minute in speed to be evaluated no more highly in the 
 latter than in the former ? If it should receive a higher 
 proportional credit, what will be the correct propor- 
 tion? These are some of the questions which, though 
 not frequently expressly raised by investigators, and 
 less frequently answered, are nevertheless always im- 
 plicit in the method of group control. 
 
 To obviate them and to secure a more refined type 
 of checking device one that is based more directly 
 and intimately upon the detailed facts in the case 
 the method of individual control has been used in the 
 present investigation. It has cost much in the way of 
 additional pains, time, and labor. The writer be- 
 lieves, however, that the more incisive and clear-cut 
 basis of comparison which resulted was worth the price. 
 
 The following extracts from a circular letter sent to 
 all schools participating will show the method em- 
 ployed by the teachers in securing both the group and
 
 156 
 
 SILENT READING 
 
 the individual control. The explicit character of the 
 directions indicates the pains taken to insure uniformity 
 of procedure in this important phase of the experiment. 
 
 Directions for Division of Class 
 
 "We are mailing you - - 1 copies of the Courtis Silent 
 Reading Test. Kindly have the teachers you have selected 
 administer these tests to their pupils on Friday, April 4. 
 
 " On the basis of the results of these tests have them divide 
 each of the above mentioned grades into two groups of 
 approximately equal speed in silent reading. In this division 
 only the score in speed (or words read) should be regarded. 
 The score in comprehension (questions answered) should 
 be marked, but should not be taken into consideration in the 
 question of dividing the grade into two groups. These 
 groups should be built up as follows : 
 
 ILLUSTRATION FOR CLASS OF TWENTY 
 
 
 GROUP A 
 
 GROUP B 
 
 Pair 1 
 
 190 
 
 188 
 
 Pair 2 
 
 185 
 
 185 
 
 Pair 3 
 
 179 
 
 180 
 
 Pair 9 
 
 68 
 
 67 
 
 Pair 10 
 
 50 
 
 55 
 
 
 
 
 Total 
 
 1452 
 
 1449 
 
 
 
 
 " Rank the pupils in each entire class on the basis of speed 
 from the best to the poorest. Number them in order, be- 
 ginning with the best (1, 2, 3, etc.), assigning them a rank. 
 Put all the odd numbered children in Group A, and all the 
 even numbered children in Group B. You will then have 
 1 Number inserted.
 
 STATEMENT OF PROCEDURE 157 
 
 two groups of approximately the same scores. Write in the 
 scores of the pupils. The tendency will be for the scores of 
 Group A to run higher. Now make minor adjustments by 
 shifting pupils from one group to the other until you have 
 (a) pairs of pupils of as nearly the same score as possible, 
 and (6) as a result, the total group scores nearly the same. 
 
 " Group A will be called the Experimental Group and 
 Group B the Control Group. This means that the method 
 of instructions set forth in the inclosed directions is to be 
 followed for Group A only, and that Group B is to be taught 
 in the usual way. If there is an odd number of pupils in a 
 class, put the odd pupil in Group A. 
 
 " After having scored the papers and used them to divide 
 the classes into two groups, kindly send the papers to us 
 for purposes of diagnosis and further study. Also kindly 
 inclose the test blanks which were not used. Ship by ex- 
 press collect. 
 
 " We are also sending you a sample form of chart for the 
 individual pupil, and a sufficient number of sheets, so that 
 each pupil can make his own individual chart under the 
 direction of the teacher. It is suggested that this may be 
 done in a drawing or arithmetic period on Friday, April 4. 
 Actual work on the reading project should begin on Tuesday, 
 April 8. The testing and division of the classes into two 
 groups, as well as the preparation of charts, should therefore 
 be finished before Tuesday, April 8." 
 
 In the Courtis Silent Reading Test, the results of the 
 performance are divided into the score for speed and 
 the score for comprehension. Unlike the Kansas 
 Silent Reading Test, there is no compound score serving 
 as the single index of the efficiency of the combined 
 qualities of speed and comprehension. In the above
 
 158 SILENT READING 
 
 set of directions, it will be noticed that the score in 
 speed, and not the score in comprehension, constituted 
 the basis for the division of the class into groups and 
 pairs. This was because the investigator, being pri- 
 marily interested in the development of speed, was 
 thereby enabled to compare the progress of the two 
 groups in this particular phase of reading more satis- 
 factorily than if either the score for comprehension 
 or the score for the combined qualities of speed and 
 comprehension were used as the basis for the division. 
 
 The Individual Chart 
 
 Every pupil in the experimental groups was re- 
 quested to keep an individual chart. Since uniformity 
 in the computation of the chart was necessary in order 
 that the graphs of the pupils in the different grades 
 might be readily comparable, a set of directions was 
 sent to every teacher to insure uniformity of procedure 
 on the part of the pupils. As many of the pupils were 
 in the third and fourth grades, and had had no previous 
 experience in graphing, it was found necessary to make 
 the directions very explicit and detailed. Besides the 
 following set of directions, a sample copy of the indi- 
 vidual chart was sent to each teacher. 
 
 DIRECTIONS TO TEACHERS FOR MAKING INDIVIDUAL 
 
 CHARTS 
 
 On Friday, April 4, have each child make a blank form like 
 the sample submitted. On account of the capacity of the
 
 STATEMENT OF PROCEDURE 159 
 
 sheet, the diagram is arranged so as to record the number of 
 words per half-minute. 
 
 Observe that the dates during which the special instruction 
 takes place are arranged from left to right along the base line 
 and that the speed in words per half -minute is expressed along 
 the vertical line. The idea is this : If a child reads on a 
 given day at an average speed of 220 words per minute, 
 divide 220 by 2, giving 110 as the average number of words 
 per half-minute. Suppose that this record was made by the 
 pupil in question on April 8. Have the pupil put a dot above 
 the date and at 110. He will make no other entry on that 
 day. Suppose, however, that on the next day (April 9) he 
 reads at the rate of 240 words per minute, or 120 words per 
 half-minute. He will then put a dot over April 9 and oppo- 
 site 120, at the same time drawing a line between the dot he 
 entered the day before and the one he has just entered. 
 
 On subsequent days he will record his speed in the same 
 way, each time putting a dot over the date and opposite the 
 number of words he read per half-minute, and drawing a line 
 to connect this dot with the last one he has made. If the 
 average number of words per half-minute for a given child 
 on a certain day lies between two numbers written at the left 
 side of the diagram, let him estimate how high to place the 
 dot. Suppose, for example, that he has read 268 words per 
 minute, or 134 words per half -minute. The dot will then be 
 not quite halfway up from 130 to 140. 
 
 It has been found not only that children rather readily 
 understand the making of such a chart but that they enjoy 
 watching their curves grow. Of course, the interest lies 
 in seeing the extent to which the slopes upward indicate 
 improvement. 
 
 If there is a break in the record due, for example, to the 
 absence of a child the dot should be entered above the date 
 of the day when the record is resumed, and a line should be
 
 160 SILENT READING 
 
 drawn from the last dot to the one just made. This line 
 will cross over the intervening day or days, and to indicate 
 that a break has occurred, the line should be dotted. 
 
 Observe that two school days are omitted along the base 
 line, namely, May 2 and May 29. These are the days on 
 which you are requested to test your pupils with the Courtis 
 Silent Reading Tests. You will not be expected to conduct a 
 reading recitation in either Group A or Group B on those days. 
 
 It is suggested that added interest may be given to the 
 work if you have a class chart on which you will enter the 
 average speeds of reading for the class in precisely the same 
 way that the children will enter their individual results. If 
 you have sufficient blackboard space, a chart may be drawn 
 on the board. A large sheet of mounting paper will, however, 
 be fully as satisfactory. 
 
 Reading Conference 
 
 Despite the pains taken to formulate directions that 
 would guide the teachers in every important step in the 
 procedure, it was recognized that differences in the 
 application of the method by the various teachers might 
 still result. These differences might mar or at least 
 weaken the accuracy of the conclusions that would be 
 drawn from the study. If the teachers could witness 
 the concrete application of the method, and the mode 
 of computation, besides having ari opportunity to 
 present orally any difficulties which occurred to them, 
 and have them cleared away, a greater guarantee for 
 uniformity of procedure would be secured. Accord- 
 ingly, a representative of each school system 1 partici- 
 
 1 There was no delegate from Bloomington. The writer went there 
 personally to explain the method, its application, etc., to the teachers.
 
 STATEMENT OF PROCEDURE 
 
 161 
 
 pating in the experiment was invited to attend a con- 
 ference at the University of Illinois, where the method 
 was demonstrated and explained in detail. 
 
 The conference lasted an entire day. In order that 
 all the teachers might benefit from the conference, a 
 stenographic report of the demonstration class and of 
 the discussions by the delegates and the promoters of 
 the experiment was sent to every teacher. This served 
 to supplement the report which each delegate brought 
 back to her colleagues. The following extracts from 
 the circularized report of the conference will convey 
 some idea as to the nature of the meeting and the points 
 discussed. 
 
 "On Saturday, April 12, representatives of the various 
 school systems engaged in the Silent Reading Project as- 
 sembled for a conference at the University of Illinois. The 
 following is the list of those present : 
 
 Mr. D. E. McCraken 
 Mr. F. R. Nichols 
 Miss Clara F. Klass 
 Miss M. D. Wheaton. 
 Mr. C. B. Smith 
 Mr. D. L. O'Sullivan 
 Mr. F. G. Taylor 
 Miss Mary T. Duncan 
 Miss F. C. Lenskey 
 Miss Rose M. Boyd 
 Miss Maude Law 
 Miss Hazel B. Mason 
 
 " Miss Davis brought her fifth-grade class from the Lincoln 
 School, Champaign, and demonstrated the method of train- 
 ing the pupils in rapid silent reading. The total time allowed 
 
 Principal 
 
 Manierre School 
 
 Chicago 
 
 Principal 
 
 Hendricks School 
 
 Chicago 
 
 Teacher 4 
 
 Goodrich School 
 
 Chicago 
 
 Teacher 3 
 
 Goodrich School 
 
 Chicago 
 
 Superintendent 
 
 
 Rushville 
 
 Principal 
 
 
 Argo 
 
 Superintendent 
 
 
 Oregon 
 
 Principal 
 
 
 LaSalle 
 
 Teacher 3 
 
 Grant School 
 
 Streator 
 
 Teacher 6 
 
 Greeley School 
 
 Streator 
 
 Teacher 7 
 
 Hewitt School 
 
 Taylorville 
 
 Teacher 4 
 
 Central School 
 
 Sycamore
 
 162 SILENT READING 
 
 for the period was 30 minutes. The teacher secured seven 
 reading stretches consisting of 3, 4, 2, 4, 3, 3, 3 minute periods, 
 making a total of 22 minutes spent in actual reading. The 
 instructions, questions, and answers on the matter read 
 occupied only eight minutes." 
 
 There was considerable discussion by the delegates 
 as to the method of procedure in computing the aver- 
 age number of words per minute and of graphing. 
 The following extracts from the conference report will 
 serve as a resum6 of the chief points discussed and of 
 the conclusions finally agreed upon. 
 
 "1. There was a discussion as to the expediency of com- 
 puting the number of words read per minute from only one 
 of the reading stretches of say, two or three minutes' dura- 
 tion, or of computing the average number of words read per 
 minute from the sum total of all the reading stretches during 
 the 30-minute period. The representatives decided to 
 adopt the latter method. Consequently, if there have been 
 eight reading stretches of two and three minutes' duration, 
 so that, say, the sum of 20 minutes was spent in actual read- 
 ing, it would mean that the total number of words read, say 
 5000, is to be divided by 20, giving 250 as the average num- 
 ber of words read per minute. 
 
 "2. The question as to whether or not the teacher should 
 take down the number of the page reached by the pupil 
 after each reading stretch or merely at the end of the total 
 reading time was discussed at length. It was agreed to 
 leave that point to the discretion of each teacher which- 
 ever method she found the more effective she should be free 
 to adopt. Miss Davis reported that after trying it both 
 ways, she preferred to take the number of the page reached 
 by each pupil only at the end of the period, as this allowed
 
 STATEMENT OF PROCEDURE 163 
 
 more time to be spent in actual reading than was possible 
 under the other method. 
 
 "3. In making the class chart, the delegates decided to 
 use the median rather than the average. The median is 
 simply the score of the pupil middlemost in the order of rank. 
 In a class of 21 pupils, the score of the llth pupil would con- 
 stitute the median for the class. If there are an even number 
 of pupils, the median is the score midway between the scores 
 of the two middle pupils. In a class of 20, therefore, the sum 
 of the scores of the tenth and eleventh pupils when divided 
 by 2 gives the median. 
 
 "4. Suitable reading material for the different grades was 
 discussed. It is important that the material chosen be not 
 too difficult for the grade. Easy material is to be preferred 
 to make possible the setting up of habits of rapid eye- 
 movements." 
 
 Inasmuch as the pupils in the experimental groups 
 were covering two, three, four times the usual 
 amount of reading matter, the problem of supplying 
 them with sufficient suitable material soon became a 
 pressing one. The reading material that was on hand 
 for the particular grade was exhausted by the experi- 
 mental pupils in the course of a few weeks' practice. 
 A list of selections suitable for the different grades was 
 submitted by the delegates. This list was incorporated 
 into the Report of the Reading Conference. In addi- 
 tion to this, Mr. James F. Hosic of the Chicago Normal 
 School kindly selected a list of books especially suit- 
 able for the purposes of the experiment from a much 
 larger number which he had prepared for use in certain 
 Chicago schools designated for the time as "English
 
 164 SILENT READING 
 
 centers." A copy of this list was also sent to each 
 teacher. Both of these lists will be found in the ap- 
 pendix to this study. 
 
 The Work Curve for One Reading Period 
 
 The reading period is divided into five, six, or seven 
 stretches of two, three, and occasionally of four min- 
 utes' duration. The method of computation which the 
 teachers decided to adopt secures the average number 
 of words read per minute by each pupil during the total 
 reading period. It does not determine, however, the 
 average number of words read per minute during each 
 of the reading stretches. It left unanswered the 
 questions : In which of the stretches do the pupils ob- 
 tain the highest speed? Does the intensive nature of 
 the work cause a marked decline towards the end of the 
 period ? Does the curve for this type of activity bear 
 much similarity to the work curve described by Offner 
 (70), Rivers and Kraepelin (71), and others ? To throw 
 some light upon these interesting queries, a number of 
 teachers in different grades were requested to secure 
 a record of the average number of words read by each 
 pupil during each of the various stretches on a particu- 
 lar day. 
 
 At the conclusion of the experiment, the test papers, 
 the individual graph, the class charts, teachers' notes 
 and observations were sent to the writer. A care- 
 ful examination of the scoring of the Courtis Tests, 
 especially in computing the index of comprehension,
 
 STATEMENT OF PROCEDURE 165 
 
 revealed too large a percentage of error. Most of the 
 computation of the scores had been done by the pupils, 
 as Courtis suggests. For the sake of accuracy, all of 
 the Courtis tests, including the three forms I, II, and III, 
 were reexamined and rescored by trained clerks under 
 the writer's supervision. The tremendous amount of 
 labor involved in recorrecting approximately 4000 test 
 papers, determining the score in speed per minute for 
 each one, computing the index of comprehension, mak- 
 ing the necessary readjustments in the matching of 
 the pairs of pupils, has been chiefly responsible for pre- 
 venting the publication of this investigation at an 
 earlier date.
 
 CHAPTER VIII 
 STATEMENT OF RESULTS 
 
 THE chief instrument used in this investigation to 
 measure the growth in speed of reading was the Courtis 
 Silent Reading Test No. 2. There are three different 
 forms of this test, each of which is approximately 
 equal in difficulty. The triplicate character of the test 
 made it possible to test the pupils at the beginning, 
 middle, and end of the training with the same kind of 
 measuring device, and yet avoid the complicating 
 factor of special familiarity with the specific subject 
 matter of the test. The uniform character of the three 
 forms of the test likewise renders feasible the compari- 
 son of the scores made in the beginning with the scr>res 
 made at the middle and at the end of the training. 
 The employment of different standardized tests on the 
 two latter occasions would have destroyed this uni- 
 formity in the character of the measuring device and 
 would have rendered such comparison hazardous. 
 
 Another advantage arising from the selection of the 
 Courtis test is that its division into two parts a 
 measurement of rate and a measurement of compre- 
 hension enables the investigator to ascertain the 
 
 166
 
 STATEMENT OF RESULTS 167 
 
 speed of a pupil's reading as distinct from his com- 
 prehension. It disentangles these two factors and 
 weighs them separately. 
 
 The Courtis Test is not, however, without its short- 
 comings. While it is helpful to know the score of the 
 two factors speed and comprehension separately, 
 it is not less helpful to know the score for these two 
 factors when functioning conjointly on the same pas- 
 sage. This the Courtis Test through its score of rate 
 and index of comprehension does not determine. At 
 the time the score for rate of reading is determined, 
 there is no measure taken of the comprehension that 
 accompanied such reading. Herein lies the chief weak- 
 ness of the Courtis Test. 
 
 By supplementing the index of comprehension - 
 which reflects only the proportion of the correct to the 
 incorrect answers, independently of the absolute num- 
 ber attempted by the number of correct answers, 
 this difficulty is in a large measure obviated. For the 
 number of correct answers reflects not only the ability 
 to comprehend the matter but the speed with which the 
 pupil reads and comprehends. 
 
 According to the method of computing the index of 
 comprehension as stated by Courtis, a pupil who at- 
 tempts only five questions in the five minutes allowed 
 and answers them correctly secures a mark of 100 per 
 cent in comprehension. So likewise does the pupil 
 who attempts 60 questions and answers them all cor- 
 rectly receive a mark of 100 per cent. Yet it is obvious
 
 168 SILENT READING 
 
 that there is a great difference in the performance of the 
 two pupils the latter is much superior. This differ- 
 ence is not reflected in Courtis' Index of Comprehension. 
 Accordingly, in this investigation, the number of ques- 
 tions correctly answered was used as a score of the 
 pupil's comprehension to supplement Courtis' Index, 
 and to express actual differences in the performances 
 of the pupils differences in levels of achievement. 
 This device enables one to determine whether the train- 
 ing was effective in increasing the rate of reading when 
 the comprehension is employed conjointly and is being 
 tested simultaneously. In fact, in many respects the 
 number of questions correctly answered is a better 
 measure of the efficiency of comprehension on the part 
 of the reader than the index as devised by Courtis. 
 
 Standardized tests are all open to the criticism that 
 they test but one performance of the subject. The 
 performance of the pupil on the particular day on which 
 he was tested may have been normal or it may not 
 have been. Due to any one of a thousand causes which 
 affect the variability of the functioning of a human 
 personality, the subject may not have been true to 
 form on that particular occasion ; he may not have 
 been able to display his highest ability. In other words, 
 the test gives but one sampling of a variable function- 
 ing. Whether that particular sampling really repre- 
 sents the pupil's ability is a matter which can be deter- 
 mined only by repeated testing. 
 
 In the present investigation the attempt has been
 
 STATEMENT OF RESULTS 169 
 
 made to obtain a picture of the pupil's daily growth in 
 speed of reading by securing a record of his daily 
 performance for a period of two school months. This 
 was accomplished by means of the individual chart, 
 which showed the average number of words read, per 
 minute by each pupil on each of the 36 days of the 
 training period. It represents the results of 36 measure- 
 ments of rate. As such, it has the advantage of a 
 certain degree of reliability which accrues from re- 
 peated testing. The disturbing factors of variability 
 are practically equalized in such a large number of 
 measurements. 
 
 While the individual graph has the advantage of a 
 much larger number of measurements over the stand- 
 ardized test, it is inferior to the latter in the un- 
 standardized character of the subject matter on which 
 the record was made. The individual chart represents 
 rates of speed in reading the ordinary type of material 
 which is found in the graded readers. In the majority 
 of cases it is safe to say that the matter is fairly well 
 adapted to the reading capacity of the pupils. But 
 it is not uniform for all the pupils. One fourth grade, 
 for example, uses one reader, another fourth grade uses 
 a different one. This is true for almost all the grades. 
 While all the reading books for one grade may be on the 
 same general level, there are doubtless minor differ- 
 ences in degree of difficulty which make the record 
 secured on one book not strictly comparable with the 
 records secured on another reading book. In the
 
 170 SILENT READING 
 
 Courtis Silent Reading Test, on the other hand, the 
 subject matter is absolutely uniform for all the pupils. 
 
 By using a fairly large number of different graded 
 readers, as the pupils did in this experiment, the scores 
 may be said to represent the rates on the average type 
 of reading material for the particular grade. It is 
 true that in any one class the average rate of speed 
 would fluctuate somewhat on account of differences in 
 degree of difficulty in different selections in the same 
 reader, and also on account of changing readers. The 
 large amount of material which the training required 
 to be read made the using of many readers inevitable. 
 In some instances the class charts give the names of 
 the reading selections which were employed on the 
 days in which the class averages in rate were made. 
 This serves as a sort of check on the rate. It explains, 
 in a number of cases, considerable variations which 
 appear in the class average from day to day. 
 
 Changes in the degree of difficulty of different 
 reading selections are somewhat of a disturbing factor 
 in measuring the gradual growth in speed for a single 
 class. They are much less of a disturbing factor, how- 
 ever, when the average represents not the mean of a 
 single class but the central tendency of a fairly large 
 number of classes. In other words, increases and de- 
 creases hi degree of difficulty of selections tend, in a 
 general way, to equalize each other when the average 
 reflects the performance made on a large number of 
 different selections. An illustration of this point may
 
 STATEMENT OF RESULTS 171 
 
 be readily had by comparing the curve of progress, 
 let us say, of a single fourth-grade class, with the curve of 
 progress of all the fourth-grade classes, as shown in the 
 composite Graph XII. The latter is seen to be much 
 smoother and more regular in its progress. 
 
 EFFECT OF THE TRAINING UPON THE RATE AND COM- 
 PREHENSION AS DETERMINED BY THE COURTIS 
 TESTS 
 
 1. Third-Grade Pupils 
 
 Table VIII shows a rather remarkable gain in rate 
 for the experimental pupils in two third-grade classes. 
 From an average rate of 152.1 words per minute at the 
 beginning of the experiment, the average rises to 239.5 
 words after one month of the experimental training. 
 At the end of the two months of training the high 
 average of 258.6 words per minute is reached. The 
 average rate of the "control" pupils advances from 
 137 words per minute to 183.8 after a month of the 
 conventional work in reading. In the course of the 
 next month the rate has advanced but six words per 
 minute, reaching a final average rate of 189. 8 per minute. 
 The average gain of the experimental pupils in rate is 
 more than twice the average gain of the "control" 
 group 106.5 words per minute as contrasted with 
 52.7 words per minute. 
 
 What was the effect of this acceleration of the read- 
 ing pace upon the comprehension ? Beginning with an 
 average index of comprehension of 76.4 per cent, the
 
 172 
 
 SILENT READING 
 
 TABLE VIII. COMPARISON OF THE RESULTS OF THE TRAINING IN RAPID SILENT READING WITH THE RESULTS OF 
 THE CONVENTIONAL WORK IN READING 
 
 Total Number of Classes 2 GRADE III Total Number of Pupils 32 
 
 M 
 
 fe 
 
 I 
 
 1 
 
 . 
 
 i 
 
 o 
 
 S^sS 
 
 To illustrate the reading of this table, the fourth line showing the averages is read thus : The rate of the experi- 
 mental (A) pupils of the two schools in the first Courtis Test at the beginning of the training is 152.1 words per 
 minute ; their comprehension is 76.4%^ The rate for the control (B) pupils is 137.1 words pe? minute, their com- 
 prehension is 75.7%. On the second Courtis Test given after one month of training, the rate for the experimental 
 pupils is 239.5 words, their comprehension 78.6% ; the rate for control pupils is 183.8, their compreSonTl% 
 Throughout this study, A is used to designate the experimental pupils ; B, the control pupils. 
 
 1 
 
 O rH CO > rH 00 
 
 < 
 
 I 
 
 CD rH CO ^ 
 rH * H 
 
 I 
 
 CO 00 J^ O 
 
 5 
 
 H 
 
 . 
 
 i 
 
 o 
 
 00 Tt 
 
 00 cC 
 
 "* CO rH (N 
 
 2 * r3 00 
 rH ^* 
 
 3 
 
 oS 
 
 tf 
 
 p O> t^ CO O 00 
 
 rH CO fli 2 00 
 
 222 822 So 
 
 - 
 
 I 
 
 O5 O 
 CN U3 
 Oi 1> 
 
 1C 00 
 
 
 1 
 
 rH r^ 
 CO CM 
 
 t> CO 
 
 rH OC 
 
 
 1 
 
 B 
 H 
 
 . 
 
 1 
 
 co * 
 
 l> t> 
 
 t> ^H 
 
 O iq 
 
 1 
 
 CO rH 
 t^ O! 
 
 O CO 
 
 CN r^ 
 
 - 
 
 1 
 
 rH CC5 
 00 t>- 
 
 l> CD 
 
 
 1 
 
 p CO 
 
 (N 1C 
 
 CO 5^, 
 
 
 H 
 
 M 
 
 ~ 
 d 
 
 O TJ< 
 
 O CN 
 
 oo t> 
 
 CM t--; 
 
 O 
 
 i-i 
 
 1 
 
 tf 
 
 p oq 
 * co 
 
 i I rH 
 
 CN 2 
 
 O5 'C 
 
 CO rH ! 
 
 - 
 
 a 
 
 o 
 O 
 
 CO O5 
 00 * 
 
 CN Tf 
 ^ CO 
 
 Total Superiority of A over B 
 Average " " A " B . . 
 
 1 
 
 TJH O 
 
 d 6 
 
 00 CO 
 
 CO rH 
 
 Is 
 
 I'i 
 
 * oo 
 
 w 
 
 i 
 
 Webster 
 Grant 
 
 V 
 
 tc 
 
 3 
 
 r > 
 
 o 
 
 Rushville 
 Streator
 
 STATEMENT OF RESULTS 173 
 
 score of the experimental pupils rises to 78.6 per cent 
 after one month's training. At the end of the second 
 month, it has mounted to 82.8 per cent a gain of 6.4 
 per cent. The increase in rate has, therefore, been 
 accompanied by an increase in comprehension. The 
 "control" pupils began with an average index of 75.7 
 per cent and at the end of two months' conventional 
 work in reading had dropped to 74.6 per cent a 
 decrease of 1.1 per cent. Consequently, the final 
 average gain of the experimental pupils over the con- 
 trol group is 53.8 words per minute in rate and 7.5 
 per cent in comprehension. 
 
 A clearer understanding of the way in which records 
 were assembled and compared perhaps requires that I 
 should again refer to the system of pairing. It is to 
 be understood that the 32 children constituting the two 
 third-grade classes were arranged in sixteen pairs, 
 each pair consisting of an "A" pupil and a "B" pupil. 
 The sixteen "A" pupils taken together constituted the 
 "A" group, and the sixteen "B" pupils taken together 
 constituted the "B" group. The "A" group was 
 taught by the experimental method, the "B" group by 
 the usual method. The pupils were paired on a basis 
 of their rate scores in the first Courtis Test. After the 
 second and third Courtis Tests had been applied, all 
 records of pupils were rejected if they were not present 
 at all three of the tests, and no pupil's record was in- 
 cluded unless the companion in the "pair" was also 
 present at all the tests.
 
 174 
 
 SILENT READING 
 
 TABLE IX. COMPARISON OF THE NUMBER OF PUPILS IN THE EXPERI- 
 MENTAL (A) GROUP WHO SURPASSED THEIR CORRESPONDING PUPILS 
 IN THE CONTROL (B) GROUP 
 
 GRADE III 
 
 
 TEST I 
 
 TEST n 
 
 TEST HI 
 
 
 Rate 
 
 Com. 
 
 Rate 
 
 Com. 
 
 Rate 
 
 Com. 
 
 Number of A equal to B 
 
 
 2 
 
 1 
 
 
 
 
 Number of B surpassing A . 
 
 7 
 
 8 
 
 3 
 
 7 
 
 4 
 
 9 
 
 Number of A surpassing B . 
 
 9 
 
 6 
 
 12 
 
 9 
 
 12 
 
 7 
 
 Group superiority A over B 
 
 2 
 
 
 9 
 
 2 
 
 8 
 
 
 Group superiority B over A 
 
 
 2 
 
 
 
 
 2 
 
 Final superiority A over B in 
 
 
 
 
 
 
 
 per cent 
 
 
 
 
 
 50 
 
 -121 
 
 
 To illustrate the reading of Table IX, the fourth line showing the 
 Group Superiority A over B is read thus : In the first Courtis Test, the 
 group superiority of the experimental (A) pupils over the control (B) 
 pupils in rate is 2 pupils; in comprehension, the "control" group excels 
 the experimental by 2 pupils. In the second Courtis Test the experi- 
 mental's superiority in rate is 9 pupils, in comprehension 2 pupils. At 
 the end of the training, there are 12 experimental pupils surpassing their 
 respective individual control pupils, while four members of the control 
 group surpass their corresponding pupils in the experimental class. 
 This leaves a final net group superiority of eight for the experimentals 
 in rate, as against a final net group superiority of two for the control 
 section in comprehension. 
 
