LB 'S8 UC-NRLF *B ?tq na. D 3 ,XACT MEASUREMENTS IN EDUCATION JAMES LEROY STOCKTON, A. M. (Columbia) SUPERINTlfeNDENT ELEMENTARY DEPARTMENT NdE>IAL SCHOOL, WINONA, MINN. CHICAGO NEW YORK ROW, PETERSON & COMPANY Digitized by the Internet Archive in 2007 with funding from IVIicrosoft Corporation http://www.archive.org/details/exactnneasurementOOstocrich EXACT MEASUREMENTS IN EDUCATION JAMES LEROY STOCKTON, A. M. (Columbia) SUPERINTENDENT ELEMENTARY DEPARTMENT NORMAL SCHOOL, WINONA, MINN. CHICAGO NEW YORK ROW, PETERSON & COMPANY o ^■d Copyright, 1915 BY James LeRoy Stockton Vv>^ ^ v^ EXACT MEASUREMENTS IN EDUCATION THESES I. Measurement in Education should have for its goal the computation of work and rate-of-work (power), in the sense in which these terms are used in Mechanics. II. Scales of force, space, and time, exist, or can be made, for school subjects; and the stand- ard units of these scales of force, and space, and time, should be combined into standard units of work and rate-of-work (power), such units directly corresponding to the foot-pound and the horse-power. (In this paper units are worked out for penmanship, and illustrated by experi- mental work involving certain applications of the Thorndike Scale.) III. Many units in many school subjects should 331230 4 EX4CJT MEASTJKEMENTS be supplemented by a single unit, making possible the computation of mental work and rate-of- mental-work (mental power) in all school sub- jects. The force involved in this computation is intelligence; the space is measured in elements of expression. (As there is no adequate scale of intelligence uncombined with any mechanical fac- tor, a theory of the necessary scale is ventured.) IV. In any case, to consider either force, space, or time, alone, or to combine them in an arbitrary manner, gives unreliable results. [This is shown, for computations in school subjects, by the pen- manship illustration. For computations of men- tal work, and mental power, experience with the Binet-Simon tests is cited in proof of the con- tention.] EXACT MEASUEEMENTS IN EDUCATION I Most persons do not any longer question the possibility of measurement in Education, because it has become apparent that measurements always have been made, and are continuing to be made. When it is said that a piece of work is good, bad, or indifferent, a measuring scale of at least three steps is evidently being used. If papers are marked A, B, C, D, E, according to the judgment of the examiner, a scale of five steps is being used. This is clearly evident; measurement is a fact in all departments of Education whenever the value of the product is expressed. There are, however, many conscientious think- ers who still question the degree of exactness to which the measurement should be carried. The common rough measurements which are con- stantly used do not seem so objectionable as the more exact scientific measurements which are being proposed. It is feared that too much exact- ness will make Education formal or mechanical. 6 EXACT MEASUREMENTS If this fear were justifiable it would furnish a very strong foundation for a stand against meas- urement, for modern Education cannot defend formalism. Fortunately, however, the difficulty can be met with the following statements : (1) Education, in so far as it can be measured, is a product, (2) Mechanical methods of measuring a prod- uct do not require mechanical methods of pro- ducing that product. Handwriting might be measured by the most mechanical means one could imagine, and yet have been produced by the freest, most spontaneous method that exists. The worst that can be said is that mechanical measurement may, in the careless and unthoughtful, tend to produce mechanical methods of production; but pre-supposing reasonable thoughtfulness in its use, nothing promises more for Education than does exact scientific measurement. In this work progress has been made through the establishment of relatively exact scales in certain school subjects ; but the progress has been slow, as it always is in a new field. Confusion, also, is beginning to result, because the plunge into this undiscovered country has naturally been IN EUUCAXroN 7 made with no very definite route marked out in advance, and with no very adequate conception of the extent of the territory to be explored. There is not much evidence that it is realized that the making of scales may be merely a scouting on the frontier — merely the beginnings of roads whose end lies in a more remote country. If this should prove to be true much wandering will be prevented if a return is made to the starting point, and an attempt made, in the light of all past experience, to map the whole route from the beginning to the end. Then if the map shows districts to be traversed in which as yet no road exists, the problem will at least be clear when these sections are reached. It is the purpose of this paper to suggest that an unexplored district does exist in the field of measurement in Education, and that the making of scales takes the investigator only part way on the road to the final goal. An attempt will be made to show that even with the scales now avail- able, or with other similar ones which may be made, still another step must be taken or Educa- tion remains in the same condition as was the science of Mechanics before the time of Watt. 8 EXACT MEASUREMENTS Before Watt the scales of feet, pounds, and min- utes were in use, but there was no attempt to use them in a computation of work and rate-of-work by means of the composite units called the foot- pound and the horse-power. The formulation of these units opened a new realm in Mechanics. From now on this discussion will deal with the hypothesis that there is such a new realm in measurement in Education, and that all of our efforts in this field, including the making of scales, will gain in definiteness and worth through being directed toward this final goal — the computation of work and rate-of-work; work being used in its technical meaning for the science of Mechanics. Any hypothesis, in order to justify itself, must show wherein it meets conditions unmet before; it gains its adherents through its ability to clear up existing confusions, and to present worthy results. Therefore the problem squarely in view is (1) to show that there is confusion, (2) to show that this hypothesis clears up at least some of it, and (3) to show that the results from the applica- tion of the hypothesis are reasonable and valuable. There are at least three points where confusion exists. The first is clearly stated by Whipple, IN EDUCATION 9 ^* Manual of Mental and Physical Tests/' as fol- lows. ^' The question arises: shall efficiency be measured in terms of quality, excellence, delicacy, or accuracy of work, or shall it be measured in terms of quantity, rate, or speed of work? For this question no general answer can be given." Certain expedients are then suggested, but no final and exact program is outlined. An attempt will be made to show that the hypothesis of work clears up the problem of the true relation between quantitative and qualitative scales, which is the real problem propounded in the foregoing quota- tion. Another source of confusion, distinct, but indirectly included by Whipple in the lines just quoted, lies in the treatment of the time element involved in testing. This, when considered at all, is ordinarily carried as a separate index; but in many cases there is a tendency to neglect it entirely, often with grave results, as happens when two schools are compared in handwriting, without any consideration of the time involved in the production of the specimens. The need for a separate index vanislies under the hypothe- sis of worJc, and time receives its legitimate and necessary emphasis. The third source of confu- 10 EXACT MEASUREMENTS sion is in the conception of efficiency itself. This conception is vague and indefinite. Various defi- nitions are contending for recognition. All school measurement is supposed to be directed toward the determination of relative efficiency, and yet there is disagreement as to what constitutes true efficiency. There can be no such disagreement under the hypothesis of work. These claims for the hypothesis must now be more closely examined and tested. This task will be furthered by an analysis of mechanical work and rate-of-work. As already indicated, before the time of Watt the scales of feet, of pounds, and of minutes, were in use. It was therefore possible to know that a force of 5047.00 pounds was at work where it was found necessary to exert another force of 5047.00 pounds against it — as in lifting against the force of gravity. It was also easily seen that another valuable formulation could be made if distance were included. To say that one machine