College and Research Libraries EGILL A. HALLDORSSON and MARJORIE E. MURFIN The Perforinance of Professionals and Nonprofessionals in the Reference Interview Twenty-five sets of "indirect" and "faulty information" questions were asked at two university library reference centers, one staffed by nonprofessionals and the other by professionals. The purpose was to determine relative success of professional and nonprofessional refer- ence staff in (1) probing beyond "indirect" questions and (2) detect- ing and correcting faulty information. Als,o considered were frequen- cy and success of nonprofessional referrals on unanswered questions. CoMMUNICATION PROBLEMS in tradi- tional reference service seldom have been explored scientifically. There is little doubt in the profession, however, that such problems often result in fail· ure to be of help to patrons and thus create an obstacle to the improvement of reference service. A difference of opinion exists on the extent and seriousness of communica- tion problems. Some do not accord these problems serious consideration. On the other hand, there is some evidence to in- dicate that reference personnel may not be aware of the true extent of commu- nication problems due to lack of feed- back.1 If this is true, it is possible that these problems may be more prevalent and more serious than has heretofore been supposed. In the Rose Bowl game of 1929 stupe- fied fans watched "Wrong Way Hie- gals" run sixty-three yards toward the wrong goal line. The well-intentioned Egill A. H aUdorsson is a doctoral student at Kent State University, and Marjorie E. Murfin is assistant professor of library ad- ministration and reference librarian at Ohio State University Libraries, Columbus. reference librarian · who fails to deter- mine the patron's information need be- fore proceeding confidently in the wrong direction is equally unfortunate. One of the greatest difficulties in de- termining the patron's actual informa- tion need comes when that need is hidden under an "indirect" or "faulty information" question. For this reason these types of questions have been chosen as the subject for this study. One observational study in an academic li- brary reports that 25 percent of all questions were "indirect" in some re,.. spect and 73 percent of these "indirect" questions were successfully resolved. 2 Cole reports the occurrence of questions which do not represent the actual needs of patrons in academic libraries to be 21 percent. 3 On the basis of these studies we might expect t:bat 20 percent to 25 percent of questions asked in an academic library might not represent patrons' actual information needs. Granted that such questions do .occur to a greater or lesser extent, it must be asked how they are dealt with by refer- ence personnel and with what success. Regrettably, studies of traditional ref- erence service have concentrated on ref:. I 385 386 I College & Research Libraries • September 1977 erence failure originating in the search process and have, for the most part, ignored failure due to communication problems. The exception is .a study by Dorman Smith where an "indirect" question was asked in twenty Boston area libraries. Smith's report indicates that communication in the reference in- terview is an aspect of service seriously in need of improvement.4 If reference service failure due to communication problems is to be reduced, it is essential that we know more about how these types of questions affect reference ser- vice. Success in resolving "indirect" and "faulty information" questions also may be influenced by type of personnel used. This has a broader significance in that the use of nonprofessional person- nel in reference service is an issue which, when resolved, ·may well deter- mine the future role of the reference librarian. Thus any data on differential performance of professionals and non- professionals should be a useful addi- tion to the literature. Some documentation on use of non- professionals in reference service exists in the literature, but the most significant study of differential ability remains that of Bunge .'> To the authors' knowledge, no similar study has yet been done com- paring the performance of these two groups in the reference interview. Some of those involved in the contro- versy over use of nonprofessionals dis- count or give little thought to the communication process. Others on both sides base their advocacy of certain points of view on unproven hypotheses as to the relative abilities of profession- als and nonprofessionals in the refer- ence interview. Thus it is important to investigate these abilities. Generally, those for or against non- professional reference service divide on the following points in regard to com- munication problems: 1. Should communication problems be the responsibility of the refer- ence librarian? 2. Should they be considered of equal importance to search problems? 3. Do they occur frequently enough to have a detrimental effect on ser- vice? 4. Is the expenditure of time and ef- fort involved in solving these prob- lems justified? 