 In Table IX the results of the method of individual 
 control are shown. Here the averages are replaced by 
 the number of pupils in one group who surpass their cor- 
 responding pupils in the other groups. The results of 
 this method of control are largely corroborative of the 
 results as shown by the average scores in Table VIII. 
 When reduced to a percentage basis, Table IX shows 
 a final net superiority of 50 per cent in rate for the ex- 
 perimental over the control group. The apparent
 
 STATEMENT OF RESULTS 
 
 175 
 
 W 
 
 
 
 
 
 I'lOOCOCsmOINeNO'O i-l>O 
 
 f r 
 
 
 as 
 
 * CO 
 
 
 P5 
 
 a 
 
 1 III 
 
 5 
 
 
 B CN 
 
 & 
 
 S a 
 
 O 
 
 
 s 
 
 SES8S7S2ci;8!8 
 
 
 
 ^ 
 
 
 
 
 
 COii-HCSt^r*OOCO^WC" C N 
 
 
 
 o 
 
 j 
 
 
 c3 
 
 1 1 r "~ " 
 
 
 
 6 * 
 
 
 
 cc 
 
 lOCO^'CSCOOi-lW'-'COIN >Ot~ 
 
 
 
 i 
 
 
 
 5 
 
 00 O W C> >-i CO i!5 C> Q CO u-5 3 
 
 
 
 H 1 
 
 
 
 s 
 
 a 
 
 ISI2ISSH?51i 
 
 Soi 
 
 
 wo ^ 
 
 
 ffl 
 
 (B 
 
 CN^CCTfr-O^OOXO O CN 
 
 CO CN 
 
 
 
 s 
 
 
 09 
 
 T}<XCNCO'^oix'^OC5CO c^Q5 
 
 |* 
 
 . 
 
 Q 
 
 1 
 
 
 
 t^)<OCNCNO'HCO^'HC> COO 
 
 
 I-H 
 I-H 
 
 is 
 
 H 
 
 
 a 
 
 OOl^-OXXXOOOOt^cO C5 X 
 
 
 > 
 5 
 
 ^< 
 
 
 ^! 
 
 
 
 f-cO^CNOCNXOt^'OO CO-^< 
 
 
 a 
 
 z 
 
 
 
 
 l^ilsislslll 
 
 
 H 
 S 
 
 
 
 
 
 
 lQ>-li-(CO>Oi-l>Oi-l>OO H 
 
 O CN 
 
 
 
 ?g 
 
 
 
 3 
 
 gsrissBK^siia s 
 
 s 
 
 i 
 
 BP 
 
 
 
 V 
 
 ^.CO^^OiOi-iXO^ tN>O 
 
 ^ ^" 
 
 P 
 
 ~3 
 
 b 
 
 
 1 
 
 SSSslSlaSS |i 
 
 isS 
 
 i 
 
 fc 
 
 g 
 
 
 
 CN O O X CO ffl O X 00 O "-ICN 
 
 
 
 8 > 
 
 
 
 o 
 O 
 
 CO 3< CO' -< -H ^' CN 3; O 1--5 S i-l 
 
 coxxoaxt^xxx^ xo 
 
 
 i 
 
 s 
 
 E 2 H 
 
 
 
 S 
 
 O"5-i'i-"W^J<i-TS5O O5CS 
 
 
 
 H l 1 
 
 
 
 <a 
 
 tf 
 
 |j||_|ii||| |s 
 
 
 i 
 
 OO n 
 
 
 
 
 
 
 
 ^ 
 
 00 
 
 
 PQ 
 
 S 
 o 
 O 
 
 x'cN'x'cd-<'a6or^'i<oiri fcx 
 
 COXt^Xt^t^XXt^XCO ^^* 
 
 T i 
 
 s 
 
 .9 
 
 *~^ S 
 
 
 
 S 
 
 r^r-cNt-coocNi-icocico 1-11-1 
 
 O CO 
 
 X 
 
 I H 
 
 J 
 
 
 
 ^SI1S1^222 2 
 
 
 -= 
 
 J3 
 
 
 i 
 
 
 
 oooTfcoocoaor^oo ao-<r 
 
 
 
 
 C oo 
 
 H 
 
 
 a 
 
 KS?:g{:SSSSgS g 
 
 
 
 H S 
 
 
 
 s 
 
 CO^OCOCQCOr^CO'OCNCO <-i ( 
 
 
 H 
 
 h| 
 
 
 
 
 CSOCOOt^^t^XCSCOO cO^ 
 
 
 2 
 
 2* 
 
 i 3 
 
 ; 
 
 i. " 
 
 ! o : 
 ;gj 
 
 t 
 
 CN 
 
 pan 
 
 
 COMPAF 
 
 r of Classes 
 
 
 SCHOOL 
 
 
 1 o a a a -3 . 
 
 a||||||l5 
 
 arity A over 
 A " 
 
 
 TABLE X. 
 Total Numb 
 
 
 1 
 
 
 
 
 Q _5 _ ggo> fc 
 
 s"3"5o-S-Eo:: of 
 
 &J** HI I"! ii 
 
 Total Superi 
 Average 

 
 176 SILENT READING 
 
 superiority of 12 per cent in comprehension for the 
 control sections over the experimental pupils is dis- 
 counted by the fact that they started with an initial 
 group superiority of two pupils or 12^ per cent. 
 
 The number of third-grade pupils who were matched 
 into pairs, one being in the experimental class, the other 
 in the control class, is too meager to justify any but 
 the most tentative kind of conclusions. The pupils 
 who survived the matching process were evidently the 
 more capable ones, and consequently their performance 
 can scarcely be said to be typical of third-grade pupils. 
 The results do show, however, that third-grade experi- 
 mental pupils profited noticeably by the training. The 
 average rate of speed which they finally obtained is 
 far in excess of the standard rate reported for that grade 
 by any of the authors of the various standardized read- 
 ing tests. Owing to the small number of third-grade 
 pupils participating in the experiment, the writer will 
 refrain from comparing these results with those obtained 
 in the other grades with a much larger number of pupils. 
 
 2. Fourth-Grade Pupils 
 
 In Table X are shown the average scores of 1 1 fourth- 
 grade classes. The total number of pupils is 236. 
 The average rates of 155.7 words per minute for the 
 experimental group, and of 155.1 for the control pupils, 
 show that the two sections start the experiment with 
 practically equal proficiency in rate. In comprehen- 
 sion, the two groups are also closely matched, the con-
 
 STATEMENT OF RESULTS 
 
 177 
 
 trol pupils having an average index of 78.6 per cent as 
 compared with 77.4 per cent for the experimentals. 
 
 TABLE XI. COMPARISON OF THE NUMBER OF PUPILS IN THE 
 EXPERIMENTAL (A) GROUP WHO SURPASSED THEIR CORRE- 
 SPONDING PUPILS IN THE CONTROL (B) GROUP 
 
 GRADE IV 
 
 
 TES 
 
 T I 
 
 TBS 
 
 T II 
 
 TESI 
 
 III 
 
 
 Rate 
 
 Com. 
 
 Rate 
 
 Com. 
 
 Rate 
 
 Com. 
 
 
 5 
 
 5 
 
 2 
 
 5 
 
 1 
 
 3 
 
 Number of B surpassing A .... 
 
 57 
 
 61 
 
 35 
 
 53 
 
 29 
 
 55 
 
 Number of A surpassing B . 
 
 36 
 
 52 
 
 72 
 
 51 
 
 88 
 
 60 
 
 Group Superiority A over B . . . 
 
 
 
 
 
 37 
 
 
 
 59 
 
 5 
 
 Group Superiority B over A ... 
 
 1 
 
 9 
 
 
 
 2 
 
 
 
 
 
 Final Superiority A over B in per cant 
 
 
 
 
 
 50 
 
 4 
 
 NOTE. Table XI is to be read in the same manner as Table IX. 
 
 After one month of training in rapid reading, the 
 experimentals have reached an average rate of 241.9 
 as compared with 189.5 for the controls. In compre- 
 hension they have more than held their own, mounting 
 to an average index of 81.2 as compared with 78.0 for 
 the controls. At the end of the second month of 
 training, the average rate for the experimentals is 236.4 
 words per minute as contrasted with 188.2 for the con- 
 trols. The average gain in rate for the two months 
 is 80.7 words per minute for the experimentals as against 
 33.1 for the controls an average superiority in gain 
 of 47.6 for the experimental pupils. 
 
 The story told in Table X is very largely reflected 
 by Table XL At the beginning of the experiment, the 
 two groups displayed practically equal ability in read-
 
 178 
 
 SILENT READING 
 
 
 
 
 sa 
 
 fe 
 O 
 
 1 
 H 
 
 . 
 
 s 
 
 a 
 
 '* 00 C3 tO 00 OS 
 T-< ^i ui CO to IN 
 
 1 1 
 
 W TO 
 
 TO 
 
 1 
 
 rH |> O t** 00 O5 
 
 O IN 
 
 11 
 
 CO M W <N W CO 
 <N 00 **3 ^ O 
 
 1 
 
 
 
 s 
 
 o 
 
 o 
 
 (N OS IN t- (N b- 
 
 t- OS 
 
 to* 
 
 
 CN 00 CM "9 O TO 
 1 1 
 
 0) 
 
 00 IN IN CO TO b- 
 
 S5 
 
 
 O 00 -H O >O -! 
 O 00 TO CO t- 
 
 1 
 
 EH 
 
 pa 
 
 J 
 
 O 00 TO 00 ^ CO 
 
 t~ to 
 
 is 
 
 CO O 
 
 ^ c*i 
 
 7 i 
 
 OS CN lO t- CN O 
 
 00 OS 00 t~ OS OS 
 
 -S 
 
 sa 
 tf 
 
 to r- ^< TO -^ o 
 
 IS 
 
 to t- 
 
 is 
 
 to o in os t^ IN 
 
 (N IN i-l i-H IN IN 
 
 
 
 S 
 
 a 
 
 oo o CM -^ TC <* 
 
 is 
 
 
 N t- "5 00 TO * 
 
 oo oooor-oSob 
 
 1 
 
 O O OS ^ TO Oi 
 
 OS 00 
 TO ** 
 
 
 t OS OS TJI U5 
 <N 00 to 00 OS 
 TO CN 1-1 CN M IN 
 
 M 
 
 EH 
 
 pa 
 
 s 
 
 o 
 O 
 
 I-H IN <N 00 OS * 
 
 OS OS 
 
 Is 
 
 CN (N ' 
 1 1 
 
 O TO (N * IN t- 
 
 Os OS OS 00 OS 00 
 
 a 
 
 O 00 t- i-l TO O 
 
 o to 
 
 oo oc 
 
 00 TO 00 TO t* O 
 "5 tN t^ 00 b- TO 
 (N <N i-l i-l N (N 
 
 < 
 
 o 
 O 
 
 iO to ^ t^ lO 00 
 
 IN q 
 o ^ 
 
 
 t- O OS 00 >O O 
 00 OS t> 00 OS 00 
 
 to 
 
 1 
 
 O! 00 "9 i-H I-H IO 
 
 2^ 
 
 li 
 
 
 CO -H IN OS tO CM 
 IN tN tN i-H CO TO 
 
 1 
 
 H 
 
 pa 
 
 3 
 o 
 O 
 
 "5 TO O i-l TO M 
 
 TO TO 
 
 is 
 
 1-1 00 
 9 
 
 TT 
 
 O t>- 00 O T)I TO 
 OS 00 56 00 t- OS 
 
 <S 
 tt 
 
 00 >O O -^ CO 9 IN 
 
 |s 
 
 00 IN 
 
 TV 
 
 h- TO i-l 9 N TO TO 
 00 -< CM O "9 r tN 
 IN <N rt TH i-l IN tN 
 
 - 
 
 g 
 
 O O *O O t^ CO 
 
 CN >n 
 
 Total Superiority A over B 
 Average " A " B 
 
 "9 00 00 TO IN TO 
 00 t*" t" 00 l* OS 
 
 a 
 tf 
 
 CN iH-00 t^ 00 OS CN 
 
 11 
 
 OS CM 00 00 00 00 Tf 
 OS OS i-H O TO -H IN 
 CN <H i-H i-l 1-1 cN CN 
 
 JO HXBKQVf 
 
 tOININTllrfieviTjl * 
 CNCNtNN-HCNCN >fl 
 
 1 
 
 - B S 
 
 S V 
 > 
 
 t- 
 
 6 
 
 a 
 o 
 
 M 
 
 < CO O 00
 
 STATEMENT OF RESULTS 
 
 179 
 
 ing. This is noticeably so in regard to rate. Among 
 118 pairs of pupils, there are 5 pairs in which the scores 
 of the two pupils are exactly the same, 57 pairs in 
 which the experimental pupil is a trifle faster than his 
 checkmate, and 56 pairs in which the reverse is true. 
 
 TABLE XIII. COMPARISON OF THE NUMBER OF PUPILS IN THE 
 EXPERIMENTAL (A) GROUP WHO SURPASSED THEIR CORRE- 
 SPONDING PUPILS IN THE CONTROL (B) GROUP 
 
 GRADE V 
 
 
 TES 
 
 T I 
 
 TES 
 
 r II 
 
 TEW 
 
 r III 
 
 
 Rate 
 
 Com. 
 
 Rate 
 
 Com. 
 
 Rate 
 
 Com. 
 
 Number of A equal to B 
 
 4 
 
 5 
 
 3 
 
 5 
 
 
 4 
 
 Number of B surpassing A .... 
 
 25 
 
 34 
 
 19 
 
 35 
 
 12 
 
 34 
 
 Number of A surpassing B . ... 
 
 48 
 
 26 
 
 48 
 
 30 
 
 54 
 
 28 
 
 Group Superiority A over B . . . 
 
 23 
 
 
 
 29 
 
 
 
 42 
 
 
 
 Group Superiority B over A ... 
 
 
 
 8 
 
 
 
 5 
 
 
 
 6 
 
 Final Superiority A over B in per cent 
 
 
 
 
 
 63} 
 
 -9} 
 
 NOTE. Table XIII is to be read in the same manner as Table IX. 
 
 At the end of the experiment, the experimental group 
 shows a superiority of 50 per cent in rate, and 4 per 
 cent in comprehension. These percentages of superior- 
 ity of the experimental group over the control group 
 almost coincide with the superiority indicated by 
 Table X. The two methods of control thus serve to 
 corroborate the same general conclusions and mutually 
 to ree'nforce each other. 
 
 3. Fifthr-Grade Pupils 
 
 The effect of the experimental training upon 
 the pupils in the fifth grade is shown in Table
 
 180 SILENT READING 
 
 XII. 1 It includes seven classes comprising a total of 154 
 pupils. Here again it will be noted that both the A and 
 B groups commenced the experiment with equal initial 
 proficiency in rate, 190.7 for the experimental as com- 
 pared with an average of 191.9 words per minute for the 
 controls. The latter have somewhat of an advantage 
 in comprehension 86.3 as against 82.5 for the experi- 
 mentals. 
 
 At the end of the first month the average of the A 
 group had mounted to 265.4 as compared with 225.6 
 for the B group. After another month of training, the 
 experimentals had increased their lead over their 
 checkmates by reaching an average of 277.8 as against 
 222.1 for the controls. The final averages for compre- 
 hension are 85.6 for the A group and 88.6 for the B 
 group. This reflects an average gain of 2.9 as compared 
 with 2.3 for the controls an average superiority in 
 gain for the experimentals of 0.6. More conspicuous 
 is the average superiority of gain in rate of the experi- 
 mentals over the controls, 87.1 as against 30.2 an 
 average superiority of 56.9. 
 
 A comparison of Table XII with Table X shows that 
 the amount of gain in rate made by both the experi- 
 mental and control sections in grade 4 is very similar 
 to the gains made by the experimental and control 
 groups respectively in grade 5. The average gain in 
 
 1 Owing to the failure to test the comprehension of the pupils in the 
 fifth grade in the Plumb School, the number of pairs whose scores in 
 comprehension were recorded is not quite the same as the number of 
 pairs whose rates of speed are compared.
 
 STATEMENT OF RESULTS 181 
 
 rate for the A group in grade 5 is 87.1 words per minute 
 as compared with 80.7 for the corresponding group in 
 grade 4 an average superiority in gain of 6.4 words 
 per minute for the fifth grade. The controls in grade 5 
 scored a gain of 30.8 as against an average gain of 33.1 
 words per minute for the controls in grade 4 a differ- 
 ence of 2.3 in favor of grade 4. On the whole, the aver- 
 age gain in rate of the fifth grade is superior to that of 
 the fourth grade ; so also is the final average rate to 
 which the fifth grade attains 277.8 words per 
 minute as against 236.4 for the experimentals in the 
 fourth grade. 
 
 The final net superiority in rate of the experimentals 
 in grade 5 over their checkmates as determined by the 
 method of individual control, as shown in Table XIII, 
 is 42 pupils. This means that out of 66 pairs of pupils 
 who were closely matched at the beginning of the 
 experiment, 54 experimental pupils surpassed their 
 checkmates in rate of reading at the end of the train- 
 ing, as against 12 control pupils who surpassed their 
 corresponding experimental pupils in rate. This leaves 
 a final superiority of 42 pupils in favor of the A group, 
 or a superiority of 63^ per cent. In comprehension, 
 Table XIII shows a superiority of 6 pupils or 9 per 
 cent in favor of the B group. This latter is discounted, 
 however, by the fact that the B section started with an 
 initial superiority of 8 over the A group. The training 
 received by the experimentals reduced this lead to 6.
 
 182 
 
 SILENT READING 
 
 
 
 j 
 
 O CO 00 1-1 
 CN 00 *' -- 
 1 1 1 
 
 00 iH 
 
 T i 
 
 O CO 
 CN -H 
 
 O 
 
 2 
 
 
 
 * co co co 
 
 IN CO 
 
 O IN 
 
 O 
 
 
 1 
 
 t- N IO >O 
 CN CN t* 1 
 
 2 w 
 
 OS <* 
 SCO 
 
 
 
 1 
 
 H 
 
 
 a 
 
 
 CO -H l>; CO 
 U5 CO ^ O 
 1 1 
 
 CO CO 
 
 Tf 
 
 
 
 
 .2 
 
 d 
 
 tf 
 
 I-H in t~- co 
 
 i-l 00 
 OS 
 
 
 
 
 
 CO t- CO t- 
 
 i-l "5 
 
 O CO 
 
 
 
 g 
 
 O 
 
 O 
 
 1-1 U5 O t- 
 
 Os 00 OS 00 
 
 -* 
 
 CO O 
 
 T 1 
 
 
 
 
 (N 00 t- * 
 
 T-l O 
 
 CO 
 
 
 
 
 j! 
 
 9 
 
 tf 
 
 O * OS O 
 CM IN IN IN 
 
 CN " 
 
 CO N- 
 
 os in 
 
 H 
 
 H 
 
 
 1 
 
 "5 OS t-; <N 
 OS 00 00 00 
 
 388 
 
 
 
 
 S 
 I 
 
 OS <}< O H 
 
 irf co pi I-H 
 co o -H co 
 
 IN CO * IN 
 
 t^ CO 
 
 11 
 
 
 
 
 g 
 
 CO IN co' 00 Os' 
 CO OS 00 00 00 
 
 OS * 
 
 q * 
 
 
 PQ 
 
 a 
 
 i-H i-l Tjl t- i-H 
 
 CN iM IN (N CN 
 
 li 
 
 p co , 
 co 
 
 H 
 
 H 
 
 
 a 
 
 a 
 
 O OJ CO CO IO 
 
 os co co o ci 
 
 OS 00 CO OS OS 
 
 OS 00 
 * OS 
 
 
 
 
 1 
 
 1-. U5 CO ^* I CO 
 
 t^ IN O b- CN 
 tN CN CO CO CN 
 
 N S 
 
 
 
 
 
 O ^ i < CO 
 
 
 O CO 
 
 
 
 J 
 
 IO OS ^ lO 00 
 OS 00 OS OS 00 
 
 
 T 
 i 
 
 
 CQ 
 
 
 CO 00 CO ^ i"* 
 
 ^ Tjl 
 
 O CO 
 
 g 
 
 a 
 
 
 1 
 
 os co os co r- 
 
 00 t- i i CN O 
 i-H i-l CN IN IN 
 
 II 
 
 1? 
 
 1 
 
 H 
 
 
 s 
 
 a 
 
 CO 3] CN CN 00 
 
 OS 00 OS OS 00 
 
 Is 
 
 
 
 
 I 
 
 
 
 IO 00 OS CO t" 
 
 -oo 
 
 Is 
 
 
 1 
 
 TliQ, 
 
 r-i>uv 
 
 i 
 
 QM 
 
 T)< IN IN CN CO 
 CN CN CO IN CN 
 
 00 
 (N 
 
 |s 
 
 
 SCHOOL 
 
 
 s J a"! 
 
 f> "g g o a 
 GQ ^ O O CQ 
 
 
 
 
 ^ 
 
 _o 
 
 'u - 
 
 
 
 U 
 
 
 ffi & 
 
 !Jl 
 
 |I1J SS 
 
 r. 
 
 5 3 
 
 > 
 
 Total Supe 
 Average
 
 STATEMENT OF RESULTS 
 
 183 
 
 4. Sixth-Grade Pupils 
 
 The results of the training upon five sixth-grade 
 classes having a total of 128 pupils are shown in Table 
 XIV. The average rate of 197.8 words per minute for 
 the experimentals mounted to 284.7 after a month's 
 training ; by the end of the training period they had at- 
 tained an average of 292.6 words per minute. The con- 
 trol sections started with an average rate of 204.7 an 
 average initial superiority of 6.6 words per minute 
 over the experimental pupils. After a month of the 
 conventional reading work, the controls averaged 235.4 ; 
 during the second month their rate remained prac- 
 tically stationary as shown by their average of 235.0 at 
 the end of the experiment. The final average gain in 
 rate for the A group is 94.8 words per minute as com- 
 pared with 30.6 for the B group, a superiority in gain 
 of 64.2 for the experimental pupils. 
 
 TABLE XV. COMPARISON OF THE NUMBER OF PUPILS IN THE 
 EXPERIMENTAL (A) GROUP WHO SURPASSED THEIR CORRE- 
 SPONDING PUPILS IN THE CONTROL (B) GROUP 
 
 GRADE IV 
 
 
 Tea 
 
 T I 
 
 TES 
 
 r II 
 
 TESI 
 
 III 
 
 
 Rate 
 
 Com. 
 
 Rate 
 
 Com. 
 
 Rate 
 
 Com. 
 
 Number of A equal to B 
 
 3 
 
 10 
 
 1 
 
 8 
 
 
 
 2 
 
 Number of B surpassing A .... 
 
 38 
 
 28 
 
 23 
 
 33 
 
 8 
 
 24 
 
 Number of A surpassing B . . . . 
 
 23 
 
 26 
 
 40 
 
 23 
 
 44 
 
 26 
 
 Group Superiority A over B . . . 
 
 
 
 
 
 17 
 
 
 
 36 
 
 2 
 
 Group Superiority B over A ... 
 
 15 
 
 2 
 
 
 
 10 
 
 
 
 
 
 Final Superiority A over B in per cent 
 
 
 
 
 
 69 
 
 4* 
 
 NOTE. Table XV ia to be read in the same manner as Table IX.
 
 184 
 
 SILENT READING 
 
 s 
 
 g 
 
 I 2 
 
 3 5 
 
 I 
 
 10 . GRADE VII Total Number of Pupils 206 
 
 g 
 O 
 
 1 
 
 H 
 
 M 
 
 S 
 
 3 
 
 -HOt~to^-*roa>t~o> i-- - 
 
 S3 
 
 NOTE. Table XVI is to be read in the same manner as Table VIII. 
 
 III 1 
 
 1 
 
 NOOOCOOOCO^COWW oc 
 
 1 ^H 00 
 
 SSS^ggS^sfS 
 
 < 
 
 j 
 
 PJtOfHtOM'MTjitO^OO M" 
 
 H 
 
 i 
 
 ci IN IH co d IN -! N' f-i i !2 - 
 
 II 1 1 "" 
 
 -2 
 
 OS 
 
 tf 
 
 t^C^COCOGOCOCOP-JC^CO i^C 
 
 5 
 
 ! 
 i 
 i 
 
 2gSSSS^8 |2 
 
 M 
 
 1 
 
 M 
 
 J 
 
 
 00 
 
 i 
 
 810 oi to co ci IN' ' IN t^ ^J - 
 OiwOCBwOBaDcboo eke 
 
 I 
 
 NCOOb-OOroffltO'-*^ Ot> 
 
 |2 
 
 ?3lNININ(NIN<N<NCO(N ^? 
 
 < 
 
 1 
 
 C^t^i-Hi (Oi>OOOCOC<IO IOC*" 
 
 
 pi i-i uj TJ! pi pi >o o pi to p 
 
 1 
 
 OOOCOT(<COO>^TJ4C<100 HCC 
 
 
 pie^ooco-^pifoPiMpa J^CT 
 
 ' 
 
 
 
 6 
 
 O^iOlfl^OMi-lPlrt* iflifl 
 
 s^ 
 
 pi to' IO CO ^ tf> CO 00 O to' S M 
 OBOBCvOlQBCvwOvGBOi c; "^ 
 
 03 
 
 O^tO-^COinoOOSOM rfN 
 
 Is 
 
 P4pJp4P)P4PJP)C<IPIPJ ^PJ 
 
 < 
 
 J 
 
 OPIXtOCOCOO'HOO JS"- 1 
 
 
 SSSSSSS gS 
 
 5 
 tf 
 
 i-HOOOWCOC^t^iOCOCO 00*O 
 
 
 HiiiiMii li 
 
 H 
 
 . 
 
 g 
 
 
 i r 
 
 C^dCiC^OOOOOClOO g)G^ 
 
 1 
 
 OOOOCOOiO^^COOS O>O 
 
 O 1-4 
 
 S2ila'i|sil 
 
 < 
 
 1 
 
 U5OOOOtOP<OOOt- PJPJ 
 
 Total Superiority A over B . . 
 Average A " B .... 
 
 SS^8^6S IS 
 
 1 
 
 ^HQOOOOiCcO-<^HOCS Q 1 ^ 
 
 ^H^H^H<N<N^^*^C^C^^H ^- j "I 
 
 sawnj jo 
 
 COOOOO^^OOOOOC^ CO 
 
 f 
 
 Illllliijii 
 
 KO-?OOOWKO^ ' 
 
 fc 
 
 
 
 
 
 g = 
 
 a o > V 
 
 is= liii II
 
 STATEMENT OF RESULTS 
 
 185 
 
 In point of comprehension Table XIV shows a slight 
 decrease for both the A and the B groups. The average 
 index of comprehension of the A group at the beginning 
 of the training is 91.0 as against 88.2 at the end. The 
 average for the control pupils drops from 92.6 to 88.5 
 a decrease of 4.1 as against 2.8 for the experimentals. 
 This would seem to indicate that the decrease, slight 
 as it is, cannot be attributed to their acceleration in 
 reading rate. For the control pupils, whose gain in 
 rate is less than one-third the gain of the experimental 
 pupils, decreased about one and a half times as much 
 in comprehension as the experimentals. The decrease 
 in the case of the experimentals is so slight as to be of 
 no special significance. 
 
 TABLE XVII. COMPARISON OF THE NUMBER OF PUPILS IN THE 
 EXPERIMENTAL (A) GROUP WHO SURPASSED THEIR CORRE- 
 SPONDING PUPILS IN THE CONTROL (B) GROUP 
 
 GRADE VII 
 
 
 TEST I 
 
 TEST II 
 
 TEST III 
 
 Rate 
 
 Com. 
 
 Rate 
 
 Com. 
 
 Rate 
 
 Com. 
 
 Number of A equal to B 
 
 5 
 39 
 59 
 20 
 
 12 
 41 
 50 
 9 
 
 1 
 25 
 77 
 52 
 
 18 
 39 
 46 
 
 7 
 
 20 
 83 
 63 
 
 61 
 
 7 
 42 
 54 
 12 
 
 11J 
 
 Number of B surpassing A .... 
 Number of A surpassing B . . . . 
 Group Superiority A over B . . . 
 Group Superiority B over A ... 
 Final Superiority A over B in per cent 
 
 NOTE. Table XVII is to be read in the same manner aa Table IX. 
 
 Table XV tells very much the same story as Table 
 XIV. The conclusions issuing from the two methods 
 of control group and individual serve to reenforce
 
 186 SILENT READING 
 
 each other. The outstanding facts shown by Table 
 XV are these : From a total of 64 pairs of pupils, 38 
 controls surpassed their corresponding experimentals, 
 as against 23 experimentals who surpassed their check- 
 mates. At the end of the experiment, from a total 
 of 52 pairs, 44 experimentals excelled their check- 
 mates, as contrasted with the eight controls who sur- 
 passed their experimentals. Reduced to a percentage 
 basis, the A group surpassed the B group by 69 per 
 cent in rate and 4^ per cent in comprehension. 
 