lifted a weight of 5047.00 pounds, and another a weight of 5556.00 pounds, led naturally to the idea that the second machine was the stronger; but as soon as the distance was taken into consideration a doubt was raised. If IX EDUCATION" H the first machine raised 5047.00 pounds four feet, and the second machine raised its 5556.00 pounds four feet or more the doubt as to the greater strength of the second madhine did not exist. But if the first machine raised 5047.00 pounds four feet and the second machine raised 5556.00 pounds tJiJ^ee feet, indefiniteness as to strength was apparent. It was possible to carry the two indexes in each case (5047.00 pounds lifted four feet, and 5556.00 pounds lifted three feet) and to get certain rather valuable results. One could say that he preferred the smaller amount lifted the greater distance, or the larger amount lifted the smaller distance; but the computation of work from these data made a single index possible, put definiteness into exact comparison of the two, and so opened the new realm as previously mentioned. Quoting from a modern text in physics: *^ "When a body acted upon by a force moves in the direction in which the force is acting, work is said to be done. * ^ * The amount of work done is measured by the product of the force by the distance which the body moves along the line of the action of the force. Thus when a two pound weight is raised three feet, it moves a dis- 12 EXACT MEASUREMENTS tance of three feet against a force of two pounds and therefore six foot-pounds of work is done against the force of attraction of the earth."* Work, therefore, in Mechanics means force acting through space, and is computed by the formula W = F X S. Where work is to be con- sidered, force alone means nothing and space alone means nothing; but force acting through space means ivorh, and a certain unit of force (the pound) acting through a certain unit of space (the foot) means a certain unit of work (the foot-pound). This unit of work may be briefly expressed as unit force acting through unit space. By means of this unit the two ma- chines above referred to may be definitely com- pared as to the work they do. One machine did work equal to 5047.00X4.00, or 20188.00 foot- pounds. The other did work equal to 5556.00 X 3.00, or 16668.00 foot-pounds. The relative work- ing ability of the two machines is definitely ex- pressed by the ratio of 20188.00 to 16668.00. But there is still another element to be con- sidered; viz., that of time. The amount of work •Kimball — '* College Physics." IN EDUCATION 13 is the same whether 5047.00 pounds be lifted 4.00 feet in one minute or in one hour or in one year; but it is often important to know for various rea- sons, at what rate this work can be delivered. Hence another unit (a certain amount of work delivered in a certain time) becomes necessary. If a definite amount of work in a definite time is taken, it is not important just what the amount or the time may be, except for considerations of convenience. But if there is no unit agreed upon, two indexes must be carried as before, and com- parisons are again cumbersome. 20188.00 foot- pounds in five seconds, must perhaps be compared with 16668.00 foot-pounds in 51/2 seconds. In order to do this it must all be put upon the basis of amount delivered in one second by dividing the number of foot-pounds of work by the time. 20188.00 foot-pounds divided by 5.00 = 4037.60 foot-pounds per second; 16668.00 foot-pounds divided by 5.50 = 3030.54 foot-pounds per second. « These can now be compared with each other. But it is still better to have a standard unit of accomplishment per second and compare all other accomplishments with the unit. Watt selected as the unit of rate-of-work the number of foot- 14 EXACT MEASUREMENTS pounds per second accomplished by the average horse (550.00 foot-pounds per second). He could have used any other number, but this number proved convenient. Using it as a unit, it is seen that the machine which did 4037.00 foot-pounds per second was a 7.34 horse-power machine. The machine which did 3030.55 foot-pounds per second was a 5.51 horse-power machine. These two results admit of immediate and perfect compari- son, and the formulation of this method of com- puting rate-of-work (or power, as the physicist calls it) opened to Mechanics the second part of the new realm, as the computation of work itself opened the first part of that realm. In attempting to appropriate for Education this new field of work and rate-of-work (power) it is necessary to formulate units of work and rate-of-work (power) based upon either an anal- ogy to, or an identity with, force acting through space in time. Examination of the situation seems to show a real identity. That which is measured in Education is always some kind of expression through movement occurring in space, which movement is controlled (changed) either in direction or magnitude by some agent. The dif- IN EDUCATION 15 ferences which we measure in handwriting are differences in direction and magnitude of motion, registered on paper in the form of letters. Even thought itself becomes manifest and can be meas- ured only in terms of expression, which expres- sion is in movement, resolved in the last analysis into changes in direction or magnitude. Now the only name the world has ever had for that which changes the motion of a body, either in direction or amount, is force. There seems to be no reason for calling the agent behind expression by any other name than force. It meets the definition of force, and is measured as all force must be ; i., e. in terms of its products. There is therefore an identity between one element in units of work and rate-of-work (power) in Mechanics, and the same element in Education. (This affirmation of identity is meant to carry only so far as the assertion that the agent behind achievement in Education is a force. This force may differ from other forces, just as electrical force probably differs from gravitational force etc.) But all of the movements which are initiated and controlled by the force, take place in space ^r4 time. That is, the force acts through the 16 EXACT MEASUREMENTS space in the production of the given movement in the given time. In handwriting when a word is written, the force (or control) acts through the space roughly measured by the linear arrange- ment of letters, this measurement being exactly parallel to the rough measurement of space by paces or other such linear units, used before the more accurate foot and inch where selected as units. The addition of the time element here as elsewhere, provides for the computation of rate- of-work, or power. This relation between force, space, and time is not an arbitrary but a natural and necessary relation. Physics demonstrated and adopted it; physics did not create it. The relation between the factors is a universal rela- tion which is found wherever the three factors are involved. Hence it seems inevitable to apply this prin- ciple in Education in a manner similar to its use in Mechanics.* ♦Reference is made earlier in this paper (page . . ) to certain attempts (see Whipple, Manual of Mental and Physical Tests) to correlate these factors. Reference should also be made to Brown's excellent article on Reading in the Elementary School Teacher for June, IN EDUCATION 17 An attempt will now be made fully to illustrate and to apply the idea in the field of handwriting, since it is there that the most suitable scales nec- essary to the formation of the units are found. In handwriting there is motion under varying degrees of control. This control which alters the direction and magnitude of motion is a force. But the force here presents a complication of two factors; viz., conscious direction, which may be called intelligent force, or intelligence ; and habit, which is mechanical. It follows that the motion, then, is a resultant of the action of more than one force; but this does not alter anything in relation to the computations. A resultant of two or more forces is dealt with under the same laws as are simple forces. The one thing which must be remembered in this connection is that because the force, intelligence, is combined with a mechan- ical factor, the work computed cannot be called purely mental work but mere penmanship work. 1914, and to others. In all cases, however, which have come under the observation of the writer of this article, arbitrary relations have been established among the fac- tors, and the necessary and permanent relation has been disregarded. 