5. Can detection and solution of such problems be done more successful- ly by professionals? 6. Do nonprofessionals frequently fail to make referrals whtm they are unsuccessful in determining patron needs? Those holding one point of view tend to answer "yes" to the above questions. Partially for this reason, they advocate all-professional refereQce service or ser- vice where questions are "fielded" by professionals. This point of view is ex- pressed by Wheeler and Goldhor as fol- lows: The idea that inquiries should be pre- sented to inexperienced persons and fed upward to those qualified to help is a disservice and inconvenience to readers, partly because the inexperi- enced do not know where the question should lead. 6 The opposite viewpoint usually an- swers some or all of the above questions in the negative and often favors the new concept of the information center where questions are fielded by nonpro- fessionals who then refer difficult ques- tions to subject specialists who are on call. This viewpoint is expressed by Jestes and Laird who advocate use of technical assistants. Professional librarians, although still immediately available to any patron, would be freed from many interrup- tions and better able to concentrate on collection development .... 7 Balay and Andrew also say, It seeq1s likely that paraprofessional assistants could handle these inquiries [information-direction], and could di- rect other, presumably more difficult, questions to reference librarians. s The specific problems with which this study will be concerned are as follows: 1. What is the relative success rate of professionals and nonprofessionals in resolving .. indirect" and "faulty information" questions? What are the reasons for any differences found? What are the implications of this for reference staffing? How can performance of reference staff members be improved in dealing with these questions? 2. How successfully do nonprofes- sionals make appropriate referrals when they fail to determine pa- trons' information needs? What are the causes of referral failure and how can such situations, if they exist, be improved? METHOD Two medium-sized midwestern uni- versity libraries were selected, both hav- ing a centralized reference service. One of these libraries had an information center staffed by nonprofessionals who had access to subject specialists for con- sultation. The other was staffed at all times by an all-professional reference staff. Seven investigators of both sexes and different ages were used. Twenty-five different reference inter- views were prepared, each consisting of an "indirect" question followed by a .. faulty information" question. These questions were taken from the actual reference experience of the second au- thor. "Faulty information" questions were in areas of the social sciences and humanities. Each prepared "set" of two questions was asked of a nonprofession- al at the information center library and of a professional at the second library. Reference Interview I 381 The procedure was as follows: All investigators were instructed to begin each interview by appearing mod- erately confused and making such com- ments as "I'm new here," "I don't really know what I'm doing," etc. It was felt that these behavioral clues and com- ments were sufficient to alert reference staff members to the possibility of "in- direct" questions. "Indirect" questions were defined as those where the patrons hide their spe- cific information need and, instead, ask for ( 1) books on a general subject or ( 2) a type of source which they think would contain the specific information they seek. Examples of these are ( 1 ) "Where are your philosophy books?" (patron wants quote by Aristotle) and ( 2) "Do you have an index to philoso- phy?" (patron wants life-styles as relat- ed to architecture). A judgment of success was made if at any time during a five-minute period the reference staff member probed fur- ther by asking for more specific infor- mation about what the patron wanted. Failure was judged if the reference staff member accepted the "indirect" question as representing the patron's real needs and did not ask further ques- tions about more specific information needs before attempting to terminate the interview or before five minutes had passed. Directly following this, the investiga- tor then proceeded to ask one of two types of "faulty information" ques-:- tions of the same staff member. These types of questions were defined as those where the patron presents a specific in- formation need directly but instead gives ( 1) a misspelling or ( 2) general faulty information. Examples are ( 1) Massaponti for Maupassant and ( 2) the poem "Agnes Eve" by Shelley ("Eve of St. Agnes" by Keats). Since these questions were considered more difficult, no time limit was set. In 388 I College & Research Libraries • September 1977 order to increase motivation, investiga- tors commented that finding the infor- mation was important to them and resisted mildly the first