 5. Seventh-Grade Pupils 
 
 The records of the seventh-grade classes totaling 
 206 pupils are presented in Table XVI. Both groups 
 commence with approximately equal averages in rate 
 and comprehension. In the latter respect, the two 
 averages differ by but 0.1. A month's training in 
 rapid reading brings the average of the experimentals 
 from 205.6 per minute to 298.5, whence it rises to 321.6 
 by the end of the second month. The experimentals 
 far outstrip their checkmates whose rate during the 
 same two months of conventional work in reading 
 reaches only 249.7 words per minute. The average 
 gain of the A group is 116.0 as compared with 47.2 for 
 the B group an average superiority in gain of 68.8. 
 In comprehension the experimentals likewise surpass 
 their checkmates. Their average rises from 91.2 to 
 92.3 as compared with the slight increase from 91.3 
 to 91.4 by the control pupils. The superiority of gain
 
 STATEMENT OF RESULTS 
 
 187 
 
 2 
 
 O 
 
 1 
 
 H 
 
 a 
 
 1 
 
 O ^5 ^ 
 00 *4 09 
 
 I I I 
 
 S * 
 1 1 
 
 S2 
 
 2 
 
 oo d co 
 
 3 6 
 
 li 
 
 
 
 S 
 
 a 
 
 IO N .H 
 
 O < ' *- 
 
 1 1 1 
 
 co q 
 
 
 1 
 
 O O CO 
 
 *" '? d 
 
 p4 OT QB 
 
 Is 
 
 
 I 
 H 
 
 . 
 
 o 
 O 
 
 00 CO O 
 OS OS Ci 
 
 Si 
 
 co t- 
 co 
 
 I 
 
 CO 00 CO 
 
 L~ f "J 
 
 !N CO CO 
 
 N 00 
 
 5S 
 
 < 
 
 a 
 
 o 
 O 
 
 ^_ OS "5 
 
 co 31' co 
 
 ~ _ . C~- 
 
 CO 00 
 
 
 1 
 
 eit~n 
 
 SOO 00 
 t~ ^< 
 
 co co *< 
 
 I- O 
 
 ll 
 
 
 1 
 
 PQ 
 
 1 
 
 <S CO OS 
 1- -^ 
 QB Oi w 
 
 00 CO 
 
 ~ ^) 
 
 T 1 
 
 2 
 pi 
 
 N CO U5 
 00 t^ t> 
 
 ^ -^ o 
 
 N CO CO 
 
 S 
 co jj 
 
 s^ 
 
 - 
 
 S 
 
 a 
 
 OS 00 CO 
 ~ ~ ~ 
 
 OS CO 
 
 
 1 
 
 at no 
 
 1-4 O -< 
 
 li 
 
 
 i 
 
 EH 
 
 PQ 
 
 S 
 
 a 
 
 iii|i?7 
 
 S 
 
 00 <-4 CO 
 
 5 ^' Os' 
 
 2SS 
 
 la 
 
 00 
 
 5 
 
 - 
 
 1 
 
 OS <N CO 
 
 " * *- 
 
 CD 00 
 
 is 
 
 Total Superiority A over B 
 
 1 
 
 CO 00 O 
 
 si ** *~ 
 
 Sao 
 
 gj 
 
 stwnj do 
 
 ??Sg 
 
 S 
 
 1 
 
 Hendricks 
 Gar field 
 
 C.rccli'V 
 
 5 
 
 e > 
 
 
 
 Chicago 
 Btreator
 
 188 
 
 SILENT READING 
 
 in comprehension by the A group is not, however, 
 very appreciable, being but 1.0. 
 
 On the basis of the method of individual control, 
 Table XVII shows a superiority of 61 per cent in rate 
 and 11 per cent in comprehension in favor of the 
 experimental group over the control pupils. 
 
 6. Eighth-Grade Pupils 
 
 From the fourth grade on, the tables have shown an 
 increase in the amount of gain by the experimentals 
 as the grade advanced. This is especially true of the 
 eighth grade, where the amount of gain is greater than 
 that of any of the lower grades. From an average 
 rate of 220.8, the A group rises to 361.2 after one 
 month's training, and to 393.0 at the end of the second 
 month. This is the highest average rate of speed 
 which any of the experimental groups attained on the 
 
 TABLE XIX. COMPARISON OF THE NUMBER OF PUPILS IN THE 
 EXPERIMENTAL (A) GROUP WHO SURPASSED THEIR CORRE- 
 SPONDING PUPILS IN THE CONTROL (B) GROUP 
 
 GRADE VIII 
 
 
 TES 
 
 T I 
 
 TBS 
 
 r II 
 
 TEST 
 
 p III 
 
 
 Rate 
 
 Com. 
 
 Rate 
 
 Com. 
 
 Rate 
 
 Com. 
 
 Number of A equal to B 
 
 6 
 
 10 
 
 1 
 
 6 
 
 
 4 
 
 Number of B surpassing A .... 
 
 9 
 
 16 
 
 11 
 
 25 
 
 13 
 
 16 
 
 Number of A surpassing B . . . . 
 
 31 
 
 20 
 
 34 
 
 15 
 
 33 
 
 26 
 
 Group superority A over B . . . . 
 
 22 
 
 4 
 
 23 
 
 
 
 20 
 
 10 
 
 Group superiority B over A ... 
 
 
 
 
 
 
 
 10 
 
 
 
 
 
 Final superiority A over B in per cent 
 
 
 
 
 
 43$ 
 
 22 
 
 NOTE. Table XIX is to be read in the same manner as Table IX.
 
 STATEMENT OF RESULTS 
 
 189 
 
 TABLE XX. TABLE SHOWING AVERAGE RATE OF READING FOR EX- 
 PERIMENTAL (A) AND " CONTROL " (B) PUPILS AT BEGINNING AND 
 END OF TRAINING PERIOD AS DETERMINED BY COURTIS SILENT 
 READING TEST FOR GRADES IV-VIII 
 
 
 NUM- 
 
 TEST I 
 
 TEST II 
 
 TEST III 
 
 GAIN 
 
 A's 
 
 
 
 
 
 
 
 Crrnn 
 
 
 PUPILS 
 
 A 
 
 B 
 
 A 
 
 B 
 
 A 
 
 B 
 
 A 
 
 B 
 
 IN GAIN 
 
 IV . 
 
 236 
 
 155.7 
 
 155.1 
 
 241.9 
 
 189.5 
 
 236.4 
 
 188.2 
 
 80.7 
 
 33.1 
 
 47.6 
 
 31% 
 
 V . . 
 
 154 
 
 190.7 
 
 191.9 
 
 265.4 
 
 225.6 
 
 277.8 
 
 222.1 
 
 87.1 
 
 30.2 
 
 56.9 
 
 30% 
 
 VI . 
 
 128 
 
 197.8 
 
 204.4 
 
 284.7 
 
 235.4 
 
 292.6 
 
 235. 
 
 94.8 
 
 30.6 
 
 64.2 
 
 33% 
 
 VII . 
 
 206 
 
 205.6 
 
 202.5 
 
 298.5 
 
 237.21 
 
 321.6 
 
 249.7 
 
 116.0 
 
 47.2 
 
 68.8 
 
 33% 
 
 VIII . 
 
 92 
 
 220.8 
 
 211.7 
 
 361.2 
 
 290.5 
 
 393.0 
 
 301.8 
 
 172.2 
 
 90.1 
 
 82.1 
 
 35% 
 
 
 
 
 
 
 
 
 
 
 
 
 
 Average gain of A for all grades = 56%; ofB = 25%; A's superiority 
 over B=31%. 
 
 TABLE XXI. TABLE SHOWING AVERAGE COMPREHENSION IN READING 
 FOR EXPERIMENTAL (A) AND "CONTROL" (B) PUPILS AT BEGIN- 
 NING* MIDDLE, AND END OF TRAINING PERIOD AS DETERMINED 
 BY COURTIS SILENT READING TEST 
 
 GRADE 
 
 Nmi- 
 
 TEST I 
 
 TEST II 
 
 TEST III 
 
 GAIN 
 
 A's 
 
 
 BER OF 
 
 PUPILS 
 
 A 
 
 B 
 
 A 
 
 B 
 
 A 
 
 B 
 
 A 
 
 B 
 
 SUPERIORITY 
 IN GAIN 
 
 IV. . 
 
 236 
 
 77.4 
 
 78.6 
 
 81.2 
 
 78.0 
 
 81.6 
 
 79.1 
 
 4.2 
 
 .5 
 
 3.7 
 
 v . . 
 
 154 
 
 82.5 
 
 86.3 
 
 87.0 
 
 89.9 
 
 85.6 
 
 88.6 
 
 2.9 
 
 2.3 
 
 0.6 
 
 VI . . 
 
 128 
 
 91.0 
 
 92.6 
 
 90.8 
 
 90.4 
 
 88.2 
 
 88.5 
 
 -2.8 
 
 -4.1 
 
 1.3 
 
 VII . 
 
 206 
 
 91.2 
 
 91.3 
 
 94.1 
 
 93.5 
 
 92.3 
 
 91.4 
 
 1.1 
 
 0.1 
 
 1.0 
 
 VIII . 
 
 92 
 
 95.8 
 
 96.5 
 
 94.3 
 
 96.3 
 
 94.8 
 
 94.1 
 
 -1.0 
 
 -2.4 
 
 1.4 
 
 Average gain of A for all grades = .09%; ofB=-0.7%; A's superi- 
 ority over B = 1.6%. 
 
 Courtis Test. The controls likewise achieved the con- 
 siderable gain of 90.1 words mounting from 211.7 
 to 301.8. It is, however, just about half the gain made 
 by the experimentals 90.1 as against 172.2 for the 
 experimental group. The averages for comprehension
 
 190 SILENT READING 
 
 have varied but little in the case of both groups a 
 decrease of 2.4 for the controls as against a decrease of 
 1.0 for the experimentals. 
 
 SUMMARY OF EFFECTS OF THE TRAINING UPON ALL 
 
 GRADES 
 
 The results of the preceding tables in regard to the 
 average rates of the different grades are summed up in 
 Table XX. The advancement in rate of reading as 
 shown in the three tests is fairly uniform as the grade 
 increases. From an average in the first Courtis Test 
 of 155.7 words per minute for the experimentals in the 
 fourth grade, the average continues to advance until 
 in the eighth grade it reaches 220.8. At the end of the 
 training, the average for the A groups rises from 236.4 
 words per minute in the fourth grade until it reaches 
 393.0 in the eighth grade. Reducing the average gains 
 in number of words read per minute to a percentage 
 basis, it becomes possible to express the amount of 
 improvement for the experimental pupils in all the 
 grades in a single term 56 per cent. The average 
 gain for the control pupils in all the grades is 25 per cent. 
 This shows a final average superiority in gain for the 
 experimental pupils over their checkmates of 31 per 
 cent. In terms of the number of words read per 
 minute the average gain of the A group is 110.2 as 
 against 46.2 for the B group an average superiority 
 in gain of 64 words per minute in favor of the experi- 
 mental or A group.
 
 STATEMENT OF RESULTS 191 
 
 A summary of the scores of the different grades in 
 comprehension is presented in Table XXI. The gains 
 or losses for neither group are especially significant. 
 The only gains which appear at all appreciable occur 
 in the experimental sections in the fourth and fifth 
 grades gains of 4.2 and 2.9 respectively. The final 
 average gain of the experimental groups in all the grades 
 is 0.9 per cent as against an average loss of 0.7 per cent 
 for the control pupils. This gives the experimental 
 groups an average superiority in gain in comprehension 
 of 1.6 per cent over the control groups. The measure 
 of comprehension here used is the index computed ac- 
 cording to the directions of Courtis. The results of 
 the application of a somewhat different measure of the 
 efficiency of comprehension, namely, the number of 
 questions correctly answered in five minutes, will be 
 presented later. 
 
 A summary of the results of the experimental train- 
 ing upon the rate and comprehension in all the grades 
 as shown by the method of individual control is pre- 
 sented in Tables XXII and XXIII. Table XXII 
 shows that at the beginning of the experiment there 
 were 401 pairs of pupils, one of each pair being in the 
 experimental and the other in the control group. The 
 pupils constituting each pair were approximately evenly 
 matched, though at times the variation between the 
 pupils in a pair was not inappreciable. The amount 
 of variation to be found in the rates of any class of 
 pupils is such as to render very small the number of
 
 192 SILENT READING 
 
 pupils who score exactly the same rate. Of the 
 401 pairs in the present inyestigation there were 23 
 experimentals whose scores were identical with those of 
 their checkmates ; while 165 controls were superior 
 to their corresponding experimentals as against 213 
 experimentals who were above their controls. In both 
 the latter cases the differences as a rule were slight, as 
 was evidenced by the close similarities of their averages. 
 
 At the end of the training the results were : one pair 
 identically equal ; 86 controls superior to their corre- 
 spondents in the experimental group as contrasted with 
 314 experimentals who were superior to their controls 
 a final net superiority of 228 pairs for the experimentals. 
 Subtracting from this total the amount of A's initial 
 superiority of 48 ; the results show a final superiority 
 in gain of the A group over the B group of 180 pupils, 
 or 45 per cent. 
 
 Table XXIII shows that the B group starts the ex- 
 periment with somewhat of a superiority over the A 
 group in comprehension. Of a total of 389 l pairs, 
 41 pairs have the same scores for each of the members 
 of the pair, while 180 controls surpassed their experi- 
 mentals as against 168 experimentals who surpassed 
 their checkmates, an initial superiority of 12 in 
 favor of the B group. At the end of the training there 
 were 20 pairs whose members have equal scores, while 
 180 controls surpassed their experimentals as against 
 
 1 Reduced from 401 pairs because, as indicated, 12 pairs in the fifth 
 grade were not rated in comprehension.
 
 STATEMENT OF RESULTS 
 
 193 
 
 oo 
 So 
 
 as 
 
 <GD 
 
 B^S 
 
 x5g 
 
 XQ> 
 
 -H ^H * CO 
 
 et 1 
 
 "5 1 
 
 o | 
 M 1 
 
 Number of A equal to B . 
 Number of B surpassing A 
 Number of A surpassing B 
 Group Superiority A over B 
 Group Superiority B over A 
 
 
 *f cj 
 
 th ** 
 
 2 1 2 
 
 co -; oo 1 w 
 
 5 
 
 S I 1 - 
 
 B 
 A
 
 194 SILENT READING 
 
 201 experimental who were superior to their controls 
 a superiority of 21 in favor of the experimentals. This 
 number is to be augmented by 12, the handicap under 
 which the A group started, giving a final superiority 
 in gain in comprehension of 32 pairs or 8 per cent for the 
 experimental pupils over their control pupils. 
 
 The results of the method of individual control serve 
 as a powerful reenforcement of a refined type of the con- 
 clusions issuing from a comparison of the average 
 scores of the two groups. This method of control 
 shows furthermore that the superiority of the experi- 
 mental group over the control group is not due merely 
 to a very marked superiority of a rather small number 
 of pupils, but that the superiority is spread very largely 
 throughout the whole group.
 
 CHAPTER IX 
 INTERPRETATION OF RESULTS 
 
 THE EFFECT OF THE TRAINING UPON READING RATE 
 
 GRAPH I illustrates the data contained in Table XX. 
 It presents very clearly the differences in amount of 
 improvement in reading rate achieved by the experi- 
 mental pupils and the controls. While in every grade 
 the two groups start at practically the same level of 
 reading rate, yet in every grade the experimental pupils 
 far outstrip the controls. The bulk of improvement, it 
 will be noticed, is effected in the first month of the training. 
 With the exception, however, of the fourth grade, in 
 which there occurs a very slight decrease, improvement 
 of a lesser character continues during the second 
 month. 
 
 The decided improvement made by the pupils as a re- 
 sult of the first month of the training would seem to indi- 
 cate that the customary reading rate of pupils was far 
 below the rate at which they could read intelligently 
 and efficiently. In other words, they were not working 
 on the maximum plane of efficiency. Thus the pupils 
 in the seventh grade are able to increase their rate 116 
 words per minute, while the eighth-grade pupils succeed 
 
 195
 
 196 
 
 SILENT READING 
 
 in almost doubling their rate without in either case 
 experiencing any appreciable decrease in comprehen- 
 sion. In fact, in one case, a slight improvement is 
 
 No. T. 
 
 dtferafe JPa/e oj JReadi'tji j/br fxperjoteata/fSV and Confro/ f) /ty>/& / 
 of Traitf/rtf Pei~/o</ A-S flefef*n//?f</ fy (sovrf/j *Jt/e/> 
 
 JfO 
 
 J76 
 J60 
 3fO 
 
 fro Slrt>,jo 
 
 I > 
 
 $** 
 
 ^ 
 
 160 
 r 
 
 I' 00 
 
 ^30 
 
 ' 00 
 
 70 
 60 
 JO 
 
 fa 
 
 JO 
 
 to 
 
 10 
 
 VI 
 
 VII 
 
 VIII
 
 INTERPRETATION OF RESULTS 197 
 
 effected. This result gives some point to the query : 
 Is not the reading of most individuals done at slow, 
 plodding rates which are far below the levels attainable 
 by a little training ? The results portrayed in Graph I 
 point strongly in that direction. 
 
 The Effect upon Comprehension 
 
 That the comprehension has not been greatly af- 
 fected either favorably or adversely is shown by Graph 
 II. The comprehension remained constant to a large 
 extent in both the experimental and control groups. 
 The slight superiority in gain that does exist, however, 
 is in each grade in favor of the experimental pupils. 
 The conclusions that would seem to follow from the 
 performances of the pupils in comprehension in this 
 study are: 
 
 1. Marked increases in speed of reading may be 
 effected without any impairment of the comprehen- 
 sion. 
 
 2. The setting up of habits of rapid reading does 
 not per se increase the accuracy of the comprehen- 
 sion. 
 
 3. To secure marked improvement hi accuracy 
 of comprehension special stress must be placed upon 
 training designed specifically to increase the accuracy 
 of the comprehension. 
 
 While the experimental training outlined in this study 
 succeeded in safeguarding and even slightly improving 
 the accuracy of the comprehension, yet its predominant
 
 198 
 
 SILENT READING 
 
 effect was the marked acceleration of the reading rate. 
 To produce such an effect upon the rate was precisely 
 the end for which the training was devised. 
 
 No 
 
 Jo 
 e/H/ ae/ of Trairnqf Kr/'tx/ gj c/f/ermt'/Jfc/ fy foan'/s <J//?/rS iPeat//^ Tost. 
 
 /oo 
 X 
 
 *? 6 
 
 Grade 
 
 The Relative Amount of Gain Made by Different Grades 
 
 A comparison of the amount of gain in rate made by 
 the experimental and control pupils in each of the 
 grades is given in Graph III. The graph brings out 
 clearly the fact that the amount of gain increases as the 
 grade advances. In every case in the experimental 
 groups the gain made by the higher grade is superior to 
 that made by the lower. With the single exception of 
 the sixth grade, this is true likewise of the control groups.
 
 INTERPRETATION OF RESULTS 
 
 199 
 
 This superiority in gain in rate by the upper grades 
 over the lower is quite the opposite of what has usually 
 been reported concerning the relative gains made by 
 the different grades. The third and fourth grades 
 have been of late generally regarded as the crucial 
 school periods during which the appreciable increases 
 in rate of reading are to be effected. The results re- 
 
 to V/// aj 
 
 A/O m 
 
 Gain in fir/e a/ & 
 ' 
 
 fo 
 
 ported by W. S. Gray, Judd, Courtis, and Waldo have 
 shown that de facto the important gains are made in 
 the lower grades, especially in the third and fourth. 
 What is the explanation of the difference of the relative 
 amount of gain made by the different grades as re- 
 ported by other investigators and those reported in the 
 present study ? In the writer's judgment the explana- 
 tion is not far to seek.
 
 200 SILENT READING 
 
 The gains reported by other investigators were the 
 results of the conventional type of training in reading 
 with almost the entire emphasis upon the oral phase. 
 The matter of rapidity in silent reading has not as yet 
 become a problem in the minds of the majority of upper 
 grade teachers, much less in the minds of their pupils. 
 During the third and fourth grades, great stress is 
 placed by the teachers upon the mastery of the mechan- 
 ics of reading. Generous portions of time are devoted 
 to this task. After this has been accomplished, the 
 pupils are allowed to slumber along at the slow, dead- 
 level plodding rates which became fixed for them in 
 the days of sempiternal drill in oral reading. Reading 
 from the fourth grade on continues largely as 
 an exhibition of word pronouncing, articulation, etc. 
 As Judd aptly phrases it, "The conscientious teacher 
 supplied with a reading book and a period in the 
 program carries on the well-known reading farce in the 
 vain hope that the effects of unsuccessful teaching will 
 be overcome by a liberal application of the same 
 methods that produced the difficulties." 
 
 The value of an acceleration of the rate in silent 
 reading never seems to dawn upon them. It is no 
 wonder that the rate does not notably rise as the grade 
 advances. The wonder, rather, is that the rate does not 
 decrease as a result of the grotesquely misplaced em- 
 phasis on oral reading and the school's utter neglect 
 of reading in the true sense of the term the silent 
 interpretation of the printed symbols.
 
 INTERPRETATION OF RESULTS 201 
 
 The increases in reading rate in the different grades 
 as reported by K. D. Waldo are shown in Table XXIV. 
 The increases effected in the upper grades as a result 
 of almost a year's teaching of reading are so meager 
 as to be scarcely perceptible 2.1 and 11.7 words 
 in the sixth and eighth grades respectively. The rate 
 is practically at a standstill from the fourth grade on. 
 A condition such as this is, in the writer's judgment, 
 nothing less than pathological. It is a serious indict- 
 ment of the present school regime in the teaching of 
 reading. 
 
 In the present investigation rapidity in silent read- 
 ing was made a problem. It was recognized as a value 
 to be striven for and to be achieved through training. 
 Graph III shows the results of the pupils' effort. They 
 secured that for which they strove. 
 
 From these results three conclusions would seem to 
 follow : 
 
 1. Marked increases in rate can be effected in the 
 upper grades when speed in reading is made a problem 
 in the minds of the pupils. 
 
 2. When training in rapid silent reading is given to 
 pupils not previously trained therein, the acceleration 
 in rate tends in a general way to advance pari passu 
 with the advance in the grade the higher the grade, 
 the greater being the increase in rate. 
 
 3. The present average rates in the upper grades 
 are unnecessarily slow, due to the absence on the part 
 of the school of any organized effort to accelerate them.
 
 202 
 
 SILENT READING 
 
 The data concerning the gains in comprehension 
 contained in Table XXI are shown in Graph IV. It is 
 
 fo VflJ as </e/ermtnf</ 
 
 faorf/s 
 
 AlO-T3 
 
 Co if re/ 
 
 4 
 
 3 
 Z 
 / 
 
 c 
 -/ 
 
 -Z 
 
 -3 
 
 -f 
 
 IV 
 
 y// 
 
 y/// 
 
 to be noted that no grade in either the experimental 
 or the control group varies more than about four points 
 from its initial average. 
 
 TABLE XXIV. THE INCREASE IN READING RATE IN THE DIFFERENT 
 GRADES AS REPORTED BY K. D. WALDO 
 
 
 
 
 INCREASE IN 
 
 GBADB 
 
 FALL RATE 
 
 SPRING RATE 
 
 NUMBER OP 
 WORDS 
 
 PER CENT 
 
 3 
 
 76.4 
 
 149.1 
 
 72.7 
 
 95.2 
 
 4 
 
 92.7 
 
 163.3 
 
 70.6 
 
 76.1 
 
 5 
 
 113.0 
 
 129.2 
 
 16.2 
 
 14.3 
 
 6 
 
 128.0 
 
 130.1 
 
 2.1 
 
 1.2 
 
 7 
 
 122.7 
 
 142.8 
 
 20.1 
 
 16.4 
 
 8 
 
 147.2 
 
 158.9 
 
 11.7 
 
 8.0
 
 INTERPRETATION OF RESULTS 
 
 203 
 
 TABLE XXV. COMPARISON OF AVERAGES IN RATE ATTAINED BY 
 PUPILS AFTER TRAINING WITH PRESENT NORMS 
 
 GRADE 
 
 IV 
 
 V 
 
 VI 
 
 VII 
 
 vin 
 
 O'Brien .... 
 
 236 
 
 278 
 
 293 
 
 322 
 
 393 
 
 Oberholtzer . . . 
 
 156 
 
 186 
 
 234 
 
 282 
 
 288 
 
 Courtis .... 
 
 160 
 
 180 
 
 220 
 
 250 
 
 280 
 
 Gary 
 
 140 
 
 166 
 
 185 
 
 198 
 
 204 
 
 Starch .... 
 
 144 
 
 168 
 
 192 
 
 216 
 
 240 
 
 Brown .... 
 
 213 
 
 269 
 
 272 
 
 279 
 
 290 
 
 Gray 
 
 180 
 
 204 
 
 216 
 
 228 
 
 234 
 
 TABLE XXVI. TENTATIVE NORMS FOR PUPILS TRAINED IN RAPID 
 SILENT READING 
 
 GRADE 
 
 ACTUAL AVERAGE 
 
 P. E. 
 
 SUGGESTED 
 NORMS 
 
 IV 
 
 236.4 
 
 12.13 
 
 236 
 
 V 
 
 277.8 
 
 8.37 
 
 266 
 
 VI 
 
 292.6 
 
 8.71 
 
 296 
 
 VII 
 
 321.6 
 
 7.01 
 
 326 
 
 VIII 
 
 393.0 
 
 12.77 
 
 356 
 
 
 
 
 
 Average Rates Attained by Pupils After Training 
 Compared with Present Norms 
 
 A comparison of the averages of the experimental 
 pupils in grades 4 to 8 with the norms reported by 
 Courtis, Brown, Gray, Starch, and Oberholtzer is pre- 
 sented in Table XXV. There is a common basis of com- 
 parison between the averages of the experimental pupils 
 and the norms reported by Courtis. Both were achieved 
 on the Courtis Test and the directions to the pupils
 
 A/alT 
 
 Graph <5hou>/r>c /Ivera^e Pate for Pupi/j 
 after' Train/nd in Rapid Rea.<J/'n4 a.3 
 compared u>/fff //arms reported 
 Court/5 for 
 
 f/ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 300 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 J70 
 3U> 
 350, 
 
 320 
 310 
 300 
 290 
 260 
 270 
 2(6 
 .250 
 ^210 
 
 ^ zto 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 fit 
 
 'ft 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 L 
 
 i/f 
 
 fa 
 
 c _ 
 
 
 _ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 . 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 /I 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 x^ 
 
 ^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^J< 
 
 ^ 
 
 ^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 _^- 
 
 
 
 ^-^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 >^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 - 
 
 ^ 
 
 
 
 
 ^ 
 
 x 1 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^. 
 
 X" 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 x 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ' 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 . 
 
 -^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 - 
 
 .** 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 x'' 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 X 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 190 
 V. ISO 
 
 
 
 
 
 
 
 
 
 ^' 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 *" 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 _,. 
 
 ^ 
 
 " 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ISO 
 
 ^ uo 
 \ loo 
 
 30 
 SO 
 70 
 60 
 JO 
 40 
 JO 
 
 JO 
 
 a 
 t 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 v- v vr vii vti 
 
 Q ra. </e*s. 
 204
 
 INTERPRETATION OF RESULTS 205 
 
 in both cases were identical. The norms reported by 
 Courtis represent the smoothed averages, the figures 
 presented by the writer are the actual averages. A 
 comparison of these two norms is shown in Graph V. 
 It brings out strikingly the difference in superiority 
 in the reading rate of pupils who have received training 
 in rapid effective silent reading and pupils who have 
 been nurtured on the conventional pabulum of drill 
 in oral reading. The superiority of the experimental 
 pupils in every grade is very marked. Not less notice- 
 able is the superiority over the norms reported by W. S. 
 Gray and by Oberholtzer, as shown in Graph VI. 
 The norms reported by Gray for the three different 
 selections in his reading test have been adjusted here 
 to the basis of the easiest selection, "Tiny Tad." 
 
 The highest norms reported are those by Brown. 
 They represent, however, not the averages of all the 
 pupils tested in the different grades, as in the case of 
 the other investigators, but the highest averages reached 
 by various single classes tested by Brown. Conse- 
 quently they are offered as norms or standards to be 
 striven after rather than indices of the present average 
 attainment of the different grades. These standards 
 mentioned by Brown come closer to the averages ac- 
 tually reached by the experimental classes than those 
 of any of the other investigators. They are, how- 
 ever, very considerably below them, as shown in 
 Graph VII. 
 
 The present norms for reading rate have all been
 
 ap/d 
 
 gate for Pupi/s a /far traia/'/rd 
 as compared 
 
 -<? 
 