18 EXACT MEASUREMENTS In the second part of this paper the discussion of the computation of purely mental work, where the force involved is intelligence alone, is con- sidered. Now in order to make the formulation of units possible, there must be a scale of the force and a scale of the space. Then the standard unit of the scale of force can be combined with the standard unit of the scale of space into the standard unit of penmanship work ; and the standard unit of pen- manship work, complicated with the standard unit of a scale of time, can be the standard unit of rate- of -penmanship work, or penmanship power. But can the force involved in penmanship work be measured? Not directly, any more than the force of gravity can be measured directly. But the force of gravity is measured by its effects (ten- sion of a spring), and the force involved in pen- manship work can be measured by one of its effects; viz., the amount of quality exhibited by the handwriting produced. This amount of quality is, roughly at least, measured by the Thorndike handwriting scale, and the idea of such a scale is apparently sound and capable of refinement. Of this more will be said later. In IN EDUCATION 19 the meantime this scale will be used as a means of continuing the illustration ; and it should con- tinue to be used for purposes of school measure- ment until a better one takes its place, or until it is further made more nearly perfect. Let unit force (or control) be that control which produces penmanship which exhibits the amount of quality designated as No. 1 of the Thorndike scale^^ Let unit space be the space measured by one letter. Then if a person writes 60.00 letters equal to No. 12.00 quality Thorndike scale, the work involved is force X space or 60.00 X 12.00 or 720.00 units of work. These units correspond to foot-pounds and should be desig- nated by some name of similar significance. It is necessary at this point to guard against the idea that the plan as outlined above identifies force with quality of handwriting, and space with quantity of handwriting. The quality of the writing is not the force, but it is the measure of the force; the number of letters is not the space, but it is the measure of the space. Since quantity and quality are here mentioned, it seems best to discuss them further in order to show that the plan does give the combination of 20 EXACT MEASUREMENTS quantitative and qualitative scales which solves the vexed question (as claimed earlier in the paper). When it is said that a person does 60.00 letters of No. 12.00 quality in a minute, and work is computed by finding the product of 60.00 and 12.00 according to the formula W = F X S, viewed superficially it seems as if force were identified with quality and space with quantity, and that the two (quantity and quality) were merely multiplied together as a solution of the quantity-quality difficulty. But force is not iden- tified with quality nor space with quantity; and when 60.00 is multiplied by 12.00 force is not being multiplied by space (as the formula F X S would seem to imply) but a measure of force is multiplied by a measure of space, as previously indicated. Neither when 60.00 is multiplied by 12.00 is quality multiplied by quantity; but a quantity of quality, used as a measure of force, is multiplied by another quantity of quality, used as a measure of space. The Thorndike scale is a quantity-quality scale. No. 1 handwriting as measured by the scale exhibits a certain amount (quantity) of handwriting quality; No. 12.00 handwriting, following the assumption of the IN EDUCATION 21 author of the scale, exhibits an amount of hand- writing quality 12.00 times as great as that exhibited by No. 1 handwriting. That is to say that what we designate as No. 12.00 quality is not quality alone, but quantity of quality. It is the same with space. The unit of space in writing is the letter. This is rough, as has been admit- ted, but letters arranged in linear fashion meas- ure the space much as it might be measured by more or less irregular paces. 60.00 paces means 60.00 movements of pace quality. Spaces and paces have many qualities all of which are not held in common, but one quality is common to both; viz., extension. Hence the extension involved in paces is often used to measure the extension of space. In like manner it is pro- posed to use the quality of extension involved in letters as a measure of the extension of space. One letter, therefore, is equal to a unitary amount (quantity) of the space quality known as extension. Therefore the multiplication of 60.00 by 12.00 in the problem above cited, and in all similar problems, while it seems to be a mul- tiplication of quantity by quality, and actually settles our confusion as to the relation of these 22 EXACT MEASUREMENTS scales, is really a multiplication of a quantity of quality by another quantity of quality, or in other words a multiplication of quantity by quantity. A summary of points thus far made follows : Exact measurement in Education is desirable and much has been done; but there is a realm into which it has not been extended ; this is the realm of work. Computation of work requires the consideration of force acting through space. There must be a quantitative scale of some meas- ure of the force, made in definite standard units which can be counted, and the steps of the scale must bear a definite and known relation to one another. There must also be a definite scale of the space, meeting the same conditions as does the scale for the measurement of the force. Then the standard units of these scales must be com- bined into a composite unit of work, comparable to the foot-pound. So far it has been shown how the conditions can be met for handwriting: the Thorndike scale is used as the measure of the force. No. 1 handwriting being the unit; letters are used to measure the space, one letter being the unit. Combining these standard units into a composite unit of work gives One Letter — IN EDUCATION ' 23 No. 1.00 T scale as the result; the 60.00 letters No. 12.00 T scale equal 720.00 units of work (using the formula W = F X S). Now it becomes necessary to compute rate-of- work, and a unit must be found. When Watt wished to compute rate-of-work (power) he had to settle upon a representative number of foot- pounds per unit of time as a unit. So for hand- writing there must be selected a certain number of letters No. 1.00 T scale per unit of time. Any number would do, provided that it was definite and agreed upon, and used by every one. But for comparative purposes (in order that the unit may stand as a sort of goal of achievement) it is desirable that the number be put at some point near, probably slightly above, the average combi- nation of speed and control possible for the average seventh and eighth grade public school pupil. However, since all seventh and eighth grade public school pupils write above No. 1.00 T scale handwriting, it is most feasible to get the average of both speed and control for such pupils, and then to reduce that number to No. 1.00 quality. If these suggestions are carried out and the 24 EXACT MEASUREMENTS right computations made, there is added to meas- urement in handwriting (and by the same meth- ods there could be added to the measurement of any other school subject) the realm of computa- tion of 'work and rate-of-work (power) which Watt added to Mechanics. The tendency in handwriting measurement has been to take the product of one school and measure by the Thorn- dike (or other) scale and get the average control. Then to take another school and do the same and compare the two results. This is exactly similar to that measurement in Mechanics which con- siders how much a machine can lift against the force of gravity but does not ask through what space the force acts, nor in what time the effort is performed. Some investigators have seen this difficulty and have set a time limit upon the making of the specimens and have counted the words or letters written in a certain time. But these results have been carried in a form not suitable for actual comparisons. It is much as if one tried to compare two machines by saying the one could lift ten pounds two feet in one second, and the other nine pounds two and one- half feet in one second, without trying to com- IN EDUCATION 25 pute the work or the rate-of-work (power; in- volved. To make units of work and rate-of-work (power) for penmanship (or other school sub- jects) solves the time problem mentioned in the early part of this paper, as it has been shown to have solved the quantity-quality problem. But in order to put the plan fully into opera- tion for handwriting, there is needed a knowl- edge of how many letters, and what quality of letters, the average seventh and eighth grade public school pupil writes per minute. To get at least preliminary light upon this matter, fifty such pupils were tested. Copying from the printed page under a set time limit was at first tried. Each pupil wrote three tests representing (1) his ordinary work, (2) his fastest work, and (3) his best work. These papers wore scored for control by the Thorndike handwriting scale, and for space by the counting of the number of letters on the paper. Next an attempt was made to get truer data for writing per se, by eliminat- ing the perception element so common in the copying. This was done by asking the children to write memorized material. There were three five minute tests as before; viz., (1) ordinary, 26 EXACT MEASUREMENTS (2) rapid, (3) best. The tests were given to the children collectively and the papers scored as before for control and space. In all of the tests the same instructions were given to all of the children, the same part of the day was used, and in general, the usual precautions were taken to insure uniform validity in the results. Below is a table giving averages and the devia- tions from the average for both control and space in the full series of six tests. Abbreviations used: 0. C. — Ordinary Copying 0. M. — Ordinary Memory H. C. — Hurried Copying H. M. — Hurried Memory B. C. — Best Copying B. M.