attempt to termi- nate the interview if it occurred before fifteen minutes by continuing to stand near the librarian and/ or by asking further questions. A judgment of success was made if the reference staff member detected the faulty information and obtained cor- rect information or showed the patron where correct information could be found before termination of the inter- view. Failure was judged if the refer- ence staff member terminated the interview without having done this. An exception was made when a reference staff member terminated an interview but continued working on the question and later returned with correct infor- mation before the "patron" left the li- brary. These "faulty information" ques- tions were designed so that in order to obtain correct information each refer- ence staff member had to exercise some or all of the following abilities, skills, and techniques: 1. Detection of possible faulty infor- mation 2. Questioning to obtain further key information 3. Asking for written material for purposes of clarification 4. Selecting appropriate reference sources to obtain correct informa- tion 5. Using personal knowledge to detect and correct faulty ·information 6. Referring appropriately when nec- essary Interviews were recorded in progress by an unobtrusive investigator nearby and also afterward by the investigator ask- ing the question. GENERAL RESULTS The results of this study show that the professional librarians in this sam- ple were clearly superior to the nonpro- fessionals in achieving successful solutions on "faulty information" ques- tions in the reference interview. Table 1 illustrates this. Professionals personally arrived a.t the correct solution in the reference in- terview on 52 percent of questions, while nonprofessionals did so on 20 per- cent. These results are, to some extent, in line with those of Bunge who found that the speed and efficiency of profes- sionals was slightly but significantly greater than that of nonprofessionals. He found no significant difference, how- ever, in percent of questions answered correctly by the two groups, 9 while this study shows that professionals were more than twice as successful as nonpro- fessionals in obtaining corrected infor- mation. These differences may be due partially to the following: 1. Obtaining corrected information TABLE 1 OvERALL SuccEss OF PROFESSIONALS AND NoNPROFESSIONALS IN CORRECTING FAULTY INFORMATION IN THE REFERENCE INTERVIEW Type of Reference Interview Success in personally obtain- ing correct information in reference interview ( without referral or consultation) Success with referral or consultation Total Number of Questions 25 25 Nonprofessionals Number Percent Correct Correct Solutions olutions 5 7 20 28 Number Correct Solutions 13 15 Professionals Percent Difference Correct with Solutions Non professionals 52 67 32 pts. 39 pts. in "faulty information" questions actually may be more difficult than finding answers to moderately dif- ficult factual questions. This could be because the outcome appeared to depend to a much greater extent on personal subject knowledge and to a much lesser extent on skill in use of reference materials. 2. The questions used in this study may have been harder, and greater differences between groups may emerge as questions become harder. 3. Perhaps public library questions, as used by Bunge, are less related to reference sources taught in library school than are academic library questions. 4. Attitude, orientation, role concep- tion, and motivational differences between groups may have been equalized in the controlled test sit- uation where each group was being observed and was attempting to perform at top level. If there are such motivational differences, they would show up more clearly when participants did not know they were being observed. Reference Interview I 389 5. The matching of professionals with nonprofessionals of equal ed- ucation and experience would tend to erase differences in the typical situation where nonprofessionals tend to have less education and ex- perience.10 There is some evidence in this study that the performance of nonprofessionals was less con- sistent, with one performing at a level equal to any professional and another at a level consistently low- er than any professional. 6. There may be a greater difference between the performance of the groups in the academic library than in the public library. "lNDIREcr" QUESTIONS "Broad Subject" Questions The "broad subject" question is de- fined as one where the patron hides a specific information need and asks, in- stead, where the books in a broad sub- ject area are located. On 75 percent of these questions nonprofessionals probed further, while on 91 percent of ques- tions professionals did so. This type of question, as expected, proved to be the TABLE 2 SuccEss BY TYPE OF QuESTION Nonprofessionals Professionals Total Number Percent Number Percent Difference Number of of of of with Type of Example of of Questions Questions Questions Questions Nonpro- Question Question Questions Solved Solved Solved Solved fessionals "Broad Subject" Example: "Where are Indirect your literature books?" 