 JffO 
 
 360 
 JSO 
 JfO 
 JJO 
 
 Jio 
 
 J/O 
 JOO 
 290 
 280 
 270 
 
 s zso 
 
 \2XO 
 Z/O 
 
 5 S&O 
 
 a /4o 
 /zo 
 
 t/0 
 
 /ao 
 eo 
 
 70 
 (0 
 JO 
 fo 
 JO 
 
 /o 
 
 
 
 1 
 
 
 
 
 J 
 
 
 
 
 / 
 
 Lrarnd ; 
 
 
 
 / 
 
 n'Br/en 
 
 
 
 / 
 
 Obfrho/1 Zt 
 
 >r - 
 
 
 / 
 
 ftra./j 
 
 , 
 
 
 / 
 
 
 
 
 / 
 
 
 
 
 / 
 
 
 
 ^s^ 
 
 
 
 
 ^s^ 
 
 
 
 I 
 
 "_ 
 
 
 
 _^^- 
 
 A 
 
 J ' 
 
 s 
 
 
 X 
 
 
 >x 
 
 
 X 
 
 
 "7 
 
 
 X 
 
 
 z 
 
 
 X 
 
 
 
 ^ 
 
 if 
 
 ( 
 
 
 "2 
 
 .^ "' 
 
 
 
 x ^< 
 
 
 
 
 F ***"^ J !ix J " 
 
 
 
 .^ 
 
 X 
 
 
 
 ^^' x j 
 
 (W 
 
 
 
 s'' 
 
 
 
 
 ^ 
 
 
 
 
 1^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 y v yt y// yi 
 Q r a. de 3. 
 
 206
 
 A]o .EX 
 
 Average afe for Pupi /$ a//fr 
 'Training 10 &af>/</ JPeadr/^ as co*yxine</ 
 <f JfarcA /or 
 
 400 
 
 jao 
 
 JtO 
 
 sso 
 
 340 
 
 3tO 
 3/0 
 3OO 
 
 *ro 
 *se 
 270 
 
 ^ 2/0 
 
 
 
 
 
 j 
 
 I f/i d ' 
 
 
 
 / 
 
 
 
 
 / 
 
 firouin 
 
 r _ 
 
 
 / 
 
 Star c A 
 
 __ 
 
 
 / 
 
 
 
 
 / 
 
 
 
 
 / 
 
 
 
 
 / 
 
 
 
 .^ 
 
 
 
 
 ^x^^ 
 
 
 
 
 ^ 
 
 
 
 ^ 
 
 
 "" 
 
 > 
 
 ~-~*^ 
 -^ 
 
 -. ' ' 
 
 
 / / 
 
 
 
 
 .X r \ x x ' 
 
 
 
 
 X^ x 
 
 
 
 
 r X 
 
 
 
 ^^ 
 
 x 
 
 
 
 ^,- 
 
 f 
 
 
 ^ 
 
 *^ 
 
 
 
 ^*'~ 
 
 
 V 19 
 \I60 
 
 iro 
 
 \ ISO 
 f/0 
 
 /eo 
 
 60 
 7* 
 
 60 
 
 
 
 ^' 
 
 
 
 ^,^^ 
 
 
 
 
 ^..' 
 
 
 
 .* 
 
 r"^ 
 
 
 
 ^'~ 
 
 
 
 
 r*~ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 JO 
 10 
 
 JO 
 
 /o 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 Q r a. </ 
 
 207 
 
 K//
 
 208 SILENT READING 
 
 derived from the performances of pupils who have been 
 trained in the conventional type of oral reading. In 
 the vast majority of cases they have received no train- 
 ing in rapid silent reading. What the norms will be 
 after the schools begin to train in rapid silent reading 
 is an interesting question which the future alone can 
 answer. As the result of a pioneer effort in a virgin field, 
 the smoothed averages attained by the experimental 
 pupils in the different grades in the present investigation 
 are suggested as tentative norms. The degree of reliabil- 
 ity of the averages for the different grades has been com- 
 puted in terms of the P. E. 1 which are presented in Table 
 XXVI. It is noted that the P. E. is relatively small, 
 indicating a rather high degree of reliability for the 
 averages. 
 
 Variability in the Amount of Gain in Rate 
 
 To show the range or variability in amount of gain 
 in rate of the experimental pupils compared with the 
 control pupils, Graph VIII is offered. A hundred pairs 
 
 1 P. E. refers to the " probable error." As Whipple observes, " The 
 term probable error is often a source of confusion to those unfamiliar 
 with its use in mathematics. For a descriptive term, we might call it 
 the ' median deviation,' since it is that deviation that is found midway 
 from the representative value in either direction. The magnitude in 
 question is not, of course, the most probable error, neither is it, in a cer- 
 tain sense, a ' mistake.' It is rather a ' probable sampling error ' : we are 
 unable to measure every possible instance of the thing we are studying, 
 but must content ourselves with a restricted number of samples, usually 
 so taken as to be ' random samples.' The P. E. serves to indicate the re- 
 liability of these random samples, the degree to which they probably de- 
 part from the true universal values."
 
 INTERPRETATION OF RESULTS 
 
 209 
 
 of pupils, each pair consisting of one experimental and 
 one control pupil, were selected at random from a much 
 larger list containing the pairs of pupils whose initial 
 scores in rate were approximately equal. To facilitate 
 comparison, the gain of each pupil was expressed as a 
 per cent of his initial score. Graph VIII shows that 
 the gains of the experimental pupils are grouped much 
 more closely to the ordinate than the gains of the ex- 
 perimentals. The larger gains are made almost en- 
 tirely by the latter group. 
 
 Al ' o 
 
 
 Jf4rfoJ a//t/- 7r*i*/'#j 
 Ay /00 Co/ifo/ AP//J - as 
 
 cotr^trft/ u/'M Gt/ 
 
 
 V 
 
 v tO 
 
 \ 
 
 / 
 
 /so
 
 210 SILENT READING 
 
 Measurement of the Comprehension by the Number of 
 Questions Correctly Answered 
 
 As mentioned previously, the Courtis Index of Com- 
 prehension is a measure merely of accuracy. It does 
 not reflect in any way the number of questions 
 answered. Now obviously efficiency in comprehension 
 consists not only in accuracy but in the number of 
 actions correctly performed. To measure this latter 
 phase of comprehension, which entirely escapes the 
 Courtis Index, the number of questions correctly 
 answered was employed as a supplementary measuring 
 device. Indeed, unpublished data at the Bureau 
 of Educational Research at the University of Illinois 
 show that a higher correlation exists between the 
 scores in comprehension as determined by other stand- 
 ard reading tests and the number of questions correctly 
 answered, than between the comprehension scores in 
 other tests and Courtis' Index of Comprehension. 
 
 The increase in the number of correct answers re- 
 flects furthermore the increase in the rate of reading of 
 a passage in which the comprehension is required and 
 on which it is tested. This device makes it possible, 
 moreover, to determine whether or not the increased 
 speed of reading effected by the training persisted in 
 different situations. For the mental attitude assumed 
 in reading a passage to answer questions on it imme- 
 diately is considerably different from that assumed 
 in reading a passage rapidly, simply "to get the gist"
 
 INTERPRETATION OF RESULTS 211 
 
 of it. It is thought that the employment of this device 
 in the present study obviates the one serious weakness 
 inherent in the Courtis Reading Test. The perform- 
 ances of 247 experimental pupils selected at random 
 from different grades were subjected to this sort of 
 analysis. The results are presented in Table XXVII. 
 
 A very marked increase is shown hi the number of 
 questions correctly answered by the experimental pupils 
 in all the grades. In fact, the number of questions 
 correctly answered by the pupils after receiving the 
 training in rapid reading is greater than the norms or 
 average number of questions attempted as reported by 
 Courtis. The average number of correct answers 
 for the fourth grade as shown in Table XXVII is 34.8 
 as against Courtis' norm of 30 questions attempted ; 
 for the fifth grade it is 42.1 as against 37 ; for the sixth 
 grade, 48.4 as against 40. No norms have been sug- 
 gested by Courtis for the seventh and eighth grades. 
 The average number of questions correctly answered 
 by the different experimental classes at the beginning 
 and end of the training are compared in Graph IX with 
 the norms reported by Courtis for the number of ques- 
 tions attempted, whether answered correctly or not. 
 
 The conclusions that would seem to follow from this 
 phase of the investigation are these : 
 
 1. Marked improvement in the efficiency of the 
 comprehension resulted from the training, as shown by 
 the gain in the number of questions correctly an- 
 swered.
 
 212 
 
 SILENT READING 
 
 A/o. IS. 
 
 r 
 
 I//// 
 
 60 
 S8 
 M 
 
 & 
 & 
 JO 
 49 
 46 
 41 
 4Z 
 
 40 
 38 
 36 
 31 
 3Z 
 30 
 26 
 Z6 
 
 /f 
 2Z 
 
 to 
 JB 
 /& 
 /<? 
 sz 
 /a 
 8 
 4 
 
 4 
 2 
 a 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 / { 
 
 / 
 
 0Q 
 
 77 
 
 _ . 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 x 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 x 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 r 
 
 . 
 
 
 
 
 
 
 
 
 
 
 
 * 
 
 - 
 
 
 
 
 
 
 
 
 
 X 
 
 <^ 
 
 
 
 
 
 
 
 
 / 
 
 ** 
 
 ^~* 
 
 
 
 
 
 
 
 
 
 
 
 S 
 
 ^ 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 S 
 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 <Sl 
 
 /* 
 
 _ 
 
 
 
 
 
 .- 
 
 
 
 ^ * 
 
 
 
 / 
 
 
 
 
 
 
 
 
 
 
 p 
 
 ^ 
 
 
 
 
 s' 
 
 -" 
 
 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 . 
 
 x^ 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 
 
 
 
 
 
 
 . 
 
 --- 
 
 
 
 
 
 
 
 
 
 
 
 
 
 < 
 
 ' 
 
 
 
 
 
 
 ^* 
 
 ^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 "*" 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 <r 
 
 , 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 J f 
 
 > >i 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 faSe // y W y// W 
 
 flribar y/ x? & Jt
 
 INTERPRETATION OF RESULTS 213 
 
 2. This phase of the efficiency of comprehension is 
 measured in no way by Courtis' Index of Comprehen- 
 sion, which is, more strictly speaking, an index of 
 accuracy. 
 
 3. The marked gain in the number of questions 
 correctly answered demonstrates the persistence of the 
 improvement in reading rate in a changed situation in- 
 volving a different mental attitude i.e. in careful 
 reading for the purpose of answering written questions. 
 
 Comparison of the Results of Method I and Method II 
 
 Which type of training No. I. Training in Rapid 
 Silent Reading or No. II. Training to Decrease 
 Vocalization in Silent Reading proved the more 
 effective? As was stated in the previous chapters, 
 these methods are substantially the same the chief 
 difference being a difference of stress. In the latter 
 method the attempt to decrease vocalization in order 
 to improve the rate of reading is made a conscious 
 problem. In Method I the suppressing of the elabo- 
 rate vocalization is accomplished indirectly, by directly 
 emphasizing speed in reading. Not much difference 
 in the results was expected. The results show even 
 less than was expected. 
 
 In the fourth grade, Type I shows a slight superiority 
 in average gain 4.1 words per minute ; in grade five 
 the advantage, 15.9 words per minute, rests with Type II. 
 A comparison between the gains made by a number of 
 sixth-grade pupils using Type II with the gains made
 
 214 SILENT READING 
 
 by a number of seventh-grade pupils employing Type I 
 shows that, in point of absolute gain, Type I has a slight 
 superiority 11.6 words per minute. In the amount 
 of gain made over the corresponding control groups, 
 however, the advantage, 14.9 words per minute, rests 
 with Type II. These results would seem to indicate 
 that the two types of training, because substantially the 
 same, are practically equal in effectiveness. 
 
 The Growth of Speed in Silent Reading as Shown by the 
 Class Charts 
 
 The measurement of the increase in rate in this study 
 was based not only upon the results of the Courtis 
 Silent Reading Test, but also upon the daily class charts. 
 The daily progress of each experimental pupil was re- 
 corded on his individual graph by showing the average 
 number of words read per minute during the total read- 
 ing period on each day of the training period. The 
 median rate of reading for each day was placed on the 
 class chart, which thus reflected the progress of the 
 whole class. The fact that the class chart records the 
 progress for each day, enables it to reflect the character 
 of the growth in speed, gradual or otherwise, which 
 the three Courtis Tests naturally could not show. 
 
 A number of class charts showing the kinds of growth 
 in speed are presented for each grade. Graph X, show- 
 ing the progress of the experimental group in the fourth 
 grade in the Lincoln School at LaSalle, reflects a fairly 
 regular growth. The only appreciable break in the
 
 INTERPRETATION OF RESULTS 
 
 215 
 
 continuity of advancement occurs during approximately 
 the middle of the training period. The curve of im- 
 provement shown in Graph XI reveals considerably 
 more fluctuations. Many of the irregularities in the 
 curve can be explained by the changes in the subject 
 
 6K 
 ttD 
 
 610 
 Jf> 
 
 sto 
 S4e 
 
 J10 
 JOO 
 
 4JO 
 410 
 
 lie 
 4te 
 
 "^ Jfo 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 1 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 '.f. 
 
 
 r ,T 
 
 ', 
 
 
 ',, 
 
 j 
 
 ., 
 
 
 
 ', 
 
 'r' 
 
 1 
 
 . 
 
 r 
 
 /, 
 
 
 / 
 
 
 
 
 
 
 
 
 
 '. 
 
 r' ' 
 
 
 
 
 
 
 H 
 
 r / 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 - 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 1 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 f 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 , 
 
 r^ 
 
 
 
 
 
 
 J, 
 
 
 
 fto 
 
 210 
 ^ 200 
 
 ^ /40 
 HO 
 
 tea 
 
 to 
 to 
 
 4O 
 
 'to 
 e> 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 rv-^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 L _ 
 
 ff 
 
 . ' 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 8 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 
 \ 
 
 
 
 ' 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 
 
 
 
 \ 
 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 -0^ 
 
 
 
 
 
 
 
 ^ "* 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ,^ 
 
 
 e 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ? 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^_ _ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 4 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 
 
 f 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 I 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 2 
 
 J 
 
 4 
 
 3 
 
 f 
 
 7 e 
 
 t > H n a tt ff i* ir / M2oii*2tJmsif*Tu yjfff/jtJOJfJtM 
 
 matter to be read. The reading books on which the 
 rates of speed were made are indicated in the graph. 
 The median growth in speed for the seven fourth-grade 
 classes comprising 117 pupils is shown in the composite 
 Graph XII. The curves are smoothed in all the 
 graphs. The dots represent the actual averages from
 
 216 
 
 SILENT READING 
 
 which the curves were smoothed. The median growth 
 in speed for the fourth-grade classes is on the whole 
 quite regular. The median increase is from 163 words 
 per minute to 310. 
 
 A/O xn. 
 
 6 to 
 
 fto 
 
 (1J 
 660 
 
 seo 
 sto 
 
 S10 
 SJO 
 
 soo 
 
 180 
 1(0 
 HO 
 
 . 400 
 
 fee 
 
 I teo 
 , lie 
 > 1(0 
 140 
 
 ' /20 
 ico 
 to 
 to 
 4o 
 20 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 t, 
 
 ft 
 
 jf 
 
 
 ft 
 
 If 
 
 ft 
 
 , - 
 
 < ( 
 
 'ft 
 
 
 
 
 I 
 
 /I 
 
 Hi 
 
 - 
 
 ft 
 
 ri 
 
 1(1 
 
 
 -f 
 
 
 
 
 /, 
 
 "'I 
 
 t - , 
 
 ', 
 
 
 
 
 
 
 71 
 
 . 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 a 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 I 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 " 
 
 
 
 
 Os 
 
 .. 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 \ 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 u 
 
 
 
 ^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 \ 
 
 
 
 -\ 
 
 
 
 \\ 
 
 
 
 
 / 
 
 
 
 
 
 
 
 \fl 
 
 
 \ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 | 
 
 ' 
 
 
 s> 
 
 
 
 
 
 
 . 
 
 
 
 1 
 
 
 
 
 >X 
 
 . 
 
 
 " 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 \ 
 
 ' 
 
 I 
 
 * 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 c 
 
 
 
 <* 
 
 ^ 
 
 s 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 i ^ 
 
 
 I 
 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 . 
 
 
 
 
 
 
 
 
 
 
 V 
 
 X 
 
 
 
 " v 
 
 S-l 
 
 ' 
 
 
 ,k 
 
 
 
 
 
 
 
 
 
 
 
 I 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 y 
 
 
 
 I 
 
 
 
 J 
 
 n 
 
 
 2 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 
 
 
 
 
 
 
 
 w 
 
 
 
 
 
 
 
 
 f 
 
 t> 
 
 
 
 
 N 
 
 
 
 
 
 
 ^ 
 
 
 >i 
 
 
 
 
 
 
 
 
 
 ^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ki 
 
 
 
 
 
 
 .S; 
 
 
 
 's 
 
 
 
 \ 
 
 
 ^ 
 
 
 
 
 r 
 
 
 
 
 
 j 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 s 
 
 
 
 i 
 
 
 
 
 
 w 
 
 
 
 
 \ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 \ 
 
 
 
 
 
 
 
 xy 
 
 
 
 
 
 
 I 
 
 
 
 j 
 
 
 
 i 
 
 
 '^ 
 
 
 
 
 
 
 
 
 
 \v 
 
 
 
 
 
 
 
 
 
 ^ 
 
 
 
 
 
 
 
 jti 
 
 
 
 
 
 
 
 
 
 ^ 
 
 
 
 
 
 \ 
 
 
 
 
 X; 
 
 
 
 
 
 N, 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 s 
 
 
 
 
 
 
 
 
 
 , 
 
 
 
 
 
 V 
 
 
 
 
 
 
 
 X, 
 
 
 \ 
 
 
 
 
 
 
 
 
 
 I 
 
 
 
 
 
 
 
 J 
 
 
 
 
 
 
 j 
 
 
 
 j 
 
 
 
 ; 
 
 
 x! 
 
 
 
 
 
 
 
 
 
 * 
 
 
 
 
 
 
 
 
 
 t 
 
 
 
 
 
 
 
 ^ 
 
 
 
 
 
 
 ? 
 
 
 
 -> 
 
 
 
 V 
 
 
 * 
 
 
 
 
 ^ 
 
 
 
 N 
 
 
 V 
 
 
 
 
 
 
 
 
 
 \l 
 
 
 
 
 
 
 
 i 2. 3 1 s 6 7 e 3 /o n u a * .a ie n ig 20 21 22 23 21 25 zt 27 zg zs x j/ 32 33 itjsx 
 
 C C f) JfCl/ r/Vf ^JCflOOf <Xtf6<5 
 
 The curves of progress for two fifth-grade classes 
 are shown in Graphs XIII and XIV. The markedly 
 irregular curve of progress in the latter graph contrasts 
 with the general smoothness of the composite chart 
 shown in Graph XV. The increase is somewhat more 
 marked than in the fourth grade being from 183 to 
 approximately 378 words per minute.
 
 INTERPRETATION OF RESULTS 
 
 217 
 
 In the curve of progress at St. Mary's School as shown 
 in Graph XVI, it is to be noted that the growth in speed 
 is much more marked in the latter than in the first half 
 of the training period. The median growth of three 
 
 A/o X77 
 
 / 13 1 f 7 6 9 10 // tf /J /f *f/ // fl It 29 II a U If tftt 17 U 2f M )< 31 JU Jf JO Jt 
 
 Con3fCivf/vf JcSroe/ 
 
 sixth-grade classes is shown hi the composite Graph 
 XVTIL The curve is not so regular as in the previous 
 composite graphs. 
 
 The curves in Graphs XIX and XX manifest minor 
 fluctuations, though the central tendency of progress 
 is strongly in evidence. The composite Graph XXI
 
 218 
 
 SILENT READING 
 
 showing the growth of seven seventh-grade classes 
 comprising 119 pupils displays more smoothness. The 
 increase is from 220 words per minute to approximately 
 490. In the composite Graph XXIII of the three 
 
 A/o KHJ 
 
 60 
 
 10 
 
 tzo 
 
 600 
 
 S60 
 S10 
 JZO 
 SCO 
 180 
 
 ^ J60 
 JfO 
 
 i; xo 
 
 I 300 
 ^teo 
 
 260 
 
 >> zzo 
 
 ^ 200 
 ^ /8O 
 
 ^ /so 
 no 
 
 110 
 
 100 
 60 
 
 eo 
 to 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 $ 
 
 rt* 
 
 _ 
 
 f t 
 
 ,- 
 
 T^ 
 
 I 
 
 _ 
 
 
 ^ 
 
 " f 
 
 J 
 
 '^ 
 
 ^ 
 
 
 
 / ( 
 
 f, 
 
 ^ 
 
 "fl 
 
 ; , 
 
 . 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 c- 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 . 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 t 
 
 
 -^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 J 
 
 
 
 V. 
 
 
 .- 
 
 r-~ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 I 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 , 
 
 
 
 
 
 
 
 
 1 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 I 
 
 
 
 
 1 
 
 
 
 
 
 ^ 
 
 ~S 
 
 1 
 
 "I 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^- - ' 
 
 ' 
 
 
 
 
 
 
 X 
 
 ^_ 
 
 S 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 '"" 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 I 
 
 ? 
 
 1 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 \ 
 
 
 
 
 /< 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 s 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 - 
 
 ^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 / r J 4 S S 76 3 IO H J2 U M fS 16 17 la IJtO 2/ U ZJJ1ZSZ6J IS* 
 
 eighth-grade classes, the steep upward slope of the curve 
 is more marked than in any of the previous composite 
 graphs. Moreover, the increase is consistent, and 
 there is no indication that in this the highest grade im-
 
 INTERPRETATION OF RESULTS 
 
 219 
 
 provement might not continue to an indefinite extent 
 if the practice period had been longer. 
 
 eoo 
 
 340 
 
 sso 
 
 J40 
 
 520 
 
 jeo 
 
 110 
 
 166 
 410 
 110 
 . 40* 
 *> J8A 
 V* 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 f, 
 
 f 
 
 a f 
 
 (1! 
 
 
 It 
 
 I t 
 
 - f 
 
 if, 
 
 
 
 P/ 
 
 {tf\ 
 
 ^h 
 
 
 ( r 
 
 ht 
 
 j 
 
 f - 
 
 ( 
 
 r- { 
 
 J rf 
 
 I* 
 
 y 
 
 _ 
 
 /,<i 
 
 ^ 
 
 
 r it 
 
 f1 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 A, 
 
 1 
 
 
 S 
 
 
 
 
 
 
 
 
 
 "^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 i 
 
 
 
 ^ 
 
 
 - 
 
 -~t. 
 
 
 
 \ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 1 
 
 
 
 ' 
 
 
 
 
 
 
 1 
 
 
 
 /- 
 
 
 >~s, 
 
 
 
 
 ^ J<tO 
 ^JZO 
 
 joe 
 
 i 
 
 \M 
 
 ^ too 
 AUO 
 
 I? 26t 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 I 
 
 
 
 
 
 
 
 
 
 V, 
 
 
 x 
 
 
 
 
 1 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ] 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 J 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 1 
 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ' 
 
 
 ^ 
 
 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 V, 
 
 
 ' 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 \tao 
 
 Q 160 
 
 ^ /^ 
 
 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 *+- 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 izo 
 loo 
 
 
 
 . 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 60 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 6& 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 40 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 *6 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 a 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 3 1 S 6 7 3 10 I/ K a H IS // K O Z021 UUt1 tS 16 f 7 2} ft 31 JZJ3M 
 
 TABLE XXVII. GAIN IN COMPREHENSION OF THE EXPERIMENTAL 
 PUPILS AS DETERMINED BY THE NUMBER OF QUESTIONS CORRECTLY 
 ANSWERED IN COURTIS TESTS 
 
 GRADE 
 
 NUMBER or 
 PUPILS 
 
 BEGINNING 
 
 END 
 
 GAIN 
 
 IV .... 
 
 97 
 
 23 9 
 
 34.8 
 
 10.9 
 
 V 
 
 47 
 
 296 
 
 42 1 
 
 12.5 
 
 VI 
 
 57 
 
 33.7 
 
 48.4 
 
 14.5 
 
 VII 
 
 50 
 
 453 
 
 58.0 
 
 12.7 
 
 VIII 
 
 46 
 
 47 1 
 
 57.8 
 
 10.7 
 
 
 
 

 
 220 
 
 SILENT READING 
 
 TABLE XXVIII. COMPARISON OF GROWTH OF SPEED IN SILENT 
 READING OF 23 CLASSES IN DIFFERENT GRADES, AS DETERMINED BY 
 THE COURTIS TESTS WITH THAT RECORDED ON THE CLASS CHARTS 
 
 
 BEGINNING 
 
 END 
 
 GAIN 
 
 PEB CENT 
 OF GAIN 
 
 Courtis Tests . . . 
 
 189 
 
 295 
 
 106 
 
 56% 
 
 Class charts .... 
 
 228 
 
 411 
 
 183 
 
 80% 
 
 Table XXVIII is to be read thus : At the beginning of the training 
 the median rate of speed of the 23 classes as determined by the Courtis 
 Tests was 189 words per minute, 295 words at the end of the training. 
 The median gain in number of words per minute was 106, the median 
 gain in per cent was 56%. 
 
 600 
 J60 
 
 J40 
 320 
 
 190 
 
 X 1OQ 
 \ 360 
 
 J*0 
 
 IM 
 260 
 ^% 
 
 * /go 
 ^ a* 
 
 /aa 
 fa 
 
 so 
 
 40 
 fO 
 O 
 
 
 /Vo ZF 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ( 
 
 fr 
 
 a 
 
 ?* 
 
 
 . 
 
 
 7^ 
 
 y// 
 
 '/ 
 
 
 ^ 
 
 i-'i 
 
 'r 
 
 ^? c 
 
 ffi 
 
 
 
 
 P 
 
 /r ; 
 
 
 fl 
 
 t i 
 
 ?r 
 
 
 7/ 
 
 / 
 
 ff 
 
 "7 f 
 
 /^ 
 
 he 
 
 
 
 
 
 fa 
 
 
 
 <f 
 
 
 j 
 
 <> 
 
 J 
 
 "^ 
 
 r^ 
 
 J C, 
 
 
 
 
 
 , 
 
 
 
 
 
 ^; 
 
 ^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 -* 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ' 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 _- 
 
 c= 
 
 1 
 
 -^ 
 
 ^s 
 
 =I 
 
 ^, 
 
 
 r ' 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ,x- 
 
 -^ 
 
 
 
 
 
 
 
 
 
 
 
 < 
 
 
 
 < 
 
 
 ^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 1 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ,, 
 
 -^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 l*~ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 ' 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 i 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 s 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 2 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 -3 * 36 7 a 
 
 fffwa
 
 No. 
 
 410 
 430 
 420 
 4/0 
 400 
 3?O 
 3SO 
 370 
 360 
 3SO 
 34O 
 330 
 32O 
 310 
 300 
 2fO 
 2SO 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 f 
 
 ff n 
 
 Ti 
 
 77 ( 
 
 Iff 
 
 fr 
 
 -i 
 
 5 
 
 A* 
 
 re' 
 
 f - 
 
 ,; 
 
 
 Lf 
 
 r f 
 
 
 ir^i 
 
 /I 
 
 -f?^ 
 
 - 
 
 ^ 
 
 ^ 
 
 /^7^ 
 
 A 
 
 
 
 7 
 
 
 \ 
 
 f 
 
 
 
 
 
 
 t 
 
 
 
 
 
 ^ 
 
 rl 
 
 fl 
 
 ^ 
 
 fi 
 
 
 
 
 
 
 
 
 
 
 
 
 j 
 
 
 ' 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ( 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 j 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 7 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ / 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ) 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 i 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 7 
 
 D 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 
 
 
 270 
 
 240 
 2JO 
 220 
 
 \i. 2/0 
 
 \zoo 
 
 *//* 
 
 < 170 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 X 
 
 7 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 J 
 
 ^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 rf^ 1 
 
 5 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 t 
 
 x 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 X 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 y* 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 X" 
 
 ) 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 >^ 1 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 f 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 yS 
 
 s 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 150 
 
 w 
 
 V /Jo 
 
 ^ '2 
 ^ //O 
 /0O 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 60 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 60 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 10 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 JO 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 2O 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 > 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 136 76 9/f0/ra/t*TI6/7 /6 IS ZOZ/ ZZflttiSH *72i 29 JO 
 
 221
 
 222 
 
 SILENT READING 
 
 TABLE XXIX. THE AVERAGE READING RATE OF BOYS AND GIRLS 
 AT THE BEGINNING AND END OF THE EXPERIMENT AND THE 
 RELATIVE AMOUNT OF GAIN 
 
 
 . BOYS 
 
 GIRLS 
 
 AVERAGE 
 GAIN 
 
 GIRLS' SUPERI- 
 ORITY IN 
 
 
 No. of 
 Boys 
 
 Begin- 
 ning 
 
 End 
 
 No. of 
 Girls 
 
 Begin- 
 ning 
 
 End 
 
 Boys 
 
 Girls 
 
 Final 
 Ave. 
 
 Gain 
 
 III . 
 
 12 
 
 137 
 
 231 
 
 11 
 
 152 
 
 254 
 
 94 
 
 102 
 
 23 
 
 8 
 
 IV . 
 