— Best Memory . Av. = average; the tables are per minute of time. SPACE Av. Av. o.c. H.C. B.C. O.M. H.M. B.M. 54.02 77.48 49.60 CO'TEOL 75.07 94.10 60.07 O.C. H.C. B.C. O.M. H.M. B.M. 11.16 10.44 11.40 10.48 9.66 10.90 IN EDUCATION 27 AvEEAGB Deviations (from average) Av. Av. SPACE o.c. H.C. B.C. CM. H.M. B.M. 12.56 9.57 8.20 CONTEOL 9.79 12.30 9.48 O.C. H.C. B.C. CM. H.M. B.M. .83 1.10 .84 1.20 1.40 .82 This table does not involve enough cases to prove anything; but, in addition to presenting other interesting information, it does throw light upon the question as to what constitutes a reason- able unit of rate-of-work (power) in penmanship. No medians are given, but they correspond very closely to the various averages, and there seems to be little choice as to whether conclusions shall be drawn from the one or from the other. Since the results are suggestive only, it is simpler to deal with the averages only. The table of devia- tions will show that the deviation from the aver- age is but eight to twelve letters (or about two ordinary words) per minute. While fast handwriting and best handwriting present much material for comparison, it is, after 28 . EXACT MEASUREMENTS all, rather certain that ordinary writing is the best general measure. More than this, the tests marked ** ordinary memory " are naturally selected, for '* ordinary copying " was interfered with by the perception element. It will be seen that the space units under ordinary memory are 75.07 and the control units 10.48. The work is 75.07 X 10.48 (F X S) or 786.73. Approximately this result has been selected as a possible unit of rate-of-work (780.00 letters of unit control in one minute). Certain undiscussed aspects of the problem make it seem that the factors here involved would represent a better standard to strive for if the relation were changed to 65.00 and 12.00. This combination represents the same number of units of work (780.00) and will be dealt with in this paper tentatively as the stand- ard. Thorndike in his monograph on hand- writing suggests the same amount of work (60.00 letters of 13.00 times unit control) as a limit beyond which it is useless to train children in this subject. The tentative units ^suggested for handwriting are therefore: IN EDUCATION 29 Unit of work = One Letter — No. 1.00 T scale. Unit of rate-of-work (power) = 780.00 letters — No. 1.00 T scale, in one minute. If this represents a fair achievement in hand- writing, or if it does not and yet can be agreed upon as a measure, it will furnish a much more accurate and fair means of comparison than has heretofore been in use. The handwriting of pupil No. 1 scaled 12.00 (Thorndike scale). The handwriting of pupil No. 5 scaled 10.00. Using the Thorndike scale as it is often used, this is as far as the matter would be carried and it would be said that pupil No. 1 was the better pupil in handwriting. What can rightly be said is that pupil No. 1 exhibited the most handwriting control. But it is impor- tant also (for complete comparison) to deal with other factors. First, through what space was this average control sustained? Through 265.00 letters for pupil No. 1, and through 387.00 letters for pupil No. 5. Now which is *^ better " — 265.00 letters of No. 12.00 control or 387.00 letters of No. 10.00 control? There has been no way of 30 EXACT MEASUREMENTS telling at all accurately, and no system of eom- putation will ever tell which is better. The ques- tion of best all depends upon the definition of best, upon the aim toward which the work is directed, and upon the degree to which the aim is accomplished. For certain purposes, 387.00 let- ters of No. 10.00 control may be much better than 265.00 letters of No. 12.00 control (or vice versa). Control may for certain purposes be preferred to space, or excessively slow writing, for certain other purposes, may not be so good as more rapid work of less control. But though the knowledge of the aim may change the judgment as to which is best, it does not at all change the amount of work delivered. This amount of work delivered is a constant (for pupil No. 1, 265.00 X 12.00 units of work) and to make use of it opeiis to Education one-half of the new realm of work added to Mechanics by Watt. Pupil No. 1 did 265.00 letters of No. 12.00 control, or 3180.00 units of work. (A name must be coined for this unit.) Pupil No. 5 did 317.00 letters of No. 10.00 con- trol, or 3170.00 units of work. To have tried to compare these two items by IN EDUCATION 31 carrying the two indexes would have been indefi- nite and burdensome; but to compare 3180.00 with 3170.00 is simple and accurate. [This means accurate to the degree to which the scales involved are accurate, and though the scales are rough as yet, they are capable of refinement. A partial discussion of this matter follows later in regard to the Thorndike scale.] It is desirable also to add the time element and to know which of these pupils worked at the faster rate, for time is always an important factor in any task, although there are, of course, occasions when one is willing to sacrifice this element to other elements. Pupil No. 1 did 3180.00 units of work in 5.00 minutes, or 636.00 units per minute. Since 780.00 units per minute has been tentatively selected as a standard unit of rate-of-work, this pupil No. 1 exhibited less than one standard unit of rate-of-work (power); i. e. 636.00 divided by 780.00 = .81 units of rate- of-work (power). A name must also be coined for this unit. This name will correspond to the ^^horse-power'' as used in Mechanics, as the name for the penmanship unit of work will corre- spond to the foot-pound. Pupil No. 2 did 3170.00 32 EXACT MEASUREMENTS units of work in 5.00 minutes, or 634.00 units per minute. 634.00 divided by 780.00 = .81 units of rate-of-work (power). Here are two pupils who in work delivered (computed to two decimal places) are equal; but no such judgment could have been made from a mere examination of the data, or from the carry- ing of separate indexes. A definite unit of rate- of-work (power) based upon a unit of work makes this comparison possible. Even at the risk of being called to account for unnecessary repetition, it must again be said that there is no thought that these computations have proved what is the best condition. They have merely expressed accurately the facts of the condition. The question of best or worst is to be decided on the basis of the aims for the work. One may go intelligently about the task (on the basis of his aim) of producing any ratio between the factors of force-time-space that he may desire. Any adjustment of these factors may be sought, just as in the movement of physical weights a small force working through a long distance may be preferred, or a large force working through a short distance. The time element may also be IN EDUCATION 33 long or short — all of these elements varying according to the aim. However, should there still be a desire to retain in these combination results, the evidence by means of which at any time the exact figures for control and space could be regained, it may be done by the following process, and at the same time a valuable element may be added to the final result. Pupil No. 1 was found to be worth .81 units of rate-of-work (power), because he wrote 265.00 letters of No. 12.00 times unit con- trol in 5.00 minutes, or 53.00 letters of No. 12.00 times unit control in one minute. His space was therefore 53/65 of normal and his control 12/12 of normal. These fractions may be observed, and on the basis of the aim for this work (which may require a preponderance of space or con- trol) judgment may be made as to whether or not the combination is a good one. To facilitate this judgment the answer may be written .81, (53/65 X 12/12). But this notation will be all the more valuable if these fractions are reduced to deci- mals, since their relation will then be much plainer. Following out this suggestion for the students just compared, it is written that 34 EXACT MEASUREMENT pupil No. 1 delivered .81 units, (.81 X 1.00) ; pupil No. 5 delivered .81 units, (.97 X 0.83). The same method of comparison may be used for two schools. Below is a table giving averages and deviations from the average, in space and in control, for fifty normal school girls, in six tests similar to those reported upon for grade chil- dren. Following the table is a comparison of certain records of the normal school girls, with corresponding records of the grade children. Abbreviations used: O.C. -^Ordinary Copying 0. M.