11 8 73 10 91 18 pts. Questions "Wrong Type of Source" Example: "Where are your almanacs to history?" 10 5 50 9 90 40 pts. Misspelling Example: Sou thy for Southey 12 3 25 9 75 50 pts. Faulty InfoFmation General faulty Questions information Example: Poem "Agnes Eve" by Shelley 13 2 16 5 38 22 pts. 390 I College & Research Libraries • September 1977 easiest for both groups of reference staff members to handle. Performances of professionals and nonprofessionals were similar on this type of question. Examination of the verbatim interviews, however, indicates that the lower success of the nonprofes- sionals appeared to be due primarily to a concept of their basic role as "teach- ing about how to use reference materi- als." Thus they were eager to explain use of catalogs, indexes, and reference books without probing further. "Wrong Type of Sourc.e" Questions The "wrong type of source" question is defined as that in which the patron hides a specific information need and, instead, asks for a type source such as· an "index to philosophy" which he or she assumes (often incorrently) will contain the specific information sought. On 50 percent of these questions non- professionals probed further, while on 90 percent professionals did so. Examination of verbatim interview records also reflected this difference. When faced with a question such as "Do you have any almanacs to litera- ture?" the professional tended to reply immediately and almost routinely, "What type of thing do you want to look up?" while the nonprofessional tended to respond by taking the patron to the catalog to look under the heading '"literature-almanacs." While the form of this question is also generally recognizable, it is consid- erably more difficult to detect as conceal- ing a hidden information need. This may be because it represents, on the sur- face, an effective approach to finding in- formation and does not spotlight the patron's obvious lack of library knowl- edge as does the "broad subject" ques- tion. Hidden information needs behind this type of question may have been un- detected because of unawareness, per- haps due to lack of feedback, of the extent to which patrons make inappro- priate source choices (one study showed 64 percent of such sources could be con- sidered inappropriate) .1 1 However, in some cases nonprofes- sionals appeared to be aware of possible poor source choices but did not probe further. This may have been because they felt there was not sufficient justifi- cation for further questioning, that they lacked authority, or that the patron might resent it. They also may have felt that it would not yield anything useful or that it was not their responsibility. Also, by letting such questions pass, one avoids at the same time the difficult in- terview which is often likely with pa- trons who make poor source choices, the need to come up with a better source, and possible failure and the resultant need to refer. ..FAULTY INFORMATION" QUESTIONS On "faulty information" questions (misspelling and general faulty infor- mation) ahout half of the nonprofes- sionals' failure (as oppost:d to 15 per- cent of professionals' failure) occurred in the first step of the process where they failed to question the patron's information and thus did not detect faulty information. This lack of expec- tation of faulty information on the part of nonprofessionals could be due to lack of orientation, lack of personal knowledge of the subject matter of the question, and lack of feedback from previous experience. Differences be- tween the two groups may have been due to the fact that most library schools encourage responsibility for resolving communication problem questions and provide some orientation and training in handling them. Familiarity with the subject matter of the questions was judged by staff members' comments during the inter- view. The professionals' greater person- al knowledge of the subject matter of the questions ( 43 percent for profes- sionals and 24 for nonprofessionals ) , perhaps arising from a higher educa- tional level, aided them in recognizing faulty information. Greater experience on the part of the professional librari- ans also · may have been a factor, but there appeared to be some new profes- sionals and some nonprofessionals with a n.