 49 
 
 144 
 
 220 
 
 53 
 
 157 
 
 234 
 
 76 
 
 77 
 
 4 
 
 1 
 
 V . . 
 
 27 
 
 154 
 
 276 
 
 22 
 
 170 
 
 266 
 
 122 
 
 96 
 
 -10 
 
 -26 
 
 VI . 
 
 19 
 
 192 
 
 288 
 
 32 
 
 196 
 
 289 
 
 96 
 
 93 
 
 1 
 
 -3 
 
 VII . 
 
 38 
 
 195 
 
 315 
 
 64 
 
 210 
 
 326 
 
 120 
 
 116 
 
 11 
 
 -4 
 
 VIII . 
 
 21 
 
 228 
 
 435 
 
 24 
 
 211 
 
 351 
 
 207 
 
 140 
 
 -84 
 
 -67 
 
 Table XXIX is to be read for grade 3 as follows : 12 boys had an av- 
 erage rate of 137 words per minute at the beginning of the training, 231 
 words at the end ; 11 girls averaged 152 at beginning, 254 at end ; boys' 
 gain was 94 words per minute, the girls' was 102 ; the superiority of the 
 girls' final average over the boys' final average was 23, the girls' superior- 
 ity in gain over the boys was 8 words per minute. 
 
 Table XXVII compares the median rate of speed of 
 23 classes in various grades as determined by the Courtis 
 Reading Test with that shown by the class charts. 
 The table shows that the amount of increase in speed 
 as recorded on the class cKarts for the different grades 
 is greater than that manifested by the Courtis Tests. 
 This is due to the greater stimulation, the greater 
 pressure, that was brought to bear upon the pupils 
 in the regular daily training in rapid reading. The 
 pressure of the individual graph and of the class 
 chart was not applied to the reading of the Courtis 
 Test. 
 
 In Graph XXIV are shown the rates of all the grades
 
 INTERPRETATION OF RESULTS 223 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 K 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 i 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 i 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 s 
 
 
 --' 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 J 
 
 
 
 
 
 
 
 
 
 
 
 
 
 \ 
 
 T 
 
 "/ 
 
 *; 
 
 f 
 
 
 ' 
 
 * 
 
 v " 
 
 ^ 
 
 . 
 
 r .. 
 
 
 \ ' 
 
 i' 
 
 
 ' 
 
 r 
 
 J 
 
 i> 
 
 
 
 
 
 
 
 
 
 
 
 
 
 \ 
 
 
 a 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 s^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 \ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ? 
 
 
 
 
 
 
 
 
 
 
 
 
 
 o 
 
 
 
 
 
 
 
 
 
 v 
 
 J _ 
 
 ^ 
 
 " 
 
 V 
 
 ' * 
 
 j 
 
 
 - 1 - 
 
 2 
 
 ^ 
 
 
 3 
 
 
 
 
 
 
 
 
 
 
 
 
 
 \ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 t; 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 s 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 1 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 85 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 Nh 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 /' 
 
 *" 
 
 / 
 
 ' 
 
 *"-* 
 
 3 
 
 t 
 
 
 ? 
 
 / 
 
 / 
 
 /' 
 
 ^ 
 
 Sj 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 
 >j 
 
 
 
 
 
 
 
 
 
 
 
 
 
 a 
 
 7 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 
 ^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 1 ~ 
 
 ) Ji 
 
 " '> 
 
 f 
 
 s S 
 
 
 3 
 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 
 e 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ftj"5 
 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 Si 
 
 
 ',- 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^"^ 
 
 
 -; 
 
 
 
 
 
 
 
 
 
 
 / 
 
 c 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 >* 
 
 
 3 
 
 
 
 
 
 
 
 
 
 
 i 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^^5 
 
 
 
 
 
 
 
 
 
 
 
 
 \ 
 
 "" 
 
 
 > 
 
 r 
 
 
 / 
 
 
 '/ 
 
 >' 
 
 
 
 ", 
 
 " 
 
 J 
 
 '^ 
 
 ' 
 
 
 -" J 
 
 
 - 
 
 V ' 
 
 
 , 
 
 
 
 
 
 
 
 
 
 
 \ 
 
 t 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 5 
 
 
 ^s 
 
 
 
 
 
 
 
 
 
 
 
 c 
 
 V 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 Vj v 
 
 
 v- 
 
 
 
 
 
 
 
 
 
 
 
 
 
 \ 
 
 
 i 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^\ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 \ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ ;\ 
 
 
 1 
 
 
 
 
 
 
 
 
 
 
 
 
 
 c 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ u 
 
 
 - 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^^ 
 
 
 1 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 * ' 
 
 '* 
 
 
 '-^ 
 
 
 - 
 
 
 ^^ 
 
 
 ^j 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 [ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 \\5 
 
 
 " 
 
 
 
 
 
 
 
 
 
 
 
 
 . 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 s 
 
 
 V 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 \ 
 
 
 - 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 K 
 
 
 N 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 4 
 
 \ 
 
 
 
 
 
 
 
 ) 
 
 ~ 
 
 , 
 
 
 
 
 
 
 
 
 
 i 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 a 
 
 
 
 
 
 ' 
 
 
 
 ' 
 
 
 . 
 
 
 
 
 
 
 Vj 
 
 
 ^. 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 v 
 
 
 > 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 V 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 - 
 
 \ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 N 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 S^ 
 
 c 
 
 
 
 
 
 
 
 
 
 
 
 V 
 
 1 
 
 ^ 
 8] 
 
 *, 
 
 .. 
 
 * 
 
 
 
 ij 
 
 j 
 
 $ 
 
 1 
 
 1 
 
 .. 
 
 ,. 
 
 5 
 J? 
 
 1 
 
 s 
 
 : ; 
 
 > 
 
 .. 
 
 N 
 
 . 
 - 
 
 - 
 i 
 . 
 
 : 
 
 5 
 
 ". 
 
 ' 
 
 \ 
 
 . 
 N 
 
 ~ 
 
 - 
 >. 
 
 i 
 
 ' 
 
 s. 

 
 224 
 
 SILENT BEADING 
 
 at the beginning and at the end of the experimental 
 training as recorded on the class chart. The shaded 
 blocks in the graph call attention to the amount of 
 gain in rate made by the different grades. 
 
 In short, the results of the class charts offer a rather 
 
 /I/O * 1/7(7 
 
 sec 
 
 S10 
 
 ja 
 
 C!> 
 
 #e 
 
 100 
 
 xe 
 jt 
 
 _ 
 j 
 
 2 to 
 
 :u 
 
 / 
 /it 
 
 M 
 
 to 
 
 / 
 C 
 
 1 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 r< 
 
 rr, 
 
 PI 
 
 ^ 
 
 i 
 
 11 r 
 
 , v 
 
 T 
 
 
 ^ 
 
 'r-c 
 
 /, 
 
 
 
 ' 
 
 I 
 
 
 ,/ 
 
 n 
 
 "("r 
 
 
 / r 
 
 ^ 
 
 
 J^. 
 
 
 
 
 
 f 
 
 
 
 ^ 
 
 ! 
 
 ?^ 
 
 -rf 
 
 ^ 
 
 ', 
 
 ^ 
 
 
 J 
 
 . 
 
 
 
 12 
 
 
 r'r 
 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ,/ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 X 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 y 
 
 
 
 
 
 
 ^ 
 
 ^ 
 
 
 
 
 
 ^x 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 i 
 
 X 4 
 
 
 
 
 
 
 1 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 s 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 I } 1 S 6 7S 9/o/lUlJWif/6l7t3O202IZfaif&U3?tO 
 Conjccvt/i/ff Jctooo/ </ay^. 
 
 striking corroboration of the general conclusions drawn 
 from the data afforded by the Courtis Test. 
 
 Correlation of Rate of Reading with Simple Reaction 
 Time and Discrimination Reaction Time 
 
 The attempt was made in the present investigation 
 to ascertain the correlation between the reading rate 
 and the rates of simple motor reaction and discrimina-
 
 INTERPRETATION OF RESULTS 
 
 225 
 
 tion reaction. The test for simple reaction was the 
 cancellation test A-3 described by Whipple in his 
 Manual of Mental and Physical Tests. As used in 
 this experiment, the pupils were requested to cross 
 out all the geometrical figures, instead of only certain 
 ones. The discrimination reaction test called for the 
 
 s 
 
 / 
 
 canceling of only the numeral 9 in lines containing va- 
 rious digits. These tests were administered to the 
 pupils in the fourth and fifth grades in the Goodrich 
 School, whose rates of reading had been previously 
 measured by the Courtis Test. 
 
 The correlation between the rate of reading and 
 simple motor reaction was found to be rather low
 
 226 
 
 SILENT READING 
 
 r being .214 for the fourth grade and .228 for the fifth. 
 The coefficient of correlation between the rate of read- 
 
 600 
 710 
 jed 
 71i> 
 
 no 
 
 760 
 610, 
 660 
 640 
 UO 
 (00 
 SfO 
 560 
 541 
 S2C 
 300 
 4/6 
 460 
 440 
 
 340 
 ^320 
 
 zeo 
 /to 
 Ito 
 
 140 
 IZO 
 
 too 
 
 10 
 
 so 
 
 40 
 
 to 
 c 
 
 S 6 7 6 9 /O // S2 /J /* /S /6 // /g /3 20 Z/ 21 
 
 ZeZ9J03IXUJ1J536 
 
 ing and the rate of discrimination reaction was com- 
 puted for the pupils of the fourth grade. The correla- 
 tion in this case was found to be somewhat higher
 
 INTERPRETATION OF RESULTS 
 
 227 
 
 r being .435. On the whole, the coefficients are too low 
 to indicate the existence of any very significant positive 
 correlation between the reading rate and the rate 
 
 - 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 j 
 
 -V 
 
 
 
 
 1 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 -, 
 
 
 
 |V 
 
 
 
 
 
 v 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 -, 
 
 B 
 
 
 
 
 
 
 
 
 
 ^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 V 
 
 
 
 
 
 
 s 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ! 
 
 
 
 .' 
 
 
 
 
 
 
 
 \ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 I 
 
 
 
 
 
 
 
 
 5 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 K 
 
 
 
 3 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ;. 
 
 
 
 K 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 I 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 . 
 * 
 
 
 
 '" 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 x -' 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ;' 
 
 *> 
 
 
 
 
 
 
 
 
 \ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 'i^ 
 
 
 
 ^ 
 
 
 
 
 
 
 
 
 
 \ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 | 
 
 u 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 j 
 
 
 
 
 
 
 
 
 
 
 
 
 
 V 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 X. 
 
 
 
 
 
 
 
 
 
 
 \ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 r 7 ^ /o // // a ft ts * /rat 
 C enjftt/ ?/rc J<At 
 
 
 
 \ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 V ' 
 
 ', 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 . 
 
 .- 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 .1 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ., 
 
 s 
 
 
 
 
 
 
 
 
 
 V, 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 'I 
 
 
 
 
 
 
 
 
 
 
 
 x^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 1 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 s 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ,, 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 - 
 
 " 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 i 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 - 
 
 
 
 
 
 
 
 
 
 
 
 
 \ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 \ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 v 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 -' 
 
 
 
 
 
 
 
 
 
 
 
 
 
 \ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ,SJ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ~^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 \ 
 
 
 
 
 
 
 
 
 
 
 
 1^^%S?|S5S5 
 
 a^^l^sg^^s^ 
 
 of either simple motor or discriminatory reaction, 
 though the correlation is certainly more in evidence 
 in the latter case.
 
 228 
 
 SILENT READING 
 
 The Work Curve for One Reading Period 
 
 Besides determining the average number of words 
 read per minute by each pupil for each of the total 
 reading periods of, say, 20 or 25 minutes during the 
 
 OU 
 
 fet 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ao 
 uc 
 
 610 
 
 no 
 
 tin 
 J8C 
 Sfi 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ,y 
 
 fr 
 
 "rt 
 
 *fi 
 
 , 
 
 (f 
 
 n, 
 
 n 
 
 _ 
 
 Cf 
 
 (!< 
 
 / 
 
 f / 
 
 z 
 
 ") 
 
 L/ 
 
 M 
 
 /- 
 
 / 
 
 ca 
 
 4 
 
 
 'II 
 
 ' - 
 
 1 
 
 S 
 
 ft 
 
 aZ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 . 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 ^ao 
 
 486 
 460 
 
 44C 
 
 ^ 400 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ' 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 1 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 f 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 \ XC 
 J4i 
 
 
 
 
 
 
 
 
 
 i 
 
 
 
 
 / 
 
 ' 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 
 
 
 J 
 
 *** 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 /< 
 
 1 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 \JOC 
 
 ao 
 
 vji 2U, 
 
 
 
 
 ( 
 
 /i 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 f 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 X 1 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 \ 210 
 
 ^ ttx 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 a 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ao 
 
 160 
 141. 
 110 
 
 /oo 
 
 <x 
 
 
 46 
 2O 
 O 
 
 j 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 Z J 4 6 7 9 10 II li.a H 15 16 17 18 IB Z02I 222J21 2526 Z7 26 
 Conje cvfii/e Jcaoo/ aat/-s. 
 
 experiment, two days were specified on which more 
 detailed measurements were requested. The pupils 
 were asked to record the number of words read per 
 minute during each of the several briefer stretches
 
 INTERPRETATION OF RESULTS 
 
 229 
 
 constituting the total reading period for the day. 
 This was done in an effort to determine the daily work 
 curve for reading. 
 
 A study of the work curves of the various classes 
 shows there is no uniform type of curve. Variability 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 -L a 
 
 _ 
 
 4 
 
 
 h 
 
 
 m. 
 
 
 
 
 ' 
 
 
 ^ 
 
 13 
 
 
 Ff 
 
 f 
 
 ' 
 
 f 
 
 ' 
 
 rf 
 
 f 
 
 . r 
 
 
 r 
 
 ( 
 
 -rfr* 
 
 >3 
 
 
 
 
 
 
 
 a 
 
 
 ^ .-/ 
 
 r ' 
 
 "/ fl 
 
 ,, 
 
 f>1 
 
 t 
 
 /I 
 
 
 r^ 
 
 / 
 
 '. 
 
 . 
 
 
 J 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ' 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 S* 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ~~ 
 
 ^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 ^ 
 
 *~~ l 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 |X 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 f 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 > 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 ^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 f> 
 
 
 
 
 
 ^^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 1 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 f 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 j. 
 
 S 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 J 1 
 
 Coo-sectlfiit* 
 
 is the striking characteristic. In some classes, pupils 
 read at maximal rate in the middle stretches, in others, 
 toward the end, while in still other classes, pupils 
 reached their highest rate in the second and third 
 stretches. These results would seem to support the
 
 230 
 
 SILENT READING 
 
 conclusion of Thorndike that there is no uniform type 
 of work curve, with clearly defined characteristics, re- 
 
 showing tfoe A vera$e number 
 of Words J3ea<J per PtJp/7 >e/ore av</ after Train/' 06 // 
 radej /V to VIII aj </e term wet/ & "the ina/'t//ac/a/ 
 
 eoa 
 
 VIII 
 
 vealing the successive occurrence of " initial spurt," 
 "warming-up" (vinregrwngr), a adaptation" (Gewohnung),
 
 INTERPRETATION OF RESULTS 231 
 
 and the other processes which Rivers and Kraepelin, 
 and especially Offner, have predicated of work curves 
 in general. 
 
 The Factor of Sex 
 
 Table XXIX shows that the average rate of the 
 girls is slightly higher than that of the boys. At the 
 beginning of the training the average rate for the girls 
 was higher in every grade except the eighth ; at the end 
 of the training in every grade save the eighth and the 
 fifth. In amount of gain as a result of the training 
 the boys somewhat surpassed the girls. The training 
 in rapid reading was rather intensive and somewhat 
 strenuous and it is possible that the neuro-muscular 
 system of the boys enabled them to profit by it slightly 
 more than the girls. Both sexes, however, showed very 
 marked gains in rate in all the grades and the differences 
 in the amount of gain made by the two sexes are too 
 slight to be of any special significance.
 
 CHAPTER X 
 
 PHOTOGRAPHIC RECORDS OF EYE-MOVEMENTS 
 DURING READING 
 
 A STANDARDIZED silent reading test is doubtless 
 capable of measuring the rate of reading satisfactorily. 
 It can determine the increase or decrease in speed 
 effected in a given time. Behind that simple factual 
 statement, however, a reading test does not penetrate. 
 It tells nothing about the factors which lie behind 
 that increase or decrease in rate nothing about the 
 causes proximate or remote. In the present in- 
 vestigation an effort was made to penetrate behind the 
 simple determination of increase in speed, as shown 
 by the Courtis Silent Reading Test and the individual 
 charts, into the nature of the modifications of the physi- 
 ological factors which accompanied the increase in rate. 
 In other words, how is an increase in speed to be ex- 
 plained in terms of the physiological processes con- 
 stituting the reading complex ? 
 
 Is the habit of increased speed achieved chiefly by 
 the shortening of the duration of the fixation-pauses? 
 Or by a lessening of their number ? Or by a combina- 
 tion of both these results? This Schmidt (18) char- 
 
 232
 
 PHOTOGRAPHIC RECORDS OF EYE-MOVEMENTS 233 
 
 acterizes as one of the "interesting, but as yet unsolved 
 problems" in the psychology of reading. What, 
 moreover, are the changes in the character of the eye- 
 movements of an individual who has developed from 
 a slow to a rapid reader? Have the movements be- 
 come more regular and rhythmical, or the opposite? 
 Has the acquirement of the habit of increased speed 
 resulted in an increase in the number of regressive 
 movements which are signs of mental confusion 
 or have they decreased ? Is the rate of the inter-fixa- 
 tion movements and the interlinear "sweeps" affected 
 by the habit of increased speed in reading? 
 
 These are some of the interesting questions which 
 have been broached in recent studies* in the technical 
 psychology of reading. In the present investigation 
 the attempt was made to secure, some data for the 
 formulation of at least tentative answers to the above 
 queries. 
 
 For this purpose, photographic records were taken 
 of the eye-movements of ten pupils while they were 
 engaged in silent reading. These ten pupils, five boys 
 and five girls, representing various degrees of silent 
 reading ability, were selected from the classes in the 
 Goodrich, Hendricks, and Manierre schools in Chicago, 
 which were receiving training in rapid reading as out- 
 lined in Type I and Type II. The elaborate photo- 
 graphic apparatus, the evolution of which was men- 
 tioned in Chapter I, and which has been recently 
 improved upon by Judd, C. T. Gray, Freeman, Schmidt,
 
 234 SILENT READING 
 
 and others at the University of Chicago, was used in 
 this connection. Records were taken both before and 
 after the training in rapid reading. 
 
 PLATE I. SILENT READING BY A. P., A FOURTH-GRADE SUBJECT 
 BEFORE TRAINING 
 
 ' 3 S M. 10 T (i tO It f // H 
 
 hfh^n I fiave t|ie)n a]|, "he sjiid, "I'jl l|eaTJe|this| | 
 ? g IJ 175 I' * x S io 107 *7 
 
 a,/JTjf*7 sr q to n 
 
 dirty jwaierj and goj up int$ the oj-chclrd. What I 
 
 Vi 10 li 8 x a n /3 aty I0 ' ' 
 
 ^ ^ ' S L 7 <? g 10 || M /L l^f 
 
 flwj ijb frffll b^ to hop andlhopjanp hop, If elnl^ I 
 lol7 4 13 '* li 7 5 J3 k I* 1+ 
 
 a a. + I J- t 8 1 7 <o /I jx 
 
 | hjidja |lit^le brother] toj h|op Jwith mej I should ^>e 
 (if 4 4 i* /o ii x i/ if u 
 
 i ' 
 
 I so happy." 
 aS (I 
 
 X indicates that it was impossible to determine with accuracy the duration of the 
 fixation-pause. The vertical lines indicate the position of the eye's fixation; the 
 numbers at the top of the line show the order in which the fixations occurred. The 
 numbers at the bottom of the line give the duration of each fixation in fiftieths of a 
 second. Thus Plate I is to be read : The first fixation occurred on the quotation 
 mark (') before the word "when"; its duration was A of a second; the second 
 fixation was on the letter " e " in the word " when " ; its duration was A of a second. 
 
 In order that comparisons might properly be made 
 between the eye-movement records of pupils in differ- 
 ent grades, it was necessary to secure subject matter 
 which was adjusted to the comprehension of the differ-
 
 PHOTOGRAPHIC RECORDS OF EYE-MOVEMENTS 235 
 
 ent grades, and which was of approximately equal 
 difficulty for the pupils in each grade. The selections 
 in Gray's Silent Reading Test Tiny Tad, The Grass- 
 hoppers, Ancient Ships having been carefully evalu- 
 ated as to their degree of difficulty, proved well suited 
 
 PLATE II. SILENT READING BY A. P., A FOURTH-GRADE SUBJECT 
 AFTER TRAINING 
 
 "ThejsW TOs harBli up flhe next iorni^g when k 
 fyfcit fy jT> ?o if 8 * 
 
 W* * s 
 jumped out if thelwatelr and h 
 ; " 9 9 
 
 olt on the (bank. Hi was very] small, but InorU 
 |V i 1 T 4 l V 7 
 
 I JL J -ff 
 
 too enkll for his littlejlegs that wibblfcd under 
 iV // J -* 
 
 him. It iris Tfii(y> the yolng toad."/ 
 /I ? ?tu i? 
 
 X indicstes that it WM impossible to determine with accuracy the duration of th. 
 fixation-pause. 
 
 for this purpose. Different passages in the various 
 selections were used at the beginning and at the end 
 of the training, so that, in no case, was the complicat- 
 ing factor of special familiarity with the specific subject 
 matter introduced. The instructions given to the
 
 236 SILENT READING 
 
 pupils before reading Gray's test at the beginning and 
 at the end of the training period were substantially 
 the ones which preface every standardized silent read- 
 ing test " Read only as rapidly as you can under- 
 stand what you are reading." 
 
 The situation hi which the pupils read while their 
 eye-movements were being photographed was naturally 
 considerably different from the one to which they were 
 accustomed in the classroom. The room was darkened, 
 the selection alone being illuminated . A ray of light was 
 reflected into the subject's eyes, thence into the lenses 
 of the camera. The head of the subject was prevented 
 from moving by means of a headrest. The latter had 
 four points of contact one at the forehead, one at 
 the back, and two at the cheeks. After it was ad- 
 justed, the subjects did not find it especially disturbing. 
 A few of the younger subjects seemed to find the situ- 
 ation quite amusing, as shown by their smiles. 
 
 I. EYE-MOVEMENT RECORDS OF FOURTH-GRADE 
 SUBJECTS BEFORE AND AFTER TRAINING 
 
 1. A Slow Fourth-Grade Reader 
 
 Plate I shows the record of A. P. before receiving 
 the experimental training. A. P. is a ten-year-old 
 boy in the fourth grade in the Goodrich School. His 
 general class average is rated by the teacher as "fair," 
 his oral reading as "poor." His rate in silent reading
 
 PHOTOGRAPHIC RECORDS OF EYE-MOVEMENTS 237 
 
 was only 73 words per minute as measured by the 
 Courtis Silent Reading Test ; his comprehension was 
 very low. He ranked among the lowest in his class 
 in both rate and comprehension. He is of foreign-born 
 parentage. His father does not speak English, though 
 his mother does. 
 
 The record in Plate I reflects very clearly the physi- 
 ological factors back of the pupil's deficiency in reading. 
 The number of fixations is excessively large ; regressive 
 movements are too numerous. Not a single line is 
 without them. They show traces of unmistakable 
 confusion in the interpretation of the printed symbols. 
 They indicate that the reader was obliged to go back 
 frequently to previous words in order to get the mean- 
 ing of the present word or phrase. Regular rhythmical 
 eye-movement habits are conspicuous by their ab- 
 sence. The visual span is very narrow. Thus, in 
 line 2, the short word "water" receives three fixa- 
 tions. In line 3 there are four backward or regres- 
 sive movements. 
 
 The record in Plate II for A. P. after training shows 
 considerable improvement over the subject's previous 
 performance. Two months of the experimental train- 
 ing of Type II had effected a marked increase in speed. 
 His record on the Courtis Test showed that he had 
 more than doubled his rate, while his comprehension 
 had likewise improved. The amount of pressure that 
 was brought to bear upon the pupils to increase their 
 rate in the regular reading period in the school was
 
 238 SILENT READING 
 
 naturally not applied to their reading before the camera. 
 Under the novel circumstances under which the latter 
 was done, and the very brief interval allowed, it is 
 doubtless true that the subjects did not reach as rapid 
 a pace as they did in reading the Courtis Test under 
 normal classroom conditions. Nevertheless, the second 
 record of A. P.'s reading shows that the training 
 had produced a marked decrease in the number of fixa- 
 tion-pauses per line, as well as a decrease in the number 
 of regressive movements from 17 to 9 a decrease of 
 almost 50%. 
 
 Table XXX gives a detailed comparison of the two 
 complete records of A. P.'s reading. Only samples 
 of the complete record were presented in Plates I 
 and II. 
 
 The decrease in the average number of fixation-pauses 
 from 11.7 to 7.8 per line a decrease of almost 4 per 
 line is probably the most striking modification in 
 the motor behavior of the eyes which the photographic 
 records evince. It shows clearly that the experimental 
 training was effective in widening the visual span which 
 functioned in the pupil's reading. The teacher reports 
 that the interest displayed by this subject in training 
 was "intense." At the beginning of the training, his 
 silent reading was accompanied by elaborate lip-move- 
 ment. After about one month's training, the lip 
 movement disappeared. Throat-movements persisted 
 for about two weeks longer, after which time they 
 could no longer be detected by the teacher. The data
 
 PHOTOGRAPHIC RECORDS OF EYE-MOVEMENTS 239 
 
 from the photographic records, the Courtis Test, the 
 individual chart, and the teacher's observations all 
 combine to show that the training of Type II proved 
 very effective in improving the reading ability of this 
 subject to a marked degree. 
 
 TABLE XXX. COMPARISON OP THE EYE-MOVEMENT RECORDS OF A 
 FOURTH-GRADE SUBJECT, A. P. BEFORE AND AFTER TRAINING 
 IN RAPID READING 
 
 LINK 
 
 NUMBER OF 
 PAUSES 
 
 AVERAGEDURA- 
 TION OF PAUSE 
 
 AGGREGATE 
 DURATION or 
 PAUSE 
 
 NUMBER or 
 REGRESSIVE 
 MOVEMENT* 
 
 Record 
 
 Record 
 
 Record 
 
 Record 
 
 I 
 
 II 
 
 I 
 
 II 
 
 I 
 
 II 
 
 I 
 
 II 
 
 1 
 
 8 
 9 
 15 
 13 
 11 
 14 
 12 
 
 9 
 18 
 8 
 12 
 10 
 6 
 7 
 4 
 6 
 
 13.0 
 
 14.4 
 12.3 
 11.7 
 16.0 
 12.5 
 12.9 
 
 10.9 
 12.5 
 15.8 
 12.8 
 13.3 
 10.0 
 11.3 
 17.0 
 15.3 
 
 104 
 130 
 185 
 152 
 176 
 175 
 155 
 
 98 
 100 
 126 
 153 
 133 
 60 
 79 
 68 
 92 
 
 1 
 1 
 
 3 
 3 
 2 
 4 
 3 
 
 
 
 
 1 
 2 
 2 
 
 1 
 
 2 
 
 2 
 
 3 
 
 4 
 
 5 .... 
 
 6 
 
 7 .... 
 
 8 .... 
 
 9 . . 
 
 
 Total. . . . 
 Average . . . 
 
 82 
 11.7 
 
 70 
 
 7.8 
 
 92.8 
 13.1 
 
 118.9 
 13.0 
 
 1077 
 13.1 
 
 909 
 13.0 
 
 17 
 2.4 
 
 9 
 1.0 
 
 Table XXX should be read for line 1 as follows : In the first record 
 there were 8 fixations, in the second, 9. The average duration of the fixa- 
 tion-pauses in the first record was 13 fiftieths of a second, in the second 
 record, 10.9 fiftieths ; the aggregate duration was 104 fiftieths of a second 
 in the first record, 98 fiftieths in the second record. There was one re- 
 gressive movement in the first record, none in the second record.
 
 240 SILENT READING 
 
 PLATE III. SILENT READING BY FOURTH-GRADE SUBJECT, N. C. 
 BEFORE TRAINING 
 
 \ v 3 Y * r 
 
 Tiny Tad -was ft queer j.ittle fello^ irijth ^nljsr 
 
 two legs arjd a sijort tjail. He iros (nearly bl^ok, 
 
 too, and i^uclj smaller tljan most tcjdijoleq in| the 
 
 ', i ^ f J 1 
 
 big pond* I Ee coi|ld hardly tfedt fof his fy-ont/ legs 
 
 t ^ 
 to grcJJ 
 
 "When I h^v then) all, lj | e saidj ."I'll lesjr 
 dirty frater anc| go up jlnto the {orchard. Wfat 
 fun it iri|l]t)e tojhopjand hop a|d|hop. j If only f 
 had a Ii^t|e|tjroth4r to hc> with ^e, I should) ?e 
 
 so 
 
 Only the fixation-positions are marked. The subject was not in the exact focus, 
 the duration of the fixation-pauses could not be determined.
 