— Ordinary Memory H. C— Hurried Copying H. M. — Hurried Memory B. C. — Best Copying B. M. — Best Memory Av. == average ; the tables are per minute of time. Av Av. SPACE o.c. H.C. B.C. CM. H.M. B.M. r.... 74.88 99.66 73.56 CONTEOL 91.96 111.00 81.74 0. c. H.C. B.C. CM. H.M. B.M. r.... 11.50 11.00 11.80 11.56 10.30 11.78 IN ErUCATION 35 Average Deviations (from average) Av. Av. SPACE o.c. H.C. B.C. O.M. H.M. B.M. 11.52 12.46 9.42 CONTBOL 9.84 12.18 10.75 O.C. H.C. B.C. O.M. H.M. B.M. .66 .68 .56 .67 1.07 .69 To compare the records in ordinary copying the following computations are made: Normal girls space average 74.88 ; control aver- age 11.50. 74.88 X 11.50 = 861.12 ; 861.12 divided by 780.00 = 1.10, (74.88/65.00 X 11.50/12.00) or 1.10, (1.15 X. 95). Elementary school space average 54.02; con- trol average 11.16. 54.02 X 11.16 = 602.86; 602.86 divided by 780.00 = .77, (54.02/65.00 X 11.16/12.00, or .77, (.83 X .93). By merely looking at the tables it could be seen that at all points in both space and control, the normal school students were ahead of those in the elementary school. It would also be pos- sible to tell how much they were ahead in space and in control, each one being considered sepa- 36 EXACT MEASUREMENTS rately ; but without some such process as the one suggested it could never be told how much the normal school was ahead in the actual amount of work delivered. But having computed the amount delivered in each case, the comparison could be made. It is known also that more work was done than was delivered. There was loss, just as there is loss in the working of an engine where fric- tion and other causes subtract from the power actually delivered. No machine delivers as much work as it actually does, and the percentage delivered varies constantly from day to day and even from hour to hour or moment to moment. To say that a machine is ten horse-power, means that it averages ten horse-power, or that it is ten horse-power at the time that the measure- ment is made, A badly adjusted carburetor or an excess of friction at a given point may make a gas engine lose almost any per cent of its power, even to not being able to run at all. A horse grown nervous from misuse may ^* jump up and down " in one place and pull nothing. A child (metaphorically speaking) might do the same thing when nervous over being asked to IN EDUCATION 37 do his best work, or when indifferent through lack of motive (as might be true of ordinary work). He does a large amount of work, per- haps, or possibly he does not. Theoretically he should put into his work his whole self and the same se]f each time, and deliver, without waste, an equal amount of work in a given time, even though the factors of force, space, and time varied in the different cases. Practically he does not put in each time his whole self or the same self, and, also, there are many other sources of loss, so that in given periods of five minutes, or other time space, the amount of work actually delivered varies. Data computed from table No. 1 (using averages, but remembering that results would be similar for individuals) show that work delivered in ** ordinary memory ''' was 786.73 units (75.07 X 10.48) ; in '' hurried memory " 909.00 units (94.10 X 9.66) ; and in '' best mem- ory " 654.76 (60.07 X 10.90). " Under the three different sets of conditions three different amounts of work were delivered. This is exactly what should be expected because of the varying conditions under which the work was done. No one as yet can point with certainty to the proved 38 EXACT MEASUREMENTS reasons for the actual relations between the dif- ferent products, but it is easy to advance entirely reasonable explanations. Worry, perhaps, or a habitual slowness where best writing is attempted, would account for the small amount of work delivered in *^ best memory." It is difficult and nerve trying to attempt to make one's hand- writing better than it usually is. The attempt cuts down the space and does not largely increase the control. The loss in work delivered is there- fore great. On the other hand, it is usually not nearly so difficult or disconcerting to increase speed beyond one's average. The attempt to do best work is a cause not only of very slow prog- ress while letters and words are being written, but also of much loss between the separate letters or words. The attempt to do fast writing is the cause of a great gain in space (not much waste between letters or words), and while the loss in quality (control) is considerable, it is still not sufficient to overcome the gain in space, and the work is correspondingly greater. Where quality is required speed drops 1/5 from ordinary, and where speed is required, quality drops less than 1/8 (see table No. 1 — memory tests). This IN EDUCATION 39 seems to mean that there is less total loss in hurried work than in best work, and would corre- spond with the results one would naturally expect from the fact that school children are in the habit of doing much hurried writing at certain times, while they have certainly less habit of doing best work and consequently the loss is greater when best work is insisted upon. The facts just brought out tend to justify the third claim for the hypothesis of work; viz., that it settles the question of a net index of efficiency and opens the way toward a study of waste in Education. Using efficiency as it is used in Mechanics, it is the ratio of work delivered to work done. This ratio expressed as a fraction is the . net index of efficiency. It is not now known how to determine work actually done. In handwriting, for example, it is known only how to determine work delivered and this requires the use of the plan suggested in this paper. There are few, however, who would doubt that work done can eventually be measured, and when it is measured, the net efficiency index will be assured, and the way will be opened for attack upon the problem of waste. 40 : EXACT MEASUREMENTS Returning, however, to the consideration of work delivered in handwriting, it is apparent that the validity of the results depends upon the validity of the Thorndike scale, and upon the reliability of judgments of handwriting, which judgments are based upon the scale. Just as no additional accuracy is secured by carrying to three decimal places results based upon data carried to two decimal places only, so the accuracy of computations based upon imperfect scales is really no greater than the accuracy of the scales themselves, and of the judgments based upon the scales. At present both of these items (the accuracy of the scale and the accuracy of the judgment based upon the scale) may be ques- tioned, and corresponding allowance must be made in placing any dependence upon computa- tions involving units derived from the scales. Yet, even with the necessary allowance, valuable use of the units may be made. Also, the rough- ness of the scales, and of judgments based upon them, can be overcome, for handwriting as a product can be scaled accurately as to excellence. But in order to do this, a sufficiently large number of representative specimens must be m EDUCATION 41 I available. One of the main objections to the I Thorndike scale made by writing supervisors is that the specimens used were not representative of public school writing — that the specimens were all largely poor, or at least indifferent, and that the really good qualities were not fairly represented. However this may be, it at least raises the question of what grade of handwriting may properly be expected of public school chil- dren, and how is the quality of handwriting to be judged, anyhow? Did Thorndike have repre- sentative specimens and how is one to tell what is representative? The judgment of quality and therefore of the scale, must be based upon the opinion of some one as to what is or is not excel- lent. No machine can ever set up standards of excellence. No machine can ever decide as to whether