-nber of years of experience. The second area of failure came in the next part of the process where, in a number of cases, reference staff mem- bers' comments revealed that faulty in- formation was recognized but no attempt was made to obtain correct in- formation. This was the case with non- professionals on 25 percent of questions and with professionals on 35 percent. Reasons for nonprofessionals' failure to make an attempt to obtain correct in- formation may have been that they did not feel that it was their responsibility or that they wanted to avoid possible failure or the need for referral. Another reason for nonprofessional reluctance may have been uncertainty about what sources to use for verifica- tion. This is supported by the fact that on misspelling questions which were pri- marily dependent on use of reference sources for solution, nonprofessionals attempted 33 percent. Professionals, on the other hand, appearing more confi- dent in use of reference sources, at- tempted 83 percent. However, on "general faulty infor- mation" questions which were primarily dependent upon interviewing rather than use of reference sources for solu- tion, both groups were equally reluctant to attempt the interview. In these cases, when no personal knowledge was pres- e~t, the success rate of both groups dropped to zero. Professionals fre- quently showed awareness that some- thing was wrong but were singularly reluctant to interview, asking only a few perfunctory questions. Clues were Reference Interview I 391 given by the "patron" but were not followed up. This reluctance to attempt the difficult interview on questions where reference sources could not easily be consulted was a major source of pro- fessional failure. In addition to previous reasons, non- professionals may have been reluctant because they felt a lack of authority to probe further. Both groups may have felt the patron would resent their prob- ing further. They also may have been motivated by a desire to avoid the diffi- cult interview by not being clear on the nature of the problem and therefore not knowing what to ask and by the feeling that further interviewing would not reveal anything significant. In addition to these reasons, profes- sionals appeared to be reluctant due to lack of time ( they often worked alone), though there were some cases where the difficult interview was not at- tempted even though no other patrons were in sight. The strongest possibility, however, appeared to be that, probably due to constraints of the pressure of business in general, professionals had developed the habit or policy of pur- suing questions only up to a certain point-that point at which they could turn to a reference source for solution -and stopping short when the only re- course was a difficult interview where prospects of success seemed low. These results are in line with those of Dorman Smith.12 When attempts to obtain correct in- formation were made (by using person- al knowledge or by consultation of sources ) , they were successful 20 per- cent of the time for nonprofessionals and 83 percent of the time for profes- sionals. Failure on the part of the non- professionals appeared to be due primarily to failure to select the right source for verification. Professionals more frequently selected the right source but failed, due to pressure of 392 I College & Research Libraries • September 1977 business, to take sufficient time to exam- ine it carefully. One professional librarian twice se- lected the correct reference source but overlooked answers directly under her eyes, due to being in too great a hurry. In another case, she obtained the key in- formation from the patron but did not take sufficient time to examine it. Here again, the greater ability of profession- als to select the correct source could be due to library school training and pos- sibly to greater experience. Lack of time to interview and to consult sources ap- peared to be another major cause of professional failure. PERSONAL SuBJECf KNOWLEDGE An additional reason for nonprofes- sional failure to perform as well as pro- fessionals appeared to be lesser personal knowledge of the subject matter of the question, which handicapped them in recognizing faulty information and in knowing where and how to correct it. The professionals were superior in amount of personal subject knowledge, demonstrating familiarity with the sub- ject matter of 45 percent of questions, while nonprofessionals demonstrated familiarity with 24 percent. This lesser knowledge of the subject matter of questions on the part of nonprofession- als may be due, in part, to lesser educa- tion and perhaps to lesser experience. There is clear indication that knowl- edge of reference sources alone, while shown here to be important, was not sufficient for adequate performance on these types of questions. The profes- sional librarians in this study appeared to have superior know ledge of refer- ence materials; but on those questions where they had no personal subject knowledge, they achieved a success rate of only 31 percent. On the other hand, nonprofessionals who appeared to have less knowledge of reference materials had a success score of 80 percent on questions where they had personal knowledge. Personal knowledge also ap- peared to determine failure in the case of nonprofessionals who failed on 100 percent of questions where they had no personal knowledge. Professionals, how- ever, were successful on 31 percent of questions where they had no personal knowledge, due perhaps to greater knowledge of reference sources. The way in which personal know~dge aided both groups of staff members was as follows: 1. Misinformation was often detect- ed immediately, saving the verifica- tion process. 