 PHOTOGRAPHIC RECORDS OF EYE-MOVEMENTS 241 
 
 PLATE IV. SILENT READING BY FOURTH-GRADE SUBJECT. N. C. 
 AFTER TRAINING 
 
 It hrasn't loKg before tis legd begai to] grow. 
 15 >'i 12. f I* j 
 
 / a. J 4 
 
 He jumped about and kiokkd around until hit 
 iL 83 * 
 
 o. / tL f j 4 
 
 legs grew quite! strong. n l tea. goinfe out oi{ tht 
 \ n jT ? IT ff 
 
 tank ti sie if |l can hop" he said one rtight fhen 
 ? 7 8 13 I* 
 
 / 3 4 a. | f 
 
 he was jukt] six wfeeksj old* 
 
 II 8 i o i + 1 * 
 
 i. / Jl * 
 
 The sun ias hardly up the next mornidg when I a 
 i o i o I * 
 
 i 31 t f 
 
 little toejd jxraped out of the water *nd hopped 
 
 fli ^9 * 
 
 out on) the iank.f He was] very small, (but hone 
 "o I 9 ** l I* 
 
 L / 3 | ,f 
 
 too 4mall tor his little lets that wobbled] un<|er 
 f!j T '< " 
 
 him. It w^is Tiny, the youn toad. 
 
 X indicates that it was impossible to determine with accuracy the duration at UM 
 fixation-pause.
 
 242 SILENT READING 
 
 2. Record of a Rapid Fourth-Grade Reader 
 
 Plate III shows the record of N. C. before training. 
 N. C. is a nine-year-old girl in the fourth grade in the 
 Goodrich School. The teacher reports that her general 
 class average and her reading ability are "excellent." 
 Her rate of reading as measured by the Courtis Test was 
 267 words per minute ; her comprehension score was 75 
 per cent. The test confirmed the teacher's judgment 
 and showed that the subject was, by far, the fastest 
 reader in her class. She is of foreign-born parentage ; 
 her father speaks English but her mother does not. 
 Her advancement through the grades has been normal. 
 The subject manifested great interest in the experi- 
 mental training. 
 
 The record in Plate III shows that N. C. is a superior 
 type of reader for a fourth-grade pupil. The subject 
 averages but 7.1 fixation-pauses per line. The regres- 
 sive movements are not numerous, averaging but 1.1 
 per line. There is a fair degree of regularity and 
 rhythm to the eye-movements, indicating a rapid rate 
 on the part of the reader. Contrast Plate III with 
 Plate I, and the superiority of N. C. over A. P. in prac- 
 tically every aspect of the reading process is made 
 strikingly manifest. 
 
 Plate IV shows that N. C. despite her high initial 
 ability also profited by the training in Type II. There 
 is a decrease in the total number of fixation-pauses for 
 the 8 complete lines in both records from 57 to 41
 
 PHOTOGRAPHIC RECORDS OF EYE-MOVEMENTS 243 
 
 an average decrease of 2 fixation-pauses per line. The 
 perceptual span has been widened so that it grasps 
 a larger number of printed symbols in a single fixation. 
 Thus, lines 2 and 6 are each grasped in 4 fixations, 
 while the best performance on the previous record 
 (Plate III) was the perception of line 2 in 5 fixations. 
 The minimum for the other lines is 6, which is the maxi- 
 mum number of fixations required to read any line in 
 Plate IV. An improvement in regularity and rhythm 
 of eye-movement habits is also clearly noticeable. 
 The Courtis Test likewise showed an increase in speed 
 and a gain of 7 per cent in comprehension. 
 
 TABLE XXXI. COMPARISON OF THE EYE-MOVEMENT RECORDS or A 
 FOURTH-GRADE SUBJECT, N. C., BEFORE AND AFTER TRAINING IN 
 RAPID READING 
 
 Ion 
 
 NUMBER OF 
 PAUSES 
 
 DURATION or 
 PAUSES 
 
 NUMBER or 
 REGRESSIVE 
 MOVEMENTS 
 
 Record 
 
 Record n 
 
 Raooid 
 
 I 
 
 n 
 
 Average 
 
 .\Ker<-cMr 
 
 I 
 
 n 
 
 1 
 
 6 
 5 
 8 
 7 
 6 
 6 
 10 
 9 
 
 6 
 4 
 6 
 5 
 4 
 5 
 6 
 5 
 
 13.2 
 9.3 
 8.7 
 9.2 
 11.0 
 9.4 
 11.2 
 12.0 
 
 79 
 37 
 52 
 46 
 44 
 47 
 67 
 60 
 
 1 
 
 1 
 
 
 
 3 
 3 
 2 
 
 
 
 2 
 2 
 
 1 
 1 
 3 
 1 
 
 
 2 
 
 3 
 
 4 
 
 5 
 
 6 
 
 7 
 
 8 
 
 9 
 
 Total 
 Average .... 
 
 57 
 7.1 
 
 41 
 5.1 
 
 84 
 10.5 
 
 432 
 10.5 
 
 10 
 1.1 
 
 10 
 1.1
 
 244 SILENT READING 
 
 PLATE V. SILENT READING BY FOURTH-GRADE SUBJECT, M. L. 
 BEFORE TRAINING 
 
 fc JL >,</ L f 8 Q ft 
 
 Tiny fadTWllaj dueer little ffellorr hrithj only 
 5 I nffcrt '* '* * ' 3 
 
 ' ^ ? "I f i* ? f ? 
 
 I two (legs and ajsltort tail ) He wjis nearly bljusk, 
 
 ' *4 f f 4 7 f ? 'i* 
 
 toi, and muqh amallef than nbst tadpoles in] the 
 
 ,V TTl 7f A 7f f A f T 
 
 i * y -i f k 1 t 3 
 
 bid pond* I He could hardlV vlit for his I front legs 
 Te r ,'a J f o ft /V fV ? 
 
 | to grow. 
 
 "wken I hiive thlm All , Phi said/' I'll liave this 
 
 2o "4 JT V*| ft 5 7 
 
 i/j^r fc PT ? _j 
 
 dirty initer 1 arid io up k.nto[tne orchard* fliat 
 ^ rt /7 ft F * V x> 
 
 /JX^rfc 7 ^/, 
 
 fun ii will (be; to iop fend hop land hod , If onlj/ I), 
 
 /7 /| iVf V f t* ' i^ * 
 
 /ijfjr 1^7? ?'* 
 
 had la liltje bfothir to hop 4- th V. I ohouM be I 
 T 
 
 1 ' 
 
 so hWr. 1 
 f Co
 
 PHOTOGRAPHIC RECORDS OF EYE-MOVEMENTS 245 
 
 PLATE VI. SILENT READING BY FOURTH-GRADE SUBJECT, M. L. 
 AFTER TRAINING 
 
 It wasn't |ong|beforj >{is l^gs fbfgan/ijol 
 He jumped about a|id kicked) aroufd uijtil jhie 
 legs iref duite strong. "I aja goipg <|ut ]pn the 
 
 bank to| see| if I oaji hfp" he sifid one ni|ht jrhen 
 
 ' ^ 
 
 he wa just six neekjs old* 
 
 ' ^ * I f i t 
 
 The su(i fa h|rdly up f(he next/ morning/ when a 
 
 f V 3 ^ ^ 
 
 little t jafl junpef out of t^e watef and h(>pped 
 
 out on (the banlj 4 **! T * r > r ' " all P** f 01 * 
 
 ' V 4 * f ? 
 
 fcnall ] f or ]ii littl^ legal that wobbled unfer 
 
 too 
 
 / 2 i If 
 
 Ida* It |as ^-inj, the ' 
 
 Only the fixation-positions are marked. The subject wa not in the erct focus, 
 the duration of the pauses could not be determined.
 
 246 SILENT READING 
 
 Table XXXI presents the figures for a detailed com- 
 parison of the two records of N. C. 
 
 3. Record of a Fourth-Grade Subject of Medium Ability 
 in Silent Reading 
 
 Plate V shows the record of M. L. before training. 
 M. L. is a nine-year-old girl in the fourth grade in the 
 Hendricks School. Her reading ability is "fairly good." 
 She stands apparently midway between the two pre- 
 vious fourth-grade subjects. Her rate in the Courtis 
 Test is 136 words per minute ; her comprehension is 
 good. Her parents are native-born. She manifested 
 very great interest in the training Type I. 
 
 The record in Plate V shows a type of fourth-grade 
 reader that might be classed as "fair." It is not 
 nearly so poor as the initial record of A. P. (Plate I), 
 nor nearly so good as the initial record of N. C. (Plate 
 III). There is an average of 9.5 fixations per line, with 
 an average duration of 11.3 fiftieths of a second per 
 fixation-pause. Considering the number of fixations 
 per line, the number of regressive movements are not 
 numerous, averaging but 1 per line. Compare this 
 record (Plate V) with either of N. C. 's records, and the 
 excessively large number of fixation-pauses in M. L.'s 
 reading becomes plainly evident. Improvement in 
 reading rate would obviously seem to lie in the direction 
 of a reduction of the number of fixations per line. 
 There is room for improvement also in the irregular, 
 unrhythmical character of the eye-movement habits.
 
 PHOTOGRAPHIC RECORDS OF EYE-MOVEMENTS 247 
 
 The record in Plate VI shows that this is precisely 
 what happened in the previous record. The reduction 
 in the number of fixation-pauses is rather striking. 
 From a total of 76 fixations for the 8 full lines on record 
 V the number for the 8 full lines on record VI is reduced 
 to 58. This means an average reduction of 2.3 fixation- 
 pauses per line. With the exception of line 1 on record 
 VI, in which the subject evidently suffered from a poor 
 start, the maximum number of fixations for any line is 
 8, which is the minimum number of fixations for any 
 line in the subject's previous record. 
 
 Table XXXII presents the data contained in Plates 
 V and VI in tabular form. 
 
 TABLE XXXII. COMPARISON OF THE EYE-MOVEMENT RECORDS or A 
 FOURTH-GRADE SUBJECT, M. L. BEFORE AND AFTER TRAINING 
 IN RAPID READING 
 
 LOT 
 
 NUMBER or 
 PAUSES 
 
 DURATION OF 
 PAUSES 
 
 NUMBER or REGRES- 
 SIVE MOVEMENTS 
 
 Record 
 
 RecordI 
 
 Record 
 
 I 
 
 II 
 
 Average 
 
 Aggregate 
 
 I 
 
 II 
 
 1 
 
 10 
 9 
 10 
 9 
 8 
 10 
 10 
 10 
 
 11 
 7 
 8 
 7 
 6 
 5 
 7 
 7 
 
 8.5 
 8.7 
 12.1 
 10.3 
 15.5 
 10.9 
 13.8 
 11.1 
 
 85 
 78 
 121 
 93 
 124 
 109 
 138 
 111 
 
 2 
 1 
 
 
 
 1 
 
 
 2 
 1 
 1 
 
 2 
 
 1 
 
 
 
 
 
 
 2 
 
 3 
 
 4 
 
 5 
 
 6 
 
 7 
 
 8 
 
 Total 
 
 76 
 9.5 
 
 58 
 7.2 
 
 90.9 
 11.3 
 
 859 
 11.3 
 
 8 
 1.0 
 
 3 
 0.4 
 
 Average ....
 
 248 SILENT READING 
 
 PLATE VII. SILENT READING BY FIFTH-GRADE SUBJECT, G. B. 
 BEFORE TRAINING 
 
 H l ' f ' ? 
 
 a^shopjJere )were (anong Jthe worst 
 
 eniaiee of thq early] settlers ofJNebraska. 
 
 !?r ii a tV 
 
 Theit Ihotoee nejre of the (high tlailis and 
 
 7rJTF a iV n * 
 
 noil ijhe hljlB a| tlj foot of the kreai moLi- 
 
 v / J r 1 7 
 
 Ulna 4 ih M^> Hire hey Ulved and r^sed 
 Or iV f' Ts > T 7 
 
 their 
 
 i.'7 fr 
 
 In dry seajsona theqe were moije ohiliref and 
 
 lejse fdod at hole. Than they) assembled and 
 rf ?x |T ;W 7 7< 
 
 1 ' * ? f f i* r 
 
 flew iway [in g^eatf nrarms ^oj the] east and to 
 J h 7i H V i-' i" 
 
 J * f 
 
 the efutH. They tjraveled huldrede of miles. 
 
 * t 1 ! i^ i'r 
 
 on olfal we^n nooplightj nights 
 
 they trkre]|Bd ah.1 righi. More loftpn thW settled 
 i* ie ii ft i / a 
 
 do*n|latl il the afterncjon and fed, and then 
 4 > Tt to (V 
 
 oontinuep theltj waty | the next day. 
 Ulf t iTU- 
 
 X indicates that it was impossible to determine with accuracy the duration of the 
 fixation-pause.
 
 PHOTOGRAPHIC RECORDS OF EYE-MOVEMENTS 249 
 
 "LATE V1IL SILENT READING BY FIFTH-GRADE SUBJECT. G. B. 
 AFTER TRAINING 
 
 / 1 f 3 4 k 
 
 All the ccjrn -was eat 4n Jin |a| single day. 
 x. JLO X " r /o 
 
 TThere [sofrnfields stjood at sunrise nothing re* 
 J^U. 7 I T If Y 
 
 I - ^ 4 7 * ** f ? 
 
 mained at I night but] stumps of sttlks smarming 
 l'7^A > ^ V F IT ie 
 
 / x j 1 t f 
 
 with hLngi4 hopplers strlggline for! the last 
 n 5 10 is ft n 
 
 / i. j 74^S'4 
 
 biije. Thejy striked the] |ard^n| patches bare. 
 
 i / j <f f 
 
 They eiia4red ereat holes in the fugs and car- 
 ?< V /V / T " 
 
 pets put out to savje favorite plants. Tpe buds 
 andTfruilb of trees weri consumed. [They fol- 
 
 V ^ ro r]t & 
 
 lo-wedlthe plotatotes arid Anio^a into the ground. 
 
 *| - I ^ I * % ft 
 
 X indicates that it was impossible to determine with accuracy the duration of the 
 fixation-pause.
 
 250 
 
 SILENT READING 
 
 The reduction in the number of regressive movements 
 is not less noticeable than the reduction in the number 
 of fixation-pauses. From a total of 8, the number of 
 regressive movements drops to 3. Record VI shows 
 likewise the development of a habit of greater uni- 
 formity and rhythm in the character of the eye-move- 
 ments. The results of the Courtis Test, in which the 
 subject reached a rate of 284 words per minute, with 
 very good comprehension, harmonize with the story 
 told by the photographic record a "higher order of 
 eye-movement habits" was built up by the training 
 to replace the lower order habits of eye-movements 
 with which the subject began. 
 
 TABLE XXXIII. COMPARISON OP THE EYE-MOVEMENT RECORDS OF 
 A FIFTH-GRADE SUBJECT, G. B. BEFORE AND AFTER TRAINING 
 
 LINE 
 
 NUMBER or 
 PAUSES 
 
 AVERAGE 
 DURATION OF 
 PAUSE 
 
 AGGREGATE 
 DURATION OF 
 PAUSE 
 
 NUMBER OF 
 REGRESSIVE 
 MOVEMENTS 
 
 Record 
 
 Record 
 
 Record 
 
 Record 
 
 I 
 
 II 
 
 I 
 
 n 
 
 I 
 
 n 
 
 I 
 
 n 
 
 1 .... 
 
 7 
 
 6 
 
 
 12.1 
 
 
 73 
 
 2 
 
 1 
 
 2 .... 
 
 4 
 
 6 
 
 19.5 
 
 16.3 
 
 78 
 
 98 
 
 
 
 1 
 
 3 .... 
 
 8 
 
 9 
 
 13.4 
 
 10.7 
 
 107 
 
 96 
 
 1 
 
 3 
 
 4 .... 
 
 8 
 
 6 
 
 9.4 
 
 13.3 
 
 75 
 
 80 
 
 2 
 
 1 
 
 5 .... 
 
 7 
 
 8 
 
 11.9 
 
 12.9 
 
 83 
 
 103 
 
 2 
 
 1 
 
 6 .... 
 
 5 
 
 5 
 
 10.2 
 
 19.0 
 
 51 
 
 95 
 
 1 
 
 1 
 
 7 .... 
 
 6 
 
 3 
 
 13.3 
 
 14.3 
 
 80 
 
 43 
 
 2 
 
 
 
 8 .... 
 
 8 
 
 5 
 
 11.8 
 
 14.2 
 
 94 
 
 71 
 
 3 
 
 1 
 
 9 .... 
 
 5 
 
 6 
 
 15.6 
 
 19.2 
 
 78 
 
 115 
 
 1 
 
 1 
 
 10 .... 
 
 7 
 
 
 10.1 
 
 
 71 
 
 
 2 
 
 
 11 .... 
 
 7 
 
 
 12.1 
 
 
 85 
 
 
 2 
 
 
 12 .... 
 
 5 
 
 
 12.0 
 
 
 60 
 
 
 1 
 
 
 Total . . . 
 
 77 
 
 54 
 
 138.3 
 
 132 
 
 862 
 
 774 
 
 19 
 
 10 
 
 Average . . . 
 
 6.4 
 
 6.0 
 
 12.3 
 
 14.3 
 
 12.3 
 
 14.3 
 
 1.6 
 
 1.1
 
 PHOTOGRAPHIC RECORDS OF EYE-MOVEMENTS 251 
 
 II. EYE-MOVEMENT RECORDS OF FIFTH-GRADE 
 SUBJECTS BEFORE AND AFTER TRAINING 
 
 1. Record of Fifth-Grade Subject of Medium Ability 
 in Silent Reading 
 
 Plate VII shows the record of G. B. before training. 
 G. B. is a ten-year-old girl in the fifth grade in the 
 Manierre School. Her general class average was re- 
 ported as 92 per cent by the teacher. Her reading 
 ability was classified as "medium." She is of foreign- 
 born parentage, but some English is spoken hi the 
 home. She is described as doing but little reading 
 outside of school. Her rate in the Courtis Reading 
 Test was 150 words per minute ; her comprehension 
 was good. 
 
 The record hi Plate VII reflects a medium grade of 
 reading ability. The record shows a fair degree of 
 uniformity in the eye-movement habits. The average 
 number of the fixations per line for the whole record is 
 6.4, which is considerably better than the average of 
 9.3 reported by C. T. Gray (19) for his fourth-grade 
 subjects. The regressive movements are too frequent, 
 however, averaging 1.6 per line. The average duration 
 of the fixation-pauses is 12.3, which is less than the aver- 
 age of 14.3 reported by Gray for his fifth-grade subjects. 
 
 The record in Plate VIII shows somewhat of a reduc- 
 tion in the average number of the fixation-pauses per 
 line 6.4 to 6.0. The average duration of the pause 
 has increased, however, from 12.3 fiftieths of a second
 
 252 SILENT READING 
 
 to 14.3 fiftieths. The average number of regressive 
 movements per line is reduced from 1.6 to 1.1. On the 
 whole, Plate VIII shows scarcely any appreciable im- 
 provement in the reading habits of the subject G. B. 
 It is quite possible that the subject did not do her 
 best when reading before the camera. As was pointed 
 out before, the situation was quite an unusual one and 
 may have exercised somewhat of a disturbing influence 
 on this girl subject. 
 
 After the training in Type I the subject scored a 
 rate of 278 words per minute on the Courtis Reading 
 Test, with a comprehension mark of 94 per cent. This 
 shows a considerable gain over her first score of 150 
 words per minute on the Courtis Test a gain that is 
 scarcely reflected at all in her photographic records. 
 Lines 6, 7, and 8 towards the end of the record give evi- 
 dence, however, that the subject is capable of rapid 
 reading. The record for line 7 is much superior to the 
 record of any line in the subject's previous performance. 
 
 2. Record of a Very Rapid Fifth-Grade Reader 
 
 Plate IX shows the record of L. C. before training. 
 L. C. is a nine-year-old boy in the fifth grade in the 
 Manierre School. The teacher reports that his general 
 class average is 98 per cent ; his reading ability is "ex- 
 cellent." He is described by the teacher as a very 
 rapid reader who is able to retain what he reads. He 
 is "fond of reading and reads extensively outside of 
 school." His lips do not move in silent reading; he
 
 PHOTOGRAPHIC RECORDS OF EYE-MOVEMENTS 253 
 
 gives all the evidence of splendid concentration. His 
 parents are native-born. The school records and 
 the teacher's report indicate that L. C. is a "gifted" 
 pupil. His progress through the grades has been 
 accelerated. 
 
 The record on Plate IX shows that L. C. is an excep- 
 tionally able reader. The average number of fixation- 
 pauses per line is but 5, which is practically but one- 
 half of the average of 9.3 reported by Gray for his 
 fifth-grade subjects. There are but 2 regressive move- 
 ments in the whole record. The regularity and rhythm 
 of eye-movement habits displayed by this record easily 
 excel those of any of the previous records. The 
 average duration of the fixation-pause is 12.8 as com- 
 pared with the average of 14.3 reported by Gray. 
 The record shows very clearly how L. C. was able to 
 score 270 words per minute on the Courtis Reading 
 Test and secure a mark of 100 per cent in comprehension 
 by answering 47 questions correctly in the 5 minutes 
 allowed. 
 
 The record in Plate IX might be thought to represent 
 almost the physiological limit of reading prowess for a 
 nine-year-old fifth-grade subject. Plate X shows, 
 however, that L. C. improved very noticeably as a 
 result of the training. There has been a reduction in 
 the total number of fixation-pauses for the first 6 full 
 lines in each record from 30 to 21 ; the average number 
 per line has been lowered from 5.0 to 3.5. The average 
 duration of the pauses has mounted but slightly -
 
 254 SILENT READING 
 
 from 12.4 to 12.8. There occurs but one regressive 
 movement in the second record. 
 
 PLATE IX. SILENT READING BY FIFTH-GRADE SUBJECT, L. C. 
 BEFORE TRAINING 
 
 / a- 3 ^ 
 
 I In dry seasons there jrere more! children and 
 
 j ! 3 rs ri it 
 
 less] food at hone* I Then they assembled and 
 i T 3 10 \3 11 71 
 
 Fle^r ejway in great atrarms to the] east and to 
 ! r t i li ft 
 
 ', f j * f 
 
 the south* They traveled hundreds of mles* 
 io ^ 17 iV a 
 
 / X J 4 
 
 ^onetimes onj clear narmjmoonlignt nights 
 ri fs 7i 7* * 
 
 f ' 4 -"f s f 
 
 they tijavjeled all nidht. Ubre iften they setltled 
 > V 73 Vl I T 1 To 
 
 late I in the afbernoon and I fed. and i than 
 
 r* Ir 75 i 1 / i 
 
 continued their tray the next day* 
 ft n i 
 
 There is a balance, a swing, a rhythm to the eye- 
 movements of this subject which no previous record
 
 PHOTOGRAPHIC RECORDS OP EYE-MOVEMENTS 255 
 
 has evidenced. The grasping of line 3 in two fixations 
 is rather striking. It shows the wide visual span 
 
 PLATE X. SILENT READING BY FIFTH-GRADE SUBJECT, L. C. 
 AFTER TRAINING 
 
 "Where I cornfields stood! at sunrise npthing re- 
 
 ' * 3 
 
 mained at nkght but stunrds of stalks fcwarming 
 
 a /V ' 
 
 / a 
 
 with hurjgry hoppers struggling for the last 
 
 bite. They dkripped the garo]en patofies bare* 
 re /* 
 
 Thei gnawed great I holes in the Irugs andf oar- 
 pi, n t i* 
 
 petls put but to save favorite plajrrts. Thefbude 
 ?3 7 \* 5 
 
 / Q. 4 _ _ 
 
 and truit offerees were co^uaed. Tley fol- 
 ^ fr I* 
 
 which the subject utilizes in his reading. The average 
 of 3.5 fixation-pauses per line is lower than any of the 
 averages reported by Gray. In the hierarchy of eye-
 
 256 
 
 SILENT READING 
 
 movement habits which these photographic records 
 have evidenced, Plate X stands as the reflex of the high- 
 est and most efficient type. There is a coordination, 
 a harmonization of the constituent neurological pro- 
 cesses of the reading complex, evidenced in Plate X, 
 which stamps the reader as one of exceptional ability. 
 
 TABLE XXXIV. COMPARISON OF THE EYE-MOVEMENT RECORDS OF A 
 FIFTH-GRADE SUBJECT, L. C. BEFORE AND AFTER TRAINING 
 IN RAPID READING 
 
 Lnm 
 
 NUMBER OP 
 
 PAUSES 
 
 AVERAGE DURA- 
 TION op PAUSE 
 
 AGGREGATE DU- 
 RATION OP 
 PAUSE 
 
 NUMBER OF 
 PROGRESSIVE 
 MOVEMENTS 
 
 Record 
 
 Record 
 
 Record 
 
 Record 
 
 I 
 
 II 
 
 I 
 
 II 
 
 I 
 
 II 
 
 I 
 
 II 
 
 1 
 
 5 
 4 
 5 
 5 
 6 
 5 
 
 4 
 3 
 2 
 3 
 
 4 
 5 
 
 12.8 
 12.7 
 12.6 
 13.8 
 10.8 
 12.0 
 
 10.8 
 14.7 
 16.5 
 15.3 
 13.0 
 10.2 
 
 64 
 51 
 63 
 69 
 65 
 60 
 
 43 
 44 
 33 
 46 
 52 
 51 
 
 1 
 
 
 
 
 
 1 
 
 
 
 
 
 
 
 
 
 1 
 
 2 
 
 3 
 
 4 
 
 5 
 
 6 
 
 Total . . . 
 Average . . 
 
 30 
 5 
 
 21 
 3.5 
 
 74.7 
 12.4 
 
 80.5 
 12.8 
 
 372 
 12.4 
 
 269 
 12.8 
 
 2 
 0.3 
 
 1 
 
 0.2 
 
 This conclusion is further corroborated by L. C.'s 
 record in the Courtis Test given at the end of the train- 
 ing. His rate was 395 words per minute, as against 
 270 on his first record. His index of comprehension 
 was 97 per cent. He answered 66 questions out of a 
 possible 70. Only 2 of his answers were incorrect. 
 His record of 64 correct answers shows considerable
 
 PHOTOGRAPHIC RECORDS OF EYE-MOVEMENTS 257 
 
 improvement over his previous record of 47. The in- 
 crease in speed of L. C.'s reading as shown by the eye- 
 movement records is from 48 lines per minute to 66 - 
 an increase of 18 lines, or 38 per cent. This harmonizes 
 fairly closely with the increase as shown by the Courtis 
 Test of 125 words per minute, or 46 per cent. All in 
 all, Plate X gives a record of the highest type of eye- 
 movement habits photographed in this investigation. 
 To facilitate a closer comparison of the two records of 
 L. C., the data on each performance are presented in tab- 
 ular form (Table XXXIV). 
 
 III. EYE-MOVEMENT RECORD OF A SLOW SEVENTH- 
 GRADE READER 
 
 Plate XI shows the record of M. C. before training. 
 M. C. is a twelve-year-old girl in the seventh grade in 
 the Hendricks School. At the beginning of the train- 
 ing she was a slow, plodding reader. Her comprehen- 
 sion was good. Her rate on the Courtis Test was 170 
 words per minute. Her index of comprehension was 
 100 per cent, though she succeeded in answering only 
 28 questions as contrasted with the 47 correctly an- 
 swered by the fifth-grade subject, L. C., before training. 
 Her parents are native-born. The subject manifested 
 great interest in the training. 
 