a vertical or a slant penmanship is bet- ter, or as to whether legibility is worth more than beauty etc., etc. ; but agreement can be made upon these points, and when this has been done, it is conceivable that a machine could be made to tell whether or not a given specimen is up to the standard. Neither is it imperative that large numbers of persons should be employed in 42 EXACT MEASUREMENTS the setting up of the standards, except in the sense that large numbers must agree to the stand- ard as set up. That is to say, in this case as elsewhere, all could defer to a single authority, and agree to accept the grading of one expert, and to use his scale. It is better, probably, to get the judgment of many experts, as Thorndike did, and to agree to abide by the collective judg- ment. But the fundamental thing is the agree- ment, just as in any discussion there must be an agreement upon a definition of terms, after which it is possible for the persons to understand each other and to talk definitely in the terms of the agreement. Whether or not the Thorndike scale is the best upon which to agree, the future will tell. So far as the report shows, only a very general basis of judgment was proposed to the judges, and it is probable that before there can be a final scale a more definite agreement must be reached as to what constitutes excellence in handwriting. Is it beauty, neatness, size, slant, or legibility (mere legibility as was assumed in the making of the Ayres scale) ? There seems to be at present no general agreement upon these points ; but never- IN EDUCATION 43 theless the scales can be used profitably, and an understanding maintained in their use as long as it is known what they actually represent, no matter whether or not there is a final agreement as to what they might represent. If the terms are defined, the discussion can proceed on the basis of the definition, and the only further progress consists in the elaboration of a defini- tion which all can more fully accept. But granted the definition; i. e. the scale, a mechan- ical method of judging a certain specimen by the scale may be looked forward to. In the analysis of handwriting for the detection of forgeries such a method as been worked out. Enough of mechanical analysis is made so that the real essence of the particular writing is plainly seen. Looking to other fields it is recalled that a mechanical analysis of a play decides how much of it Shakespeare really wrote. In a similar way the authorship of a picture may be determined. It is in this direction that the movement will be made to remove the personal factor in scoring by such scales as the Thorndike scale in hand- writing. In the meantime Education is far better off with the imperfect scales which it has (even 44 EXACT MEASUREMENTS with imperfect use of them) than it was before it had them; but this advantage can be immeas- urably increased if these scales are at once made to yield real units of work and rate-of-work (power). With this outlook, there must be scales and units in other subjects than penmanship, and other units of work and rate-of-work (power). This means many scales and many units in many specific subjects. Some of these scales are already extant and the work in making others will be worth all that it costs, for when they are made and gradually refined, and when from them real units of work and rate-of-work (power) have been made, and used as a basis for comparison of individuals and of schools, the progress involved will be very great. IN EDUCATION n But the solution of the problem of school meas- urement should not be limited to the computation of many kinds of work, by many standards, in many school subjects. There is a need for a single measure, covering all subjects, which shall give a sort of summary of the abilities of an individual or of a group. This has been gen- erally and correctly expressed as a need for a measure of intelligence itself ; but it is also more than that. It is a need for the computation of that which, for want of a better name, may be called mental work. As already stated in another part of the paper, purely mental work cannot be computed so long as the force involved is partly intelligent and partly mechanical, as it is in hand- writing. But mental work can be computed by the plan already outlined for penmanship work, if intelligence is a force, and if it can be dealt with apart from any mechanical factors. It is believed that this can be done. Intelligence is a force, which by acting through 46 EXACT MEASUREMENTS a certain space, does work. Intelligence should not be regarded as merely analogous to force. It should be identified with force, since it meets the requirements of the definition of force. It is that which changes (controls) the motion of bodies; it is that ^' which makes it happen." Just as much is known about it (and no more), as is known about other forces. It would not be known to exist were it not seen revealing itself in action. A way is later suggested for freeing it from mechanical factors. But as already suggested, the need is not met by the use of a scale of intelligence (force) alone. The main use for such a scale is for purposes of comparison, and for such purposes there must be considered, also, the space through which a given measure of the force acts in a given time. To forget this is to violate the same principle which is violated when the penmanship of schools is compared by stating control (quantity of quality) alone, without asking through how much space (number of letters) the control acts in a given time. So while intelligence is to be dealt with as a force, it must also be dealt with as acting through space. But does it act through space? IN EDUCATION 47 It is certain that in so far as the presence of intelligence can be proved, the proof results from the observation of some kind of expression or action through which intelligence reveals itself. Action takes place in space, and the space may always be measured in the elements of the action. In handwriting the elements used were letters. In the broader attempt to measure the space through which intelligence acts, the expression is found to be not only in written words (or let- ters), but also in vocal sounds or in gestures which involve the body in whole or in part. While these different types of expression present many difficulties when the attempt is made to count their elements as a measure of space, yet the difficulties do not seem insurmountable, at least for the purpose of a rough scale. The problem, therefore, which is clearly in view, is that of regarding intelligence as a force acting through space, of scaling the force and of scaling the space, and of combining the stand- ard units of these two scales into a stand- ard unit of mental work (thereby also making possible a unit of rate-of-work) (power). A ten- tative method of scaling the space has already 48 EXACT MEASUREMENTS been suggested; but what can be done with regard to a scale of the force, intelligence? It has already been said that we do not know what intelligence is, but its existence is proved just in the same way that the existence of other forces is proved ; viz., by its effects. It is not known what gravitation is, but the falling of unsupported bodies is accepted as evidence of gravitation. A study of the literature of psychology will show that in like manner, adjustment to environment is generally accepted as an indication of intelligence. Intelligence is here conceived as the indefinable subjective force which, through all degrees of consciousness and intention, ^^ makes it happen." But from this point of view mere adjustment cannot be taken as definite proof of intelligence, for there are recognized certain unintelligent adjustments, and certain others which may or may not be intelligent. The tropisms of Loeb and the '^ pure instincts " of other writers are unintelligent because they are described as utiliz- ing a sort of hair-trigger mechanism which re- quires for its release no impulse of any kind from within, but merely the appropriate external stimulus. Other reactions which once required IN EDUCATION 49 intelligence, may through use, come to be per- formed sometimes with and sometimes (rela- tively at least) without that factor, and there is no way for the observer to make a distinction. In the case of the responses gained by use, it can be known that a past intelligence has been exercised, but not proved that a current, imme- diate intelligence is being exercised. Current, immediate intelligence can be proved in one way only. The proof lies not in the tropisms which give an invariable response to fixed conditions, nor yet in the reactions which have become mechanized to a greater or lesser degree, and so may or may not be at the