2. It helped suggest possible solutions, as when the librarian identified Massaponti as Maupassant, being aware that the latter was well known, widely read, and likely to be asked for. 3. If verification was still needed, it helped to selecting the best sources. On a question concerning the Masada, for example, the librarian knew it was famous enough to be verified in the encyclopedia under "Jews-History." 4. It saved the librarian from pro- ceeding on a false course. For ex- ample, because of personal knowl- edge, the librarian was not misled when told Tini Kling was a game. These findings suggest that lack of personal knowledge is a great handicap when dealing with "faulty informa- tion" questions. THE CONSULTATION AND REFERRAL PROCESs A significant reason for differential performance was nonprofessional fail- ure to utilize referral and consultation to the fullest. The concept of the in- formation center, staffed by nonprofes- sionals, but backed up by professional reference librarians who are available for consultation, is partially based on the premise that the less experienced personnel will refer whenever they fail • J to find the answer to a question. This study demonstrated that, in the case of "faulty information" questions, this premise is questionable. Out of twenty-one questions which nonprofes- sionals failed to answer, only six were referred-five to professional librarians and one by consultation with another nonprofessional. Thus the nonprofes- sionals referred or consulted on only 28 percent of those questions they were un- able to answer. Of those referrals or consultations, two, or 33 percent, were successfully resolved. On the six questions referred or con- sulted on, the following problems oc- curred: 1. The subject specialist was busy. 2. The subject specialist was not there. 3. The patron was referred to the wrong subject specialist. 4. T4e nonprofessional called the professional on the phone, but even though the professional knew the answer, due to a failure in communication, the patron did not get the information. 5. The nonprofessional repeated the patron's misleading information to the professional, who accepted it at face value and failed to recon- duct the interview. 6. The nonprofessional distorted the information slightly in transmis- sion. On eight unreferred cases, the non- professionals apparently did not refer because they did not detect the faulty information. On another three, they de- tected the faulty information but did not appear to know how to verify it. On another four, where lacking informa- tion rather than misinformation was the problem, they apparently did not refer because they did not realize that by using more in-depth interview tech- niques they could have obtained the key information. Basically, it appeared that they failed Reference Interview I 393 to refer because ( 1) they did not detect the faulty information, or ( 2) they be- lieved they had done all that could be done and that the professional librarian could not add anything to the answer. This is reinforced by the comment of one nonprofessional as she terminated the interview, "'I'm sorry, there is just no ,approach to this problem." They also may have failed to refer because they considered the question un- important, because they did not wish to disturb the professional librarian, or because they felt too many referrals might reflect on their capability. It also appeared that the nonprofessional sel- dom referred unless the professional li- brarian was in the vicinity. In both cases where nonprofessionals made successful referrals or consultations, the individ- ual referred to was in the immediate vicinity. Professional referrals ( two referrals and one consultation) were more successful ( 67 percent) than those of nonprofessionals ( 33 percent). The professional librarians, however, did not consult with each other as much as might have been expected. In three cases, the librarian was on duty alone (and in the others, the faulty informa- tion was not detected ) . The success score of both groups undoubtedly would have been greatly increased by more frequent consultation. The failure of the referral and con- sultation process in the information center setting suggests that if communi- cation problem questions are considered important, re-evaluation should be made of the information center concept. It also suggests the possibility that similar types and numbers of referral failures may occur on straightforward questions. CoNCLUSIONs In assessing the results of this study it is important to remember that it was not designed to evaluate the potential of reference personnel without library 394 I College & Research Libraries • September 1977 degrees. Instead, it was designed to com- pare the performances of the two groups under ordinary day-to-day condi- tions, "as it is," rather than "as it could be" at its best. The results of this study in regard to "indirect" questions indicate that pro- fessionals performed adequately on both