 The record hi Plate XI would seem to indicate that 
 the subject was of the over-careful type painstak- 
 ingly fixating practically every word some of them 
 twice. For a seventh-grade pupil such simple words
 
 258 SILENT READING 
 
 as are contained in lines 1 and 6 should not require 11 
 and 12 fixations respectively. The average number of 
 fixation-pauses per line for the complete record is 8.4 as 
 compared with an average of 7.7 reported by Gray 
 
 PLATE XI. SILENT READING OF SEVENTH-GRADE SUBJECT, M. C. 
 BEFORE TRAINING 
 
 t '. M 9 I* F T ? '\ l l 
 
 there ik hb hnbre interesting study to sarinc 
 
 IT 10 7s- Vfiij i i* '* 7 
 
 i { 4 * f 7 
 
 1 architects than tttat of the growth of modern ships 
 
 t I T O fy Fa. ^ iV F 
 
 4 f f f ' . ? 
 
 their barliestJ form* Ancient/ ships of war 
 
 ' ' ^' 
 
 a. f a 4 f j 
 
 and of bommeree ebually interest tttem; but as 
 
 !r JV i^. TJ 13 V r* 
 
 a. ' 3 > f if T f 
 
 thiy stjudy th4 douiLptures and iritingsl o? the 
 
 10 T T y T 'f f 
 
 X /. f,7 3 fc f f '0 ? V 'f 
 
 akwAentsJ|th!ey!fin4 relo or dsi of I Tranships f s> out- 
 
 C 7'T I ^^ "* ' ' '" '* 
 
 " 
 
 ntUberingl sliips of commerce. 
 if i s iW 13 
 
 for his 7 seventh-grade subjects. The total number 
 of regressive movements for the 12 complete lines is 
 30 ; the average number per line is 2.5. The number 
 is excessive. It indicates an over-cautiousness, a metic-
 
 >are 
 
 re- 
 
 PHOTOGRAPHIC RECORDS OF EYE-MOVEMENTS 259 
 
 I'LATE XII. SILENT READING BY SEVENTH-GRADE SUBJECT, M. C. 
 AFTER TRAINING 
 
 Tnefe is jiq quesijicjn nor? ^hat the ships of (the 
 ancient a ni^de extended voyages urgeq by p 
 
 / -4 ^f i f 
 
 alone* Aj thousand Aarsraen ncjre sometimes 
 
 quired to) nan tjhe sireeps pesides a prev of] five 
 
 41 4 S" ^ ^ * 
 
 n^red soldi^rsjand aailfrs* Written de^orip- 
 
 tion^ giv^ us splendid piotures of fljeets of these 
 
 4 i- T M t C 7 
 
 ancient phips ^>|ri|ig fwi^tly along pe villa-dotted 
 
 '. t ^ 
 
 shores of Greece, orj majestically sweeping into 
 
 4{ frf V t f 7 
 
 olrror-like harbor, and wijbh sounding trpm4> 
 
 4' ^l ^ f * 
 
 utijjg the setting of the lo'jr, jre stern kun 
 
 Only the fixation-positions are marked. It waa impossible to determine the duration 
 of the fixation-pauses on this film.
 
 260 SILENT READING 
 
 ulousness, a dawdling over the words which would ap- 
 pear to be quite unnecessary for a subject whose com- 
 prehension is as good as M. C.'s. It is not unlikely 
 the result of a slow, plodding, leisurely reading habit, 
 which as Huey says, was probably "set and hardened 
 in the days of listless poring over uninteresting tasks, 
 or in imitation of the slow reading aloud, which was so 
 usually going on." An average of 2.5 regressive move- 
 ments to a line shows very clearly that the subject has 
 not formed efficient habits of regular, systematic, 
 rhythmical eye-movements, but suffers from ineffi- 
 cient, wasteful, time-consuming, repetitive ocular motor 
 reactions, which have become habitual. 
 
 The record in Plate XII shows a considerable im- 
 provement. The average number of fixations per 
 line has been reduced from 8.4 to 6.5. With the ex- 
 ception of line 1, no line required more than 8 fixations. 
 The functioning of a wider perceptual span is clearly 
 evident. Line 8 is grasped in 3 fixations, while the 
 smallest number on any complete line in the subject's 
 previous record was 6. The regressive movements 
 dropped from an average of 2.5 to 1.6 per line, most of 
 which occurred in the initial fixation in each line. 
 
 The subject is evidently finding some of her numerous 
 fixation-pauses unnecessary and is setting up more 
 efficient habits of regular rhythmical eye-movements. 
 The improvement effected in the reading habits of this 
 subject by training in Type I, which is reflected in the 
 photographic records, is further shown by the data
 
 PHOTOGRAPHIC RECORDS OF EYE-MOVEMENTS 261 
 
 from the Courtis Test administered at the end of the 
 experiment. The subject scored a rate of 368 words 
 per minute as contrasted with her previous score of 
 170. Her index of comprehension is 96 per cent, but 
 she answered 67 questions, of which 65 were correct as 
 compared with her previous record of 28 correct answers. 
 The data from the two eye-movement records of 
 subject M. C. are presented in Table XXXV. 
 
 TABLE XXXV. COMPARISON OF THE EYE-MOVEMENT RECORDS OF A 
 SEVENTH-GRADE SUBJECT, M. C. BEFORE AND AFTER TRAINING 
 
 lam 
 
 NUMBER OF 
 PAUSES 
 
 DURATION OF 
 PAUSES 
 
 NUMBER OF RE- 
 GRESSIVE MOVE- 
 HE NTS 
 
 Record 
 
 Record I 
 
 Record 
 
 I 
 
 II 
 
 Average 
 
 Aggregate 
 
 I 
 
 . II 
 
 1 
 
 11 
 
 7 
 7 
 7 
 8 
 12 
 6 
 9 
 7 
 6 
 9 
 12 
 
 10 
 6 
 5 
 6 
 7 
 6 
 8 
 3 
 8 
 6 
 
 13.2 
 10.6 
 10.9 
 10.4 
 10.1 
 12.0 
 11.7 
 8.1 
 8.0 
 10.8 
 9.2 
 8.8 
 
 145 
 74 
 76 
 73 
 81 
 144 
 70 
 73 
 56 
 65 
 83 
 
 4 
 1 
 1 
 1 
 3 
 5 
 
 3 
 2 
 1 
 4 
 5 
 
 3 
 H 
 1 
 
 1 
 2 
 1 
 3 
 
 3 
 1 
 
 2 
 
 3 
 
 4 
 
 5 
 
 6 
 
 7 
 
 8 
 
 9 
 
 10 
 
 11 
 
 12 
 
 Total 
 
 101 
 8.4 
 
 65 
 6.5 
 
 123.8 
 10.3 
 
 1037 
 10.3 
 
 30 
 2.5 
 
 16 
 1.6 
 
 Average .... 
 
 A few other photographic records were taken of ele- 
 mentary pupils, but it is thought that the records shown
 
 262 SILENT READING 
 
 above reflect sufficiently well the important modifica- 
 tions in the eye-movement habits, i.e. in the number 
 and duration of fixation-pauses, number of regressive 
 movements, and regularity of the eye-movements, 
 which were effected by the experimental training. 
 
 EYE-MOVEMENT RECORDS OF AN ADULT IN ORAL AND 
 SILENT READING 
 
 Plate XIII shows the record of an adult, J. A., in silent 
 reading. The subject is a graduate student at the 
 University of Illinois who has read rather extensively. 
 He had received some training in Type I. It was 
 thought that a comparison of the eye-movement record 
 of a trained adult reader with the records of elementary 
 school pupils would yield some interesting results. 
 The record seems to reflect a greater degree of maturity 
 than any of the previous records, with the single ex- 
 ception of the record of L. C. after training, as shown 
 in Plate X. With the exception of the latter record, 
 the average number of fixation-pauses per line, 4.3 in 
 J. A.'s reading, is less than the average achieved by any 
 of the grade pupils. The average duration of the fixa- 
 tion-pauses in J. A.'s record is 11.3, which is 1.5 less 
 than the average duration of the fixations in L. C.'s 
 record. There is a greater regularity and more of a 
 rhythmical character to the ocular motor reaction in 
 J. A.'s record than appears in most of the previous 
 records. The average number of regressive movements 
 per line is very small but 0.4. The perceptual span
 
 PHOTOGRAPHIC RECORDS OF EYE-MOVEMENTS 263 
 
 is wider than that reflected in most of the records of the 
 elementary pupils. 
 
 TABLE XXXVI. COMPARISON OF EYE-MOVEMENT RECORDS OF SILENT 
 AND ORAL READING BY ADULT SUBJECT, J. A. 
 
 LINE 
 
 NUMBER or 
 PAUSES 
 
 AVERAGE 
 DURATION OF 
 PAUSE 
 
 AGGREGATE 
 DURATION OF 
 PAUSES 
 
 NUMBER OF 
 REGRESSIVE 
 MOVEMENTS 
 
 
 Oral 
 
 Silent 
 
 Oral 
 
 Silent 
 
 Oral 
 
 Silent 
 
 Oral 
 
 Silent 
 
 1 .... 
 
 6 
 
 3 
 
 13.5 
 
 14.3 
 
 81 
 
 43 
 
 1 
 
 
 
 2 .... 
 
 8 
 
 4 
 
 13.2 
 
 10.7 
 
 106 
 
 43 
 
 2 
 
 1 
 
 3 .... 
 
 6 
 
 5 
 
 14.8 
 
 9.8 
 
 89 
 
 49 
 
 1 
 
 1 
 
 4 .... 
 
 5 
 
 4 
 
 15.2 
 
 11.0 
 
 76 
 
 
 
 
 
 
 5 .... 
 
 7 
 
 4 
 
 12.6 
 
 10.0 
 
 88 
 
 40 
 
 1 
 
 
 
 6 .... 
 
 6 
 
 4 
 
 11.7 
 
 10.6 
 
 70 
 
 43 
 
 
 
 
 
 7 .... 
 
 7 
 
 3 
 
 16.9 
 
 12.3 
 
 128 
 
 37 
 
 1 
 
 
 
 8 .... 
 
 5 
 
 6 
 
 21.0 
 
 10.7 
 
 105 
 
 64 
 
 
 
 1 
 
 9 .... 
 
 5 
 
 5 
 
 13.2 
 
 10.8 
 
 76 
 
 54 
 
 
 
 1 
 
 10 .... 
 
 5 
 
 4 
 
 16.4 
 
 14.0 
 
 82 
 
 56 
 
 1 
 
 
 
 11 .... 
 
 7 
 
 3 
 
 14.8 
 
 15.3 
 
 94 
 
 46 
 
 1 
 
 
 
 12 .... 
 
 7 
 
 4 
 
 12.4 
 
 11.5 
 
 87 
 
 46 
 
 1 
 
 
 
 13 .... 
 
 5 
 
 7 
 
 14.4 
 
 10.0 
 
 72 
 
 70 
 
 
 
 1 
 
 Total . . . 
 
 79 
 
 56 
 
 190.1 
 
 151.0 
 
 1154 
 
 591 
 
 9 
 
 5 
 
 Average . . 
 
 6.1 
 
 4.3 
 
 14.6 
 
 11.3 
 
 14.6 
 
 11.3 
 
 0.7 
 
 0.4 
 
 After the subject J. A. had read the selection silently 
 before the camera, he was requested to read it orally 
 to see how the two records would compare. Though 
 the subject had the advantage of being more familiar 
 with the passage when he was called to read orally, 
 yet the record in Plate XIV shows that oral reading 
 occasions much more cumbrous and awkward ocular 
 motor adjustments than its silent prototype. The 
 average number of fixation-pauses per line immediately
 
 264 SILENT READING 
 
 rises from 4. 3 to 6.1, showing that the perceptual process 
 is forced to wait upon the slower process of vocalization. 
 
 PLATE XIII. SILENT READING BY ADULT SUBJECT, J. A. 
 '. % ^ 
 
 There is no metre interesting study to marine 
 IT IS rj 
 
 ihat of the growth of modern fhipi 
 > $ IV 
 
 architects than thkt of the growth of modern 
 
 M 
 
 frok their earliest form*] Ancient ships bf far 
 
 n 
 
 . 
 
 and of oonaerok equally interest them; ijut as 
 
 * ' t ' If * 
 
 / a. j 4 
 
 tney stuojr the sculptures and -writings of (the 
 
 / J. If 
 
 antientsj they find records of -warships J f ar out- 
 
 t b i /V 
 
 ing ships of oonnrce. 
 
 4 i j t 
 
 ill 
 
 Among ancient nations, the Greeks and Ho- 
 
 * '5 7? 
 
 It is significant also that the average duration of the 
 pauses mounts from 11.3 to 14.6. - The total number of
 
 PHOTOGRAPHIC RECORDS OF EYE-MOVEMENTS 265 
 
 regressive movements jumps from 5 to 9. The uni- 
 formity of the ocular motor reaction habits is disturbed 
 by the attempt to synchronize the perceptual process 
 
 PLATE XIV. ORAL READING BY ADULT SUBJECT, J. A. 
 
 / o. J * 4 f 
 
 I TJhere is no more) interesting sti|idy to JiarinJB 
 
 /. * 4 S <i I f 7 
 
 ardhitects thkn tViat of thi grjowth of modern shipb 
 V 13 to If W L* i l o 10 
 
 frorA their elrliest jform. Ancieftt ships of war I 
 3 l\t fc 1% if " 
 
 l ^ 4 f 
 
 and olf conmeroA equally interest fthem; but as 
 7 /V ?7 X3 l/ 
 
 ' * 4 f f t 1 
 
 they itT the sculptures arjd writings of thb 
 
 \ y /V flF <V iv iv 
 
 ( * 4 f t 
 
 anciejtts, they find records of watships far jnit- 
 /i n ji /v /* /J 
 
 nui-jbering ^hips of commerce. 
 *" '& 
 
 Among anoieVit natiiis, ihe GrAeks anfl ^o- I 
 /V ? T8 Ji '* " 
 
 with the slower and more unwieldy process of articu- 
 lation with the latter's more complete dependence 
 upon an elaborate physiological mechanism. In short,
 
 266 
 
 SILENT READING 
 
 the record of silent reading shows a marked and un- 
 mistakable superiority over the oral reading in prac- 
 tically every important phase of the eye-movements, 
 i.e. the number of fixations, the length of fixations, the 
 number of regressive movemen s, and regularity of 
 eye-movement habits. It shows clearly the physi- 
 ological basis for the unmistakable superiority of silent 
 over oral reading, both in point of time and energy 
 expended. 
 
 The data from the complete records of J. A.'s silent 
 and oral reading are thrown into tabular form to facili- 
 tate comparison of the two records. (Table XXXVI.) 
 
 TABLE XXXVII. COMPARISON OF THE NUMBER OF FIXATION-PAUSES 
 AND REGRESSIVE MOVEMENTS OF FIVE ELEMENTARY SCHOOL 
 PUPILS BEFORE AND AFTER TRAINING IN RAPID READING 
 
 POTILB 
 
 AVERAGE NUMBER 
 or PAUSES 
 
 AVERAGE NUMBER OF 
 REGRESSIVE MOVEMENTS 
 
 Record 
 
 Record 
 
 I 
 
 II 
 
 I 
 
 II 
 
 L C . 
 
 5.0 
 8.4 
 11.7 
 7.1 
 9.5 
 
 3.5 
 6.5 
 7.8 
 5.1 
 
 7.2 
 
 0.3 
 
 2.5 
 2.4 
 1.1 
 1.0 
 
 0.2 
 1.6 
 1.0 
 1.1 
 0.4 
 
 M C 
 
 A P 
 
 N. C . . 
 
 M. L 
 
 Total 
 
 41.7 
 8.3 
 
 30.1 
 6.0 
 
 7.3 
 1.5 
 
 4.3 
 
 0.8 
 
 Average 
 
 
 Table XXXVII epitomizes the effect of the experi- 
 mental training upon the number of the fixation-pauses
 
 PHOTOGRAPHIC RECORDS OF EYE-MOVEMENTS 267 
 
 and the number of regressive movements in the reading 
 of five elementary school pupils. The records of their 
 eye-movements have been presented in this chapter. 
 The subject G. B. is omitted from the table because the 
 improvement, if any, in this case is less clearly discerni- 
 ble than in the others. In an endeavor to determine 
 whether the increase in speed of reading is due to the 
 lessening of the number of pauses, or to the shortening 
 of the duration of the pauses, his record, showing no 
 increase in speed, obviously could not be used. The 
 table shows that the acquisition of a habit of rapid 
 reading such as these subjects achieved, results in a 
 noticeable reduction in the number of fixation-pauses. 
 The average number of fixations per line for the five 
 subjects before training is 8.3 ; after training in rapid 
 reading, it is reduced to 6.0 a decrease of 2.3 per line. 
 The reduction in the number of regressive movements 
 is also quite marked. From an average of 1.5 per line 
 for the first reading it drops to the small average of 0.8. 
 An effort was also made to compare the rate of the 
 interfixation and the interlinear movements of the 
 eyes in the first and second reading records of three of 
 the subjects. The rate was measured in <r ( io * 66 of a 
 second). The apparatus was scarcely delicate enough, 
 however, to measure <r with perfect accuracy. The 
 average time per line consumed by the interfixation 
 movements in the first reading record was 122<r as com- 
 pared with 101<r in the record taken after training. 
 The average for the interlinear movements in the first
 
 268 SILENT READING 
 
 record was 42cr as compared with 57cr for the second 
 record. The results do not point to any significant 
 change in the rate of the interfixation or interlinear 
 movements. 
 
 Unfortunately the duration of the fixation-pauses for 
 both the first and second records could be determined 
 for but three of the six subjects whose records are 
 shown. In two of these, the duration of the pauses is 
 slightly longer in the second record ; in the other one it 
 remains approximately the same. It is very probable 
 that a marked reduction in the number of fixations 
 results in an increase, more or less, in the average 
 duration of the fixation. While the average duration 
 of the fixation becomes slightly larger, the aggregate 
 duration of the pauses per line is much less than before 
 the reduction in the number of fixations. The records 
 of L. C. in Table XXXIV and of A. P. in Table XXX 
 are good illustrations of this. Though the average 
 duration per pause for L. C.'s second record is 12.8 as 
 compared with 12.4 for his first record, yet the aggre- 
 gate duration of the pauses per line in the second record 
 is less for every line than in the first record. The re- 
 duction in the number of fixations means that the 
 fixations must embrace a wider area. The grasping of 
 more material in a single fixation would naturally seem 
 to be accompanied by a slight increase in the duration 
 of the pause. 
 
 Considering the limited number of subjects whose 
 eye-movements were photographed in this investiga-
 
 PHOTOGRAPHIC RECORDS OF EYE-MOVEMENTS 269 
 
 tion, the conclusions which an analysis of these records 
 would clearly seem to justify are : 
 
 1. A habit of speed in silent reading acquired 
 through persistent training is accompanied physiologi- 
 cally chiefly by a reduction in the number of fixation- 
 pauses. 
 
 2. The types of training utilized hi this investiga- 
 tion effected a noticeable reduction in the number of 
 regressive movements as shown by the eye-movements 
 which were photographed.
 
 CHAPTER XI 
 SUMMARY OF CONCLUSIONS 
 
 IT was not the purpose of this experiment to isolate 
 and weigh separately the effect of different single factors 
 upon the rate in silent reading. That is one of the 
 interesting and fascinating problems in the experimental 
 psychology of reading which yet awaits the hand of the 
 investigator. The purpose of this investigation was to 
 determine the effectiveness of a group of factors synthe- 
 sized into an organized method or type of training upon 
 the rate in silent reading. The query of practical 
 pedagogical consequence back of the investigation was : 
 How can speed in silent reading be developed without 
 impairing the comprehension of the matter read ? 
 
 Practically all the conclusions issuing from the pres- 
 ent investigation are contained in the two previous 
 chapters. For the sake of convenience, some of the 
 more important generalizations are expressed here. 
 
 This investigation has shown that rate in silent 
 reading may be accelerated to a marked degree by train- 
 ing in rapid silent reading. This was demonstrated 
 for all the grades investigated, namely, grades three 
 
 270
 
 SUMMARY OF CONCLUSIONS 271 
 
 to eight. In the course of two months' training in rapid 
 reading, an average gain for the experimental pupils 
 in the five grades of 110 words per minute was effected, 
 as compared with an average gain of 64 words per 
 minute for the control pupils. Reducing the average 
 gain in number of words read per minute to a percent- 
 age basis, the amount of improvement for the experi- 
 mental pupils in all the grades was 56 per cent, as 
 against 25 per cent for the controls an average 
 superiority of 31 per cent for the experimental pupils. 
 Inasmuch as the control pupils were taught by the 
 same teacher in the same classroom with the experi- 
 mental, they were not entirely unaffected by the 
 experimental training. Their increase in speed was 
 consequently greater than that of pupils receiving only 
 the conventional training in reading in classrooms 
 where no experimental training was in evidence. This 
 is shown by a comparison of their final averages in rate 
 with the present norms. Consequently, the superior- 
 ity of the average gain in rate of the experimental pupils 
 over the control pupils great as it is does not 
 fully reflect the degree of superiority of the rates of 
 the experimental pupils over the present norms for the 
 different grades. This is shown by a direct comparison 
 of the final averages of the experimental pupils with the 
 norms reported by Courtis, Starch, Gray, and others, 
 for the different grades. This comparison shows also 
 that the present reading rates of pupils in the different 
 grades are unnecessarily slow and inefficient. A com-
 
 272 SILENT READING 
 
 paratively brief period of training can increase them 
 more than 50 per cent. 
 
 Concomitant with the marked increase in speed there 
 resulted a slight improvement in the accuracy of the 
 comprehension. The average gain of the experimentals 
 was 0.9 per cent as against a loss of 0.7 per cent for the 
 controls a final average superiority in gain over the 
 controls of 1.6 per cent in comprehension. This is the 
 improvement as measured by the Courtis Index, which 
 is chiefly an index of accuracy. As measured by the 
 number of questions correctly answered, the improve- 
 ment in comprehension is notably greater an average 
 increase of 12.5 questions correctly answered, or 29 
 per cent. In fact the number of questions correctly 
 answered by the experimental pupils in each grade after 
 training is greater than the norms for the total number 
 of questions attempted, whether answered correctly or 
 wrongly, as reported by Courtis. 
 
 This increase indicates furthermore a persistence of 
 the improvement effected hi rate in different situations 
 an different mental attitudes, as in reading simply 
 to " get the gist," and in reading to answer questions. 
 
 The eye-movement records show that the improve- 
 ment is effected physiologically chiefly by a lessening 
 of the number of fixations rather than by a shortening 
 of the average duration of the fixations. The im- 
 provement is also accompanied by a decrease in the 
 number of regressive movements and by the "setting 
 up" of habits of regular rhythmical eye-movements.
 
 SUMMARY OF CONCLUSIONS 273 
 
 The pedagogical implication of this finding is that 
 short exposure exercises, in which the amount of ma- 
 terial exposed is gradually increased, rather than the 
 amount of exposure time being decreased, tend to de- 
 velop speed in reading. It demonstrates that the 
 directions in the types of training, outlined in this 
 study, to "see more at a glance," etc., are well based. 
 It shows that the development of speed in silent reading 
 is, in reality, reducible to the more effective utilization 
 of the perceptual span in reading.
 
 BIBLIOGRAPHY 
 
 1. Holmes, Henry W. "Time Distribution by Subjects and Grades in 
 
 Representative Cities," Fourteenth Yearbook of the National 
 Society for the Study of Education, part 1, pp. 21-27. Chicago: 
 University of Chicago Press, 1915. 
 
 2. Charters, Werrett W. Methods of Teaching. Chicago : Row, Pe- 
 
 terson & Co., 1909. 225 pp. (Chapter 13.) 
 
 3. Cattell, James M. "Reactions and Perceptions," Essays Philo- 
 
 sophical and Psychological in Honor of William James. New 
 
 York: Longmans, Green & Co., 1908. 610 pp. (pp. 569-84.) 
 Cattell, James M. " On the Relations of Time and Space in Vision," 
 
 Psychological Review, 7 : 325-43, July, 1900. 
 Cattell, James M. "The Inertia of the Eye and Brain," Brain, 
 
 8 : 295-312, October, 1885. 
 Cattell, James M. " Ueber die Zeit der Erkennung und Benennung 
 
 von Schriftzeichen, Bildern und Farben," Philosophische Studien, 
 
 2 : 635-50, 1885. 
 
 4. Goldscheider, A. and Muller, R. F. "Zur Physiologic und Pathol- 
 
 ogie des Lesens," Zeitschrifl fur klinische Medicin, 23 : 131, 1893. 
 
 5. Erdmann, B. and Dodge, R. Psychologische Untersuchungen iiber 
 
 das Les n auf experimenteller Grundlage. Halle: M. Niemeyer, 
 1898. 360 pp. 
 
 Erdmann, B. and Dodge, R. "Zur Erlauterung unserer tachistos- 
 kopischenVereuche,"Zeitec/iri/V/ur Psychologic, 22 : 241-67, 1900. 
 
 6. Huey, Edmund B. The Psychology and Pedagogy of Reading. New 
 
 York: The Macmillan Company, 1913. 469 pp. 
 
 7. Schmidt, William A. "An Experimental Study in the Psychology 
 
 of Reading." (Supplementary Educational Monograph, v. 1, 
 no. 2, April, 1917.) Chicago : University of Chicago Press, 1917. 
 126 pp. 
 
 8. Javal, Emile. "Sur la physiologic de la lecture," Annales rf'Ocw/w- 
 
 tique, 1878-1879. 
 
 275
 
 276 BIBLIOGRAPHY 
 
 9. Lamare. " Des Mouvements des yeux pendant la lecture," Comptes 
 rendus de la Societe frangaise d'Ophtalmologie, pp. 354 ff., 1893. 
 
 10. Ahrens, August. Die Bewcgung der Augen beim Schreiben, Rostock, 
 
 1891. 
 
 11. Delabarre, E. B. "A Method of Recording Eye Movements," 
 
 American Journal of Psychology, 9 : 572-74, July, 1898. 
 
 12. Huey, E. B. "Preliminary Experiments in the Physiology and 
 
 Psychology of Reading," American Journal of Psychology, 9 : 575- 
 86, July, 1898. 
 
 Huey, E. B. "On the Psychology and Physiology of Reading," 
 American Journal of Psychology, 11 : 293-302, 12 : 292-312, 1900- 
 1901. 
 
 13. Dodge, Raymond. "An Experimental Study of Visual Fixation," 
 
 Psychological Review, Monograph Supplement, v. 8. 
 Cf. also Dodge, R. and Cline, T. S. "The Angle Velocity of Eye- 
 Movements," Psychological Review, 8: 145-57, March, 1901. 
 
 14. Dearborn, Walter F. "The Psychology of Reading," Columbia 
 
 University Contributions to Philosophy and Psychology, v. 14, no. 1, 
 1906. 
 
 15. Judd, C. H., McAllister, C. N., and Steele, W. M. "Introduction 
 
 to a Series of Studies of Eye Movements by Means of Kinetoscopic 
 Photographs," Psychological Review, Monograph Supplement, 
 v. 7, no. 1, pp. 1-16, March, 1905. 
 
 Judd, C. H. "Photographic Records of Convergence and Diver- 
 gence," Psychological Review, Monograph Supplement, v. 8, no. 3, 
 pp. 370-423, June, 1907. 
 
 16. Holt, E. B. "Eye Movement and Central Anaesthesia," Psycho- 
 
 logical Review, Monograph Supplement, v. 4, January, 1903. 
 
 17. Dearborn, Walter F. "The Psychology of Reading," Cohimbia 
 
 University Contributions to Philosophy and Psychology, v. 14, 
 no. 1, 1906. 
 
 18. Schmidt, William A. An Experimental Study in the Psychology of 
 
 Reading. Chicago : University of Chicago Press, 1917. 126 pp. 
 (p. 79.) 
 
 19. Gray, Clarence T. Types of Reading Ability as Exhibited through 
 
 Tests and Laboratory Experiments. (Supplementary Educational 
 Monograph, v. 1, no. 5, August, 1917.) Chicago: University 
 of Chicago Press, 1917. 196 pp. 
 
 20. Gray, William S. "Principles of Method in Teaching Reading as 
 
 derived from Scientific Investigation," Eighteenth Yearbook of
 
 BIBLIOGRAPHY 277 
 
 the National Society for the Study of Education, part 2, pp. 26-51 
 (pp. 40, 41). Bloomington, Illinois: Public School Publishing 
 Co., 1919. 
 
 21. Judd, Charles H. Reading: Its Nature and Development. (Supple- 
 
 mentary Educational Monograph, v. 2, no. 4, July, 1918.) 
 Chicago : University of Chicago Press, 1918. 192 pp. 
 
 22. Pintner, Rudolf. "Oral and Silent Reading of Fourth Grade 
 
 Pupils," Journal of Educational Psychology, 4 : 333-37, June, 1913. 
 
 23. Oberholtzer, E. E. "Testing the Efficiency of Reading in the 
 
 Grades," Elementary School Journal, 15 : 313-22, February, 1915. 
 
 24. Mead, Cyrus D. "Silent versus Oral Reading with One Hundred 
 
 Sixth Grade Children," Journal of Educational Psychology, 
 6 : 346-48, June, 1915. 
 
 25. Pintner, Rudolf and Gilliland, A. R. "Oral and Silent Reading," 
 
 Journal of Educational Psychology, 7 : 201-12, April, 1916. 
 
 26. Quantz, J. O. "Problems in the Psychology of Reading," Psycho- 
 
 logical Review, Monograph Supplement, v. 2, no. 1, December, 
 1897. 
 
 27. Hendricks, Eldo L. A Preliminary Report of a Study of Reading. 
 
 Boston: Silver, Burdett & Co., 1911. 32 pp. 
 
 28. Judd, Charles H. Measuring the Work of the Public Schools. 
 
 Cleveland, O. : The Survey Committee of the Cleveland Founda- 
 tion, 1916, 290 pp. 
 
 29. Mann, Horace. "Second Annual Report of the Secretary of the 
 
 Board of Education, 1838," Life and Works of Horace Mann, 
 2 : 531-32. Boston : Lee and Shepard, 1891. 5 v. 
 
 30. Judd, Charles H. Reading: Its Nature and Development. Chicago : 
 
 University of Chicago Press, 1918. 192 pp. (p. 184.) 
 