moment intelligently directed; but it lies in consistent and effective reaction to variable conditions — to conditions tvMch are novel at the time at which the reaction occurs. Such reactions introduce a factor of selection, and where this factor is observed a subjective control, intelligence, is inferred. Hence if intelligence is to be graded it seems logical to make the grading in terms of this same principle of effective reaction to novel conditions. This can be done, for there are already recog- nized a number of very distinct ways by which 50 EXACT MEASUREMENTS living creatures respond to novel conditions with varying degrees of success. One list quite com- monly accepted contains (1) trial and error, (2) imitation, (3) *' free ideas." Some psychologists prefer to alter this list, using in the lower stages other terms, such as tropism, instinct, or circu- lar process, and discriminating in the ^* free idea " stage distinct divisions, such as sugges- tion, dictation, association, and thought. There is probably no universally accepted list ; but since some definite list could certainly be agreed upon when the need for such agreement becomes evi- dent^ any list may be taken for purposes of illus- tration. If such a list included (for example) (1) trial and error, (2) imitation, (3) suggestion, (4) association, (5) thought, then these would be the five grades in a scale of intelligence. But these grades must not be known, merely. They must be known in series, one higher in a scale than is another. And not only that, but the numerical relation between the grades must be known in order that any one of them may be expressed in terms of any other. In a general way it is accepted that imitation is higher (in a scale of intelligence) than is trial and error, and IN EDUCATION 51 thought higher than either imitation or trial and error. That creature which can respond by trial and error only, would probably be least success- ful in meeting novel conditions, and therefore called least intelligent. That creature which could respond by imitation or by thought, would be more successful and would therefore be said to possess a higher grade of intelligence. But in relation to the principle of response to novel con- ditions, the order in which these types of response stand can be definitely determined ; and not only the order, but also the mathematical relation between them. Through reasonable pa- tience in experimentation, the average percentage of success (per thousand or other large number of cases), occurring for each one of the types can be determined. The relation between successive percentages will be the relation between the steps of the scale. Tropism or some other method agreed upon as one which brings no success will be the zero. Preliminary work on such a scale is now being done by the writer, and if the results are suffi- ciently promising they will later be offered for criticism, and the aid of educational experts 52 EXACT MEASUREMENTS sought in their revision. When there is such a scale of force and also a scale of space (which is likewise being worked upon), unit force and unit space can be combined into a composite quantity-quality unit of mental work. When this unit is complicated with unit time it will furnish a unit of rate-of -mental-work (mental power), which unit will be a fair one to use in compari- sons of individuals, of schools, or any other groups, because all factors will have been con- sidered. In connection with the unit there will be needed a series of tests, by the use of which data may be obtained for computations in terms of the units. This series of tests (largely similar, probably, to those by which the scale must be established) will, if successful, be such as to make it possible for subjects of all ages to be tested for the various grades of intelligence, and for the space through which the intelligence can act in a certain time. From these data the number of units of mental work can be computed. A purely arbitrary illustration is as follows : An intelligence of grade 6.00 acts through a space of grade 50.00 in one minute. 6.00 X 50.00 (F X S) =300.00 units of mental work per min- IN EDUCATION 53 ute. Suppose the standard (arbitrarily placed or found by experiment — see penmanship illustra- tion) to be 275.00 units per minute. 300.00 di- vided by 275.00 = 1.09 units of rate-of-mental- work (the rating of this imaginary individual at that time). It should be noted here that an increase or decrease in mental work does not necessarily mean an increase or decrease in the one factor, intelligence. The change may be in either of the other factors, the space or the time. The weight of this fact will be evident in what follows con- cerning the Binet-Simon tests of intelligence. No such discussion as this would be complete without a consideration of the Binet-Simon tests (the most popular of all the tests already avail- able which make any attempt to scale the force involved in a computation of mental work). These tests have given, and are continuing to give, very valuable service; but even their most enthusiastic advocates do not claim them to be tests of immediate (current) intelligence alone. Probably the most that can be said is that they are tests of a combination of immediate and past intelligence and of inherited mechanism. They .54 EXACT MEASUREMENTS are tests of what a subject can do, as a result of his total equipment and experience. The intel- ligence which enabled him to meet novel condi- tions and to adjust himself to them, may or may not be present, for many of the tests do not call for a meeting of novel conditions at the time of the test, but for a repetition of acts the capability for doing which may have demanded a meeting of novel conditions in the past. This is one lim- itation upon the use of these tests for the pur- pose of obtaining data to be used in the computa- tion of mental work, for the computing of the mental work accomplished in any unit of time requires that the force involved in the computa- tion shall be immediate intelligence (adjustment to novel conditions at that time). It is also noted that the steps in the Binet- Simon scale are expressed in years, and it hardly seems possible that the mathematical relation between the accomplishment of the various years can ever be known accurately enough (even by the law of averages). Mental development does not seem readily measurable in years of life as a standard. It is at least unproved (and probably unclaimed) that the scale as it stands presents IN EDUCATION 55 any way by which one grade couid be expressed in terms of any other grade. But if an expres- sion in equivalents could be approximated, and if the fact that immediate intelligence is not set squarely by itself could be temporarily over- looked, the scale would then be useable for at least rough computations ; but even then it would not be fair to test subjects and to use the data for comj^arison unless allowance were made for the error resulting from a failure to consider the space through which the force acts in a given time. As the tests now stand, this question is not raised except in a few specific instances. The examiner waits for the dull child, encourages or possibly forces him, as the case seems to require, in order to get his maximum effort without regard to time. Then he is given a certain ^' mental age '^ because he passes a certain num- ber of points. If, now, a bright child be taken — one younger chronologically but quicker and more alive in every way — and if, in half of the time, results are easily obtained which indicate the same mental age, this procedure does not show that the two children belong together. Children belong together who are equal or approximately 56 EXACT MEASUREMENTS equal in ability to do mental ivork. The real test must regard the space, and the time, and the force, and make work and rate-of-work (power) the basis for comparison. Then it will be found that two children such as those cited are far apart. It is found that children tested by the Binet-Simon tests, assigned the same mental age (below chronological age) and put in classes of normal children of the given mental age, are unable to do the work. Also, children found to be below normal age and segregated in special classes of supposed approximate ages are found not to work well together after a time. They are like swimmers who go at various rates and who constantly draw away from each other. These are the natural conditions which should be expected under a system of partial testing. The recognition of the fact that they work out as they do, throws weight in favor of the conten- tion that time, force, and space should all be con- sidered, and that the computation of work smd rate-of-worh (power) is the ultimate goal