types of "indirect" questions ( 91 percent and 90 percent). Nonprofes- sionals performed adequately on "broad subject" questions ( 73 percent) but failed to perform adequately on "wrong type of source" questions (50 percent). In regard to "faulty information" questions (misspelling and general faul- ty information), nonprofessionals did not perform adequately ( 28 percent success with referral or consultation). The success of professional librarians ( 67 percent with referral or consulta- tion), while perhaps adequate, also is below what we would hope for in pro- vision of top quality service. In addition to lack of orientation, the lower performance of nonprofessionals appeared to be primarily due to lesser personal knowledge of the subject mat- ter of the questions (perhaps, in turn, due to lesser education and/ or experi- ence) and lesser knowledge of refer- ence materials. Professional failures ap- peared to be due, on the other hand, to reluctance to attempt the question which depended primarily on the diffi- cult interview for solution and to lack of time to conduct proper interviews and examine reference sources to ob- tain correct information. Thus one effective program, in terms of successful resolution of communica- tion problem questions, would be that where interviewing is done by profes- sional reference librarians who then re- fer easier questions to nonprofessional assistants. Assuming that departmental policy encouraged responsibility for communication problem questions and that professionals were properly moti- vated and trained to deal with these questions, this type of arrangement might result in increased time for inter:- viewing and locating correct informa- tion. If this way is not utilized, other ways of improving service on communication problem questions suggested by this study are as follows: 1. Orientation and training in han- dling communication problem questions should be given, includ- ing training in techniques for gaining information in the reference interview. 2. Personnel should be encouraged to develop a sense of responsibility and concern for communication problem questions and should take pride in their successful solution. 3. In-service education should be pro- vided and encouraged, aimed at in- creasing subject knowledge and knowledge of reference sources. 4. Selection of nonprofessional per- sonnel for reference should be made both on the basis of high ed- ucational level and library science courses. In regard to referral and consulta- tion, this study suggests the following: 1. Professional personnel should be available in the immediate area at times that reference service is pro- vided. 2. Referral and answering of ques- tions by phone should be avoided. In cases where this is necessary the nonprofessional should not trans- mit the patron's information but should allow the professional to reconduct the interview ·over the phone. 3. The person to whom the question is referred should reconduct the interview, though taking pains to be brief and tactful. 4. A policy should be established to J refer all questions for which an- swers cannot be found or for which no approach to the problem can be discovered. 5. Personnel should avoid judging the boundaries of others' knowledge by their own boundaries. Thus, in- stead of concluding ''There is just no approach to this problem," the conclusion should be "I can think Reference Interview I 395 of no approach to this problem, but it is possible someone else could." 6. Frequent referrals should be en- couraged and rewarded as resulting from high standards and concern with good service. The attitude should be discouraged that refer- rals are an annoyance and reflect lac~ of competency and failure. REFERENCES 1. Marjorie Murfin, "A Study of the Refer- ence Process in a University Library," (Master's thesis, Kent State University, 1970), p.177-78, p.194. 2. Murfin, "A Study of the Reference Pro- cess," p.7l. 3. Dorothy Cole, ''Some Characteristics of Reference Work," College & Research Li- braries 7:49 (Jan. 1946). 4. Dorman Smith, "A Matter of Confidence," Library journal 97:1239-40 (April 1972). 5. charles Bunge, Professional Education and Reference Efficiency (Springfield, Ill.: Illi- nois State Library, 1967). 6. Joseph Wheeler and Herbert Goldhor, Practical Administration of Public Libraries (New York: Harper and Row, 1962), p. 324. 7. Edward ]estes and \V. David Laird, "A Time Study of General Reference Work in a University Library," Research in Librari- anship 2:16 (Jan. 1968). 8. Robert Balay and Christine Andrew, "Use of the Reference Service in a Large Aca- demic Library," College & Research Li- braries 36:25 (Jan. 1975). 9. Bunge, Professional Education and Refer- ence Efficiency, p.61. 10. Laura Boyer and William Theimer, Jr., "The Use and Training of Nonprofessional Personnel at Reference Desks in Selected College and University Libraries," College & Research Libraries 36:19 (May 1975). 11. Murfin, "A Study of the Reference Pro- cess," p.76. 12. Smith, "A Matter of Confidence," p.1239- 40.