 31. Thorndike, Edward L. "Reading as Reasoning: A Study of Mis- 
 
 takes in Paragraph Reading," Journal of Educational Psychology, 
 8 : 323-32, June, 1917. (p. 332.) 
 
 32. Judd, Charles H. Reading: Its Nature and Development. Chicago : 
 
 University of Chicago Press, 1918. 192 pp. (pp. 150, 151, 155, 
 146.) 
 
 33. Bowden, Josephine. Learning to Read. (Master's thesis, Uni- 
 
 versity of Chicago, 1911.) 
 
 34. Abell, Adelaide M. "Rapid Reading : Advantages and Methods," 
 
 Educational Review, 8 : 283-86. October, 1894. 
 
 35. Dodge, Raymond. Die Motorischen Wortvorstettungen. Halle: 
 
 N. Niemeyer, 1896. 78 pp. (p. 65.)
 
 278 BIBLIOGRAPHY 
 
 36. Whipple, Guy M. "The Effect of Practice upon the Range of 
 
 Visual Attention and of Visual Apprehension," Journal of Edu- 
 cational Psychology, 1 : 249-62, March, 1910. 
 
 37. Peters, Charles C. "The Influence of Speed Drills upon the Rate 
 
 and Effectiveness of Silent Reading," Journal of Educational 
 Psychology, 8 : 350-66, June, 1917. 
 
 38. Gray, William S. Studies of Elementary School Reading through 
 
 Standardized Tests. (Supplementary Educational Monograph, 
 v. 1, no. 1.) Chicago: University of Chicago Press, 1917. 
 
 39. Fordyce, Charles. "Testing the Efficiency in Reading," Addresses 
 
 and Proceedings of the National Education Association, 55 : 818-21, 
 July 7-14, 1917. (p. 821.) 
 
 40. Whipple, G. M. and Curtis, Josephine N. " Preliminary Investiga- 
 
 tion of Skimming in Reading," Journal of Educational Psychology, 
 8 : 333^9, June, 1917. 
 
 41. Gray, William S. "Reading." Survey of the St. Louis Public 
 
 Schools, v. 2. St. Louis : Board of Education, 1917. 
 
 42. Ruediger, William C. "Field of Distinct Vision," Columbia Con- 
 
 tributions to Philosophy and Psychology, v. 16, no. 1, 1907. 
 
 43. Messmer, Oskar. "Zur Psychologic des Lesens bei Kinder und 
 
 Erwachsenen," Archiv fiir die Gesamte Psychologic, Band II, 
 Heft 2-3, pp. 190-298 Vorlesungen, B. 3S. December, 1903. 
 
 44. Ribot, Th. "Les Mouvements et leur importance psychologique," 
 
 Revue Philosophique, 8 : 371-86, October, 1879. 
 
 45. Bain, Alexander. The Senses and the Intellect. 1868. p. 336. 
 
 46. Egger, Victor. La Parole interieure. Paris: G. Bailliere et Cie. 
 
 1881. 
 
 47. Ballet, Gilbert. Le Langage interieur et les diverses formes de 
 
 I'aphasie. Paris : F. Alcan, 1886. 
 
 48. Pintner, Rudolf. "Inner Speech during Silent Reading," Psy- 
 
 chological Review, 20 : 129-53, January, 1915. 
 
 49. Strieker, S. Studien uber die Sprachvorstellungen. Vienna : 
 
 Ebendas, 1880. 106 pp. 
 
 50. Stumpf, Carl. Tonpsychologie. Leipzig : Hirzel, 1883. 427 pp. 
 
 (p. 154 ff .) 
 
 51. Paulhan, Fr. "Le Langage interieur et la pensee," Revue Philoso- 
 
 phique, 21 : 26-58, January, 1886. 
 
 52. Baldwin, James. Mental Development in the Child and the Race. 
 
 1895. p. 442. 
 
 53. Bastian, Henry C. Brain as an Organ of Mind. 1891. p. 595 ff.
 
 BIBLIOGRAPHY 279 
 
 54. Collins, J. The Genesis and Dissolution of the Faculty of Speech, a 
 
 Clinical and Psychological Study of Aphasia. New York : 
 The Macmillan Company, 1898. 432 pp. (p. 195 ff.) 
 
 55. Dodge, Raymond. "Die motorischen Wortvorstellungen," Psy- 
 
 chological Review, 4 : 326-27, May, 1897. 
 
 56. Bawden, H. H. "A Study of Lapses," Psychological Review, Mono- 
 
 graph Supplement, v. 3, no. 4, 1900. p. 154 ff. 
 
 57. Curtis, H. S. "Automatic Movement of the Larynx," American 
 
 Journal of Psychology, 1 1 : 237-39, January, 1900. 
 
 58. Courten, H. C. "Involuntary Movements of the Tongue," Studies 
 
 from the Yale Psychological Laboratory, v. 10, pp. 93-95, 1902. 
 
 59. Hansen, F. C. C. and Lehmann, Alfred. "Ueber unwillkiirliches 
 
 Flustern," Wundt's Philosopische Studien, 11:471-530, 1895. 
 
 60. Secor, W. B. "Visual Reading: A Study in Mental Imagery," 
 
 American Journal of Psychology, 11 : 225-36, January, 1900. 
 
 61. Dodge, Raymond. "The Psychology of Reading," Cyclopedia of 
 
 Education, 1917. v. 5, p. 117. 
 
 62. Pfeiffer, Ludwig. Ueber VorsteUungstypen. Leipzig : O. Nemnich, 
 
 1907. 129 pp. 
 
 Pfeiffer, Ludwig. Experimentelle Untersuchungen uber qualitative 
 Arbeitstypen. Leipzig : O. Nemnich, 1907. 281 pp. 
 
 63. Segal, J. "Ueber den Reproduktionstypus und das Reproduktion 
 
 von Vorstellungen," Archiv fur die Gesamte Psychologic, 12 : 124- 
 235, 1908. 
 
 64. Lobsien, Marx. Ueber den V or stettungsty pus der Schulkinder. 1911. 
 
 457 pp 
 
 65. Cohn, Jonas. "Experimentelle Untersuchungen iiber das ziisam- 
 
 menwirken des akustisch-motorischen und des visuellen Gedacht- 
 nisses," Zeitschrift fur Psychologic, 15: 161-83, 1897. 
 
 66. Baade, Walter. Experimentelle und kritische Beitrage zur Frage nach 
 
 der sekundaren Wirkungen des Unterrichts. Leipzig : O. Nemnich, 
 1907. 124 pp. 
 
 67. Stoning, Gustav. Mental Pathology. Translated by Thomas 
 
 Loveday. London, 1907. 298 pp. 
 
 68. Netschajeff, A. "Die Entwickelung des Gedachtnisses bei Schul 
 
 kindern," Zeitschrift fur Psychologic, 24 : 321-51, 1900. 
 Netschajeff, A. Ueber Memorieren und Gedachtnis. Berlin : Reuther 
 and Reikard, 1903. 39 pp. 
 
 69. Meumann, Ernest. The Psychology of Learning. Translated from 
 
 3d ed. of "The Economy and Technique of Learning," by John
 
 280 BIBLIOGRAPHY 
 
 W. Baird. New York: D. Appleton & Co., 1913. 393 pp. 
 (p. 193.) 
 
 70. Offner, Max. Mental Fatigue. Translated from the German by G. 
 
 M. Whipple. Boston : Warwick and York, 1911. 133 pp, 
 
 71. Rivers, W. H. and Kraepelin, E. "Ueber Ermiidung und Erholung," 
 
 Psychologische Arbeiten, 1 : 627-78, 1896. 
 
 72. Courtis, S. A. "Standard Tests in English," Elementary School 
 
 Journal, 14 : 374-92, April, 1914.
 
 APPENDIX 1 
 
 LIST OF BOOKS FOR SILENT READING 
 
 The following list of books has been selected by Mr. James P. Hosic 
 of the Chicago Normal School from a much larger list which he had 
 prepared for use in certain Chicago schools designated for the time 
 as "English Centers." 
 
 The present list is prepared especially for use in certain schools of 
 the state where a method of improving silent reading has been tested out 
 by Dr. John A. O'Brien under the direction of Dr. B. R. Buckingham of 
 the University of Illinois. The last few titles under each grade are from 
 the Supplementary Reading List adopted February 13, 1919, by the 
 Chicago Board of Education. 
 
 SECOND GRADE 
 
 Dutton In Field and Pasture American Book Co. 
 
 Schaffner Sam ; or, Our Cat Tales Atkinson, Mentzer <fe Co. 
 
 THIRD GRADE 
 
 Baldwin Fifty Famous Stories American Book Co. 
 
 Bigham Merry Animal Tales Little, Brown & Co. 
 
 Craik Adventures of a Brownie Educational Publishing Co. 
 
 1 The shift of emphasis from oral to silent reading necessitates the 
 use of a larger amount of reading material in all the grades from the 
 second to the eighth. These lists are appended in the hope that they 
 may assist teachers to solve the consequent problem of finding a suffi- 
 cient amount of material suitable for either basic or supplementary 
 reading. Many of the readers cited in these lists are published by 
 several firmK. Where no publisher is .specified, the book as a rule can 
 he procured from practically any publishing company. 
 
 281
 
 282 
 
 APPENDIX 
 
 Gale Achilles and Hector 
 
 McMurry Story of Robinson Crusoe 
 Mulock Little Lame Prince 
 Sindelar Father Thrift and His 
 
 Animal Friends 
 Sindelar Nixy-Bunny in Far Away 
 
 Lands 
 Stevenson Children's Classics in 
 
 Dramatic Form, Book 2 
 Scudder Fables and Folk Stories 
 Schmidt A Dramatic Reader, 
 
 Book3 
 Holbrook A Dramatic Reader for 
 
 Lower Grades 
 
 Burgess old Mother West Wind 
 Segur g tory o f a Donkey 
 
 Public School Publishing Co. 
 Public School Publishing Co. 
 D. C. Heath & Co. 
 
 Beckley-Cardy Co. 
 Beckley-Cardy Co. 
 
 Houghton Mifflin Co. 
 Houghton Mifflin Co. 
 
 B. D. Berry 
 
 American Book Co. 
 Little, Brown & Co. 
 D. C. Heath & Co. 
 
 FOURTH GRADE 
 
 Aanrud 
 Baker and 
 
 Thorndike 
 Bigham 
 Carrol 
 Collodli 
 Ghosh 
 Harris 
 Large 
 Mirick and 
 
 Holmes 
 Schwartz 
 
 Schwartz 
 
 Segur 
 
 Wesselhoeft 
 
 Wiggin 
 
 Warren 
 
 Warren 
 Cook 
 
 Lisbeth Longrock 
 
 Everyday Classics, Book 4 
 Fanciful Flower Tales 
 Alice in Wonderland 
 Adventures of Pinocchio 
 Wonders of the Jungle 
 Little Mr. Thimblefmger 
 A Visit to the Farm 
 Home Life around the 
 
 World 
 
 Grasshopper Green's Gar- 
 den 
 
 Wilderness Babies 
 Sophie 
 
 Jack the Fire Dog 
 Story of Patsy 
 King Arthur and His 
 
 Knights 
 Robin Hood and His 
 
 Merry Men 
 Story of Ulysses 
 
 Ginn & Co. 
 
 The Macmillan Company 
 Little, Brown & Co. 
 Rand-McNally & Co. 
 Ginn & Co. 
 D. C. Heath & Co. 
 Houghton Mifflin Co. 
 The Macmillan Company 
 
 Houghton Mifflin Co. 
 Little, Brown & Co. 
 
 D. C. Heath & Co. 
 Little, Brown & Co. 
 Houghton Mifflin Co. 
 
 Rand-McNally & Co. 
 
 Rand-McNally & Co. 
 Public School Publishing Co.
 
 APPENDIX 
 
 283 
 
 Alcott 
 Austin 
 Bachman 
 
 Baker and 
 Thorndike 
 
 Dodge 
 
 Hyde 
 
 Kipling 
 
 Long 
 
 Seton 
 
 Gale 
 
 Stevenson 
 
 Baldwin 
 Harris 
 Spyri 
 Skinner 
 
 FIFTH GRADE 
 
 Little Women Little, Brown & Co. 
 
 The Basket Woman Houghton Mifflin Co. 
 
 Great Inventors and Their 
 Inventions American Book Co. 
 
 Everyday Classics, Book 5 
 
 Hans Brinker 
 Favorite Greek Myths 
 Just So Stories 
 Wood Folk at School 
 Krag and Johnny Bear 
 Achilles and Hector 
 Children's Classics in Dra- 
 matic Form, Book 4 
 Golden Fleece 
 Little Mr. Thimblefinger 
 Moni, The Goat Boy 
 Tales and Plays of Robin 
 Hood 
 
 The Macmillan Com- 
 pany 
 
 Ginn & Co. 
 D. C. Heath & Co. 
 Century Co. 
 Ginn & Co. 
 Charles Scribner's Sons 
 Rand-McNaUy & Co. 
 
 Houghton Mifflin Co. 
 American Book Co. 
 Houghton Mifflin Co. 
 Ginn & Co. 
 
 American Book Co. 
 
 Baldwin 
 
 Carter 
 Faris 
 Hill 
 Lagerlof 
 
 Lang 
 Moore 
 
 Roosevelt 
 
 Wiggin and Smith 
 
 Spyri 
 
 Mabie 
 
 Baldwin 
 
 SIXTH GRADE 
 
 American Book of Golden 
 
 Deeds 
 
 Stories of Brave Dogs 
 Makers of Our History 
 Fighting a Fire 
 Wonderful Adventures of 
 
 Nils 
 
 Story of Joan of Arc 
 Story of Christopher 
 
 Columbus 
 
 Stories of the Great West 
 The Posy Ring 
 Heidi 
 
 Norse Stories 
 Old Greek Stories 
 
 American Book Co. 
 Century Co. 
 Ginn & Co. 
 Century Co. 
 
 Grosset & Dunlap 
 
 Houghton Mifflin Co. 
 Century Co. 
 Doubleday, Page & Co. 
 Ginn & Co. 
 Rand-McNally & Co. 
 American Book Co.
 
 284 APPENDIX 
 
 Church Story of the Iliad The Macmillan Com- 
 
 pany 
 
 Church Story of the Odyssey The Macmillan Com- 
 
 pany 
 Baldwin Thirty More Famous 
 
 Stories Retold American Book Co. 
 
 SEVENTH GRADE 
 
 Aldrich Story of a Bad Boy Houghton Mifflin Co. 
 
 Baker and Thorndike Everyday Classics, Book 7 The Macmillan Com- 
 pany 
 
 Barrie Peter and Wendy 
 
 Brooks The Story of King Arthur Perm Publishing Co. 
 
 Fitz-Hugh Boy's Book of Scouts T. Y. Crowell Co. 
 
 Hale Man without a Country Little, Brown & Co. 
 
 Jordan The Story of Matka 
 
 Mace Life of Lincoln Rand-McNally & Co. 
 
 Pierce Great Inventors and Dis- 
 
 coverers Charles E. Merrill Co. 
 
 Sanford and Owen Modern Americans Laurel Book Co. 
 
 Stoddard Red Mustang Harper Bros. 
 
 Seton Trail of the Sandhill Stag 
 
 Weed Bird Life Stories Rand-McNally & Co. 
 
 Williams Some Successful Ameri- 
 
 cans Ginn & Co. 
 
 Stevenson Children's Classics hi 
 
 Dramatic Form, Book 5 Houghton Mifflia Co. 
 
 Montgomery Heroic Ballads Ginn & Co. 
 
 EIGHTH GRADE 
 
 Baker and Thonxdike Everyday Classics, Book 8 The Macmillan Com- 
 pany 
 
 Baldwin Fifty Famous Rides 
 
 Chapin Story of the Rhinegold 
 
 Cooper Last of the Mohicans 
 
 D'Amicis Heart of a Boy Rand-McNally & Co. 
 
 Eastman An Indian Boyhood 
 
 Fabre Insect Adventures Dodd, Mead & Co. 
 
 Grenfel Adrift on an Ice Pan Houghton Mifflin Co. 
 
 Mace Life of Washington Rand-McNally & Co.
 
 APPENDIX 
 
 285 
 
 Madden Emmy Lou 
 
 Rice Mrs. Wiggs of the Cab- 
 
 bage Patch 
 
 Scott Talisman 
 
 Thompson Scouting with Daniel 
 
 Boone 
 Western Frontier Stories 
 
 Wiggin and Smith Golden Numbers 
 
 Tennyson and others Poems of Knightly Adven- 
 ture 
 
 Doubleday , Page & Co . 
 Century Co. 
 Doubleday, Page & Co. 
 
 Newson & Co. 
 
 Titles for Silent Reading Furnished by the Conference at the Uni- 
 versity of Illinois, April 12, 1919. 
 
 Blaisdell 
 
 Craik 
 
 Lucia 
 
 Lucia 
 
 Lucia 
 
 Varney 
 
 Bigham 
 
 Bigham 
 
 Bryce 
 
 Haliburton 
 
 Holbrook 
 
 Burgess 
 
 Burgess 
 Burgess 
 Burgess 
 Burgess 
 
 McManus and Har- 
 ren 
 
 THIKD GRADE 
 
 Bunny Rabbit's Diary Little, Brown & Co. 
 Bow-wow and Mew-mew Charles E. Merrill Co. 
 Peter and Polly in Spring American Book Co. 
 Peter and Polly in Summer American Book Co. 
 Peter and Polly in Winter American Book Co. 
 Robin Reader, first Reader Charles Scribner's Sons 
 
 FOURTH GRADE 
 
 Fanciful Flower Tales 
 
 Merry Animal Tales 
 
 That's Why Stories 
 
 Third Reader 
 
 Book of Nature Myths 
 
 Mother West Wind 
 "How" Stories 
 
 Mother West Wind 
 "Why" Stories 
 
 Mother West Wind's 
 Children 
 
 Mother West Wind's Ani- 
 mal Friends 
 
 Mother West Wind's 
 Neighbors 
 
 Natural Method Third 
 Reader 
 
 Little, Brown & Co. 
 Little, Brown & Co. 
 Newson & Co. 
 D. C. Heath & Co. 
 Houghton Mifflin Co. 
 
 Little, Brown & Co. 
 Little, Brown & Co. 
 Little, Brown & Co. 
 Little, Brown & Co. 
 Little, Brown & Co. 
 Charles Scribner's Sons
 
 286 
 
 APPENDIX 
 
 Schwartz 
 Scudder 
 
 Smith 
 
 Treadwell and Free 
 
 Eggleston 
 
 Lorenzini 
 
 Murray 
 
 Scottenfels 
 
 Hancock 
 
 Spyri 
 
 Thorne-Thomsen 
 
 Thorne-Thomsen 
 
 Greene 
 
 Maeterlinck 
 
 Mallory 
 
 Swift 
 
 Grasshopper Green's Gar- 
 den Little, Brown & Co. 
 
 Book of Fables and Folk 
 Stories Houghton Mifflin Co. 
 
 Eskimo Stories Rand-McNally & Co. 
 
 Reading-Literature, First 
 
 and Second Readers Row, Peterson & Co. 
 
 FIFTH GRADE 
 
 Stories of Great Americans 
 
 for Little Americans American Book Co. 
 Pinocchio Ginn & Co. 
 
 Story Land Little, Brown & Co. 
 
 Stories of the Nibelungen 
 
 for Young People A. Flanagan & Co. 
 
 SIXTH GRADE 
 
 Children of History Little, Brown & Co. 
 
 Heidi Ginn & Co. 
 
 Birch and the Star and 
 
 Other Stories Row, Peterson & Co. 
 
 East o' the Sun and West 
 
 o' the Moon 
 
 SEVENTH GRADE 
 
 Legends of King Arthur 
 
 and His Court 
 Blue Bird for Children 
 King Arthur Stories 
 Gulliver's Travels 
 
 Ginn & Co. 
 Silver, Burdett & Co. 
 Houghton Mifflin Co. 
 Harper & Brothers
 
 INDEX 
 
 Abell, A. M., 47 f. 
 
 Ability to grasp meaning of contents, 
 
 66. 
 Adult reader's photographic record, 
 
 263 ff. 
 
 Ahrens, A., 6. 
 
 Anaesthesia, visual central, 9. 
 Articulation, atrophy of, 48 ; vestigial, 
 
 107 ; unnecessary, 115 ff. 
 Assimilation, 66 ff. 
 Association, mental, 112 f. 
 
 Bain, A., 102. 
 
 Baldwin, J., 104 ff. 
 
 Ballet, G., 102 ff. 
 
 Bastian, H. C., 105. 
 
 Bawden, H. H.. 106. 
 
 Bowden, J., 41. 
 
 Brown, W., 203 ff. 
 
 Buckingham, B. R., 91. 
 
 Bureau of Educational Research, 149. 
 
 Bushwell, G. T., 18. 
 
 Cattell, J. M., 3. 
 
 Chart, class, 75 ff. ; growth in speed 
 shown by, 214 ff. 
 
 Charters, W. W., 3. 
 
 Collins, J., 105. 
 
 Comprehension, effect of training in 
 speed, upon, 197 f. 
 
 Concentration of attention, 63 ff. 
 
 Conference, reading, 160. 
 
 Consciousness, pragmatic theory of, 
 102. 
 
 Control, method of, 153 : group, 153. 
 
 Correlation between rate and com- 
 prehension, 14 ff., 19 ff. 
 
 Courten, H. C., 106. 
 
 Courtis Test, 166. 
 
 Curriculum, criterion for, 20 f. 
 
 Curtis, H. S., 106. 
 
 Curtis, J. N., 60. 
 
 Dearborn, W. F., 7, 12, 79. 
 Decrease of vocalization in silent 
 
 reading, 43 ff., 100 ff. 
 Delabarre, E. B., 7. 
 Directions to pupils, 98 f. 
 Discrimination reaction time, 224 f. 
 Dodge, R., 4 ff. 
 Dynamogenesis, theory of, 110, 113. 
 
 Egger, V., 102 ff. 
 
 Emphasis, in teaching reading, 22 ff. 
 
 Erdmann, B., 4 ff. 
 
 Eye-movement, habits of, 54. 
 
 Eyes, apparatus to measure move- 
 ment of, 5 ff. 
 
 Eyes, movements of, during reading, 
 5 ff . ; interfixation movements of, 
 8 ff . ; fusion, due to velocity of, 9 ; 
 return sweep of, 9. 
 
 Field of vision, 127. 
 Fixation-pauses, 10 ff . ; number of, 
 
 1 1 ; duration of, 1 1 f . 
 Fordyce, C., 55. 
 
 Gary, norms of, 203. 
 
 Goldscheider, A., 3. 
 
 Graph, individual, 74 f. ; how to 
 
 make, 158 ff. 
 
 Gray, C. T., 15, 40. 44 ff., 50 ff.. 60 f. 
 Gray, W. S., 15, 54. 
 
 Habits of eye-movement. 54 ff. 
 
 Hansen, F. C., 107. 
 
 History of reading, as school subject, 
 
 28 f. 
 
 Holmes, H. W., 1. 
 Holt, E. B., 9. 
 Hosic, J. F., 163. 
 Huey, E. B., 5 ff., 38, 48. 
 
 Idea, kinsrathetic, 106. 
 
 Imagery, visual type of, 78 f. ; audi- 
 
 287
 
 288 
 
 INDEX 
 
 tory, 79 ; motor, 79 ; relation to 
 speed of reading, 133 ff. ; "surro- 
 gate," 138; mixed, 137 ff. ; 
 dominant, 136 ; vocal-motor, 137. 
 
 Index of comprehension, 168 ff. ; 
 measured by number of questions 
 answered, 210 ff. 
 
 Innervation, sensations of, 102. 
 
 Interest, immediate, 65 ; mediate, 76. 
 
 Introspection, in reading, 64, 67, 103, 
 117. 
 
 James, W., 85. 
 
 Javal, E., 6. 
 
 Judd, C. H., 7 ff., 16, 30. 
 
 Kraepelin, E., 164. 
 
 Lamare, 6. 
 
 Lehmann, A., 107. 
 
 Lip-movement, 43 ff . ; origin of, 108 ff. 
 
 Mann, H., 23. 
 
 Massachusetts, Board of Education, 
 23. 
 
 Material, suitable for reading, 281 ff. 
 
 Meade, C. D., 17. 
 
 Messmer, O., 79. 
 
 Methods, of teaching reading, 3 f . ; 
 alphabetical, 4 ; word and sen- 
 tence, 4. 
 
 Meumann, E., 137 ff. 
 
 Mueller, R. F., 3. 
 
 Musculature, articulatory, 103 ff., 
 134 f. 
 
 Netschajeff, A., 136. 
 Neural factor, 77. 
 
 Norms, present compared with pro- 
 posed, 203. 
 
 Oberholtzer, E. E., 17, 41. 
 
 O'Brien, J. A., 203, 204 ff. 
 
 Ocular reaction, uniformity of, 54 ff. 
 
 Ocular span, 56 f. 
 
 Overlapping of visual spans, 58. 
 
 Paraphrase, 88. 
 
 Paulhan, F., 104 ff. 
 
 Perception, training in, 49 ff., 126 ff. 
 
 Perceptual process, by word-, phrase-, 
 and sentence-wholes, 3. 
 
 Perceptual span, relative and abso- 
 lute, 52, 132 f. ; widening of, 50 f. 
 
 Peters, C. C., 37. 
 
 Phonetics, 22 f., 151. 
 
 Photographic records, 234 ff. 
 
 Physiological, basis of reading, 14, 70. 
 
 Pintner, R., 16 f. 
 
 Practice, in rapid reading, 37 ff., 
 82 ff. ; length of, 91 ff. 
 
 Prehension, span of, 139. 
 
 Premonitions, meaning, 129. 
 
 Preparation, thought, 96 ff. ; word, 98. 
 
 Psychological test, 136 f. 
 
 Psychology, findings of, modify read- 
 ing methods, 3 ff. 
 
 Purpose, effect of, in reading, 59 ff. 
 
 Quanta, J. O., 36, 43 f., 108 ff. 
 Questioning, manner of, 94 f . 
 
 Rapid reader's photographic record, 
 253 ff. 
 
 Rate, of reading, effect of training on, 
 171 ff. ; comparison with norms, 
 203 ff. ; factors in, 36 ff. 
 
 Reaction time, of eye-movements, 
 77 f . ; correlation with reading 
 rate, 224 ff. 
 
 Reading, comparison of silent and 
 oral, 13 ff., 16 ff. ; rates of, 16 f. ; 
 comprehension of, 18 ff. ; neglect 
 of silent reading, 20 ff. ; anach- 
 ronism of oral, 25 ff. ; silent, 
 emphasis on, 29 ; alternate, and 
 reproduction, 93 ff. 
 
 Recognition of value of rapid reading, 
 70 f. 
 
 Refixation, 14. 
 
 Regressive movements variation 
 in, 61. 
 
 Relative gains in speed, in different 
 grades, 198 ff. 
 
 Repetitive reading, 53. 
 
 Reproductions of matter read, 19, 64. 
 
 Rhythm of eye-movements, 54 ff. 
 
 Ribot, T., 101 ff. 
 
 Rivers, W. H., 164. 
 
 Ruediger, W. C., 68 f.
 
 INDEX 
 
 289 
 
 Schmidt, W. A., 5, 11, 12, 14 f. 
 
 Secor, W. B., 115. 
 
 "Set," mental, 60, 62, 74, 94. 
 
 Sex, factor of, 231. 
 
 Skimming, 94. 
 
 Slow reader's photographic record, 
 258 ff. 
 
 Speech, inner, definition of, 100 f. ; 
 history of, 101 ff . ; origin of, 108 ff . ; 
 necessity of, 114 ff . ; means of in- 
 hibiting, 117 ff. 
 
 Starch, D. A., 203 ff. 
 
 St. Louis school survey, 03. 
 
 Stoelting, C. H., 142. 
 
 Strieker. S., 103 ff. 
 
 Stumpf, ('., 104. 
 
 Subject matter affecting reading 
 rate, 53. 
 
 Synaptical connections, 112. 
 
 Tachistoscopical exposures, 4, 50 f., 
 
 83 f. 
 
 Tests, standard, 16 ff. 
 Thorndike, E. L., 24. 
 Time, control, 73 f . ; amount devoted 
 
 to reading, 1. 
 Training in rapid reading, length of, 
 
 152 ff. 
 Tulsa reading survey, 41. 
 
 Variability, in amount of gain, 208 f. 
 Vision, peripheral, 127 ff. ; foveal, 
 130 ff. 
 
 Waldo, K. D., 202. 
 "Warming up" period, 96 f. 
 Whipple, G. M., 60. 
 Will to read rapidly, 71 ff. 
 Work curve. 164, 228 ff. 
 
 Printed in the United States of America.
 
 University of California 
 
 SOUTHERN REGIONAL LIBRARY FACILITY 
 
 405 Hilgard Avenue, Los Angeles, CA 90024-1388 
 
 Return this material to the library 
 
 from which it was borrowed. 
 
 131992
 
 3 1205 00083 7037
 
 il ir 
 
 Univer 
 
 Sou 
 
 Li 1