of school measurement. [In this paper the units used have been pat- terned upon what the physicist calls his arbitrary IN EDUCATION 57 units, sucli as those based upon gravitation. The physicist also uses other units called ^* absolute " units, such as those based upon acceleration. Some of the phenomena of mental activity; e. g. the ^^ warming up " process, Weber's law, and the fact that one probably accomplishes more per minute in the last minutes of a test than in the first — at least raise the question as to whether absolute units, based upon acceleration may not in the future find a place in the computation of mental work.] Credit is due to J. S. Gaylord, Phychology, Winona Normal School, and to W. H. Munson, Physics, Winona Normal School, for definite constructive criticism of this paper. Country Life and the Country School Mabel Carney, 408 pp $1.25 The book gives a true portrayal of existing rural conditions ; presents a definite, constructive program for improvement ; and strikes a clear note of inspiration for organized endeavor. Principles of Teaching N. A. Harvey, 423 pp $1.25 The aim has been to make a thoroughly practical book for all teachers. Almost every difficult problem the teacher has to face is discussed in an interesting, helpful way. Especially val- uable are the chapters on the Definition of Education, Theory of Play, Interest, Analysis of the Study Process, and Motives in School. Methods of Teaching W. W. Charters, 444 pp $1.25 Among the first to try to work out general methods of teach- ing in terms of the function of subject matter. Dr. Charters has been more consistent and has elaborated the point of view more fully and more clearly than any other writer. It is a most stimulating and informing book, especially designed for use as a text in Normal and Training schools. The Personality of the Teacher Charles McKenny, 192 pp $1.00 It is generally conceded that the prime factor in making a school is the personality of the teacher. The author shows what qualities go to make up that desirable personality and how to develop those qualities. The book cannot fail to prove a source of inspiration and strength. How to Teach Arithmetic J. C. Brown and L. D. Coffman, 384 pp $1.25 ^ The aim has been to present in a clear and definite way the principles and devices with which efficient teachers of Arith- metic should be familiar. The selection and arrangement of the material shows sound pedagogical principles; there is an abundance of illustrative material; the suggestions are concrete; every important topic is treated; and the book exploits no particular theory, method or text-book. ROW, PETERSON & COMPANY CHICAGO, ILLINOIS The Educational Meaning of Manual Arts and Industries By R. K. ROW Cloth, 250 pages Price, $1.25 The aim of the author is to present an or- ganized view of the whole problem of the signifi- cance of manual arts and industries in a system of education. To this end he — First: Defines the problem of the book. Second : Throws into perspective the history of the develop* ment of manual training as a factor in education. Third: Analyzes and explains the primary impulses and interests in manual activities. Fourth: Shows the relation of manual activities to sense training, motor control and neuro-muscular development. Fifth: Discusses the intellectual, aesthetic, ethical, eco- nomic, and social values of training in manual arts and industries. Sixth: Shows to whom such training is of most value, out- lines a general method of teaching, and gives suggestions for a course of study. SOME ESTIMATES The most complete and intelligent tliesis thus far published on the value of manual training in the schools. — Chicago Record-Herald, Mr. Row never forgets the claims of education in vocation. His study has been careful and scholastic, his treatment professional and patriotic, his methods are pedagogical, and his style is attractive. — New England Journal of Education, Mr. Row's discussion of intellectual, aesthetic, ethical, economic, and social values is especially attractive and strong. — Moderator-Topics, The large view th^t the author takes of manual training and its wide application gives his book place as an effort that has not been anticipated by any other in its field. — Chrisiim Science Monitor^ Sent postpaid on receipt of price. ROW, PEIERSON & CO. CHICAGO • • • ILLINOIS PEICE LIST Essential Studies in English, Robbins and Row. Book I, Language, 294 pp $0.45 Book II, Grammar and Composition, 356 pp 60 Practical English, A. C. Scott, 208 pp 60 Manual of English Form and Diction, Fansler 10 Exercises in English Form and Diction, Fansler 60 Types of Prose Narrative, Fansler 1.50 A Practical Spelling Book 20 The National Speller, C. R. Frazier 20 Phonology and Orthoepy, Salisbury 50 Elementary Agriculture, Hatch and Haselwood 50 A Unit of Agriculture, Eliff 50 One Hundred Lessons in Agriculture, Nolan' 65 The Educational Meaning of Manual Arts, Row L25 Methods of Teaching, Charters. Revised Edition L25 Principles of Teaching, N. A. Harvey, 450 pp. 1-25 The Theory of Teaching, Salisbury 1.25 Reading in Public Schools, Briggs and Coif man 1.25 The Psychology of Conduct, Schroeder 1.25 The Personality of the Teacher, McKenny 1.00 Country Life and the Country School, Carney 1.25 School Management, Salisbury 1.00 Index to Short Stories, Salisbury and Beckwith 50 Balonglong, the Igorot Boy, Jenks 45 Reading-Literature Primer, Free and Treadwell 32 Reading-Literature First Reader Z6 Reading-Literature Second Reader 40 Reading-Literature Third Reader 45 Reading-Literature Fourth Reader 50 Reading-Literature Fifth Reader 55 Reading-Literature Sixth Reader. Ready in May, 1914. Reading-Literature Seventh Reader. Ready in May, 1914. Reading-Literature Eighth Reader, Shryock 60 Ivanhoe, adapted by Stella Humphrey Nida 50 East O' the Sun and West O' the Moon, Thomsen 50 ROW, PETERSON & COMPANY 623 So. Wabash Avenue CHICAGO SCHOOL MANAGEMENT By ALBERT SALISBURY, Ph. D. President of the Whitewater State Normal School, author ol "The Theory of Teaching," etc. Cloth. 12mo.. 196 pages, $1.00 This book represents the fruits of a lifetime spent in the schools and in the training of teachers. School conditions have changed greatly in recent years, and books on school economy which were excellent a few years ago are now antiquated. Much more is demanded of the teacher than formerly. He has, in fact, become an official of the state, w^ith larger functions and a greater need for intelligence concerning those functions than the old-time pedagog. While endeavoring to recognize this newer conception of the teacher's office, and the greater burden which it imposes, it has been the desire of the author to make a small book rather than a bulky one, excluding padding and time-honored common-place. The book is intended to serve the needs of young teachers and those in preparation for the work, and clearness has been aimed at rather than profundity. Testimonials Frank A. Weld, Pres. State Normal School, Moorhead, Minnesota "1 have been reading Salisbury's 'School Management* with great interett. It is a stimulating book and should find its way into many Normal schools." Dr. A. E. Winship, Boston, Mass. "I have spent more time on 'School Management' than I intended, because I have enjoyed it more than I expected to. It is in the fullest sense a notable book. It gives what is needed in the least space, in the best spirit, and in a most enjoyable style. I am charmed with it." Charles A. Wagner, Dept, of Methods. S. N. S., West Chester, Pa. "I have thoroughly enjoyed the sane views, the {practical suggestions, and the vigorous treatment and language of Dr. Salisbury's 'School Management.* The book includes all the necessary new features, and omits quite as wisely as it includes. It is ri^ht in size, covers the necessary ground, and occupies safe and sane positions. ' Wisconsin Journal of Education, Madison, Wisconsin. "A long life in the schoolroom as a trainer of teachers and a man who has kept pace w^ith educational progress. President Salisbury has written a practical book with little theory and every paragraph driving home principles for the safe guidance of teachers. His vigorous style, his coming-to-the-point-quick man- ner of writing makes 'School MHnagemen»:' a volume full of meat and a most valuable addition to the present literature on this important subject." Sent postpaid to any address on receipt of price. Libera? discount to classes, ROW, PETERSON & CO., Publishers CHICAGO. ILLINOIS UNIVERSITY OF CALIFORNIA LIBRARY BERKELEY Return to desk from which borrowed. This book is DUE on the last date stamped below. DEC 29 IQ^"? LD 21-100m-9,'47(A5702sl6)476 ! GAYLAMOUNT 1 PAMPHLET BINDER Manufactured by [ SAYLORD BROS. Inc. i Syracuse, N. Y. Sto-' ■ -3 3 / jlSO : r