College and Research Libraries Student Success in Using Microfiche to Find Periodicals Paul T. Adalian, Jr., Ilene F. Rockman, and Ernest Rodie The major purpose of this user survey was to study student usage of a microfiche serials hold- ings list to determine the accessibility and retrievability of periodical issues housed in four sepa- rate locations in the California Polytechnic State University Library. In-person interviews were conducted with 14 7 randomly selected students. Results showed that 80. 6 percent of the students were successful in their searches. This dispelled the need for the library to reorganize the arrangement of periodicals. • ith the rapid increase in costs for periodical subscriptions, and the unfortunate decrease in the buying power of library dol- lars for journals, concomitant with the dwindling physical space in which to house these journals, it comes as no sur- prise that within the past five years, nu- merous articles on periodical usage have appeared in the library literature. These studies can be categorized into four broad areas: (1) those that count the usage of periodicals to determine selec- tion, retention, withdrawal, shelving loca- tion (stacks or storage), and/or format (bound or microfilm) policies; 1 (2) those that chronicle the actual usage of periodicals-how they are used, and by whom; 2 (3) those that focus on the biblio- graphic retrieval skills of the users and their corresponding success and failure rates; 3 and (4) those that describe "user frustrations'' in accessibility and retriev- ability.4 Using these studies as background in- formation, the California Polytechnic State University Library, San Luis Obispo, decided to construct its own.peri- odical usage study. Rather than replicate previous projects, it was decided that a new study would be designed to enhance and expand prior research findings, as de- scribed in categories (3) and (4) above. The Cal Poly study would emphasize the retrievability and accessibility of peri- odicals and was designed (1) to test the ef- fectiveness and accuracy of a six- thousand-item computer-output-micro- fiche (COM) Public Serials List (PSL) on five individual microfiche cards; (2) to study the user success rate for finding, and the availability of, periodicals in four separate locations within a centralized li- brary building-in unbound format from a closed stack Current Periodicals Depart- ment; in bound format shelved by call number in the stacks and integrated with the book collection; in an oversized stack area shelved and integrated with the book collection; and· on microfilm within the centralized Reference Department; (3) to ascertain patron attitudes about the PSL system based upon ease of use, accuracy of information, readability of microfiche, and accessibility of microfiche viewers; and ( 4) to assess patron attitudes about the location of periodicals. Paul T. Adalian, Jr., is head of the reference department, Ilene F. Rockman is a reference librarian in the learn- ing resources and curriculum department, and Ernest Rodie is an M.B.A. student. All are at California Polytech- nic State University, San Luis Obispo, California 93407. 48 HYPOTHESIS Several hypotheses concerning periodi- cal usage and availability underscored the study. Numerous members of the library staff perceived that problems with the PSL format (lack of cross-references for title changes and abbreviated titles, omissions of titles held, call number errors, holding statement errors, and difficulty in discrim- inating between hard copy and microfilm issues) were contributing to its ineffective usage. Therefore, hypothesis one sought to test the relationship between PSL en- tries and user success. rates for reading a periodical entry correctly without the ben- efit of staff assistance. Staff members also expressed concern about the four locations of periodicals within the 200,000-square-foot, five-story library building. Consequently, hypothe- sis two sought to test for a relationship be- tween the location entry on the PSL and the user's ability to correctly find the peri- odical in the library. Because the library houses over 67,000 bound periodicals, close to 23,000 vol- umes of microfilm, and approximately 2,800 current unbound titles, the signifi- cance of the study could not be underesti- mated. METHODOLOGY . A face-to-face interview approach, rather than a self-reporting questionnaire, was chosen as the method of data collec- tion. This is the traditional form of survey research, as personal interviews are the standard by which survey research has gained its credibility. 5 Furthermore, the most valuable studies of catalog use have been conducted through interviews with users at the time they do the actual search- ing, rather than after the fact. 6 The sampling frame employed in this study is a nonprobability sample of the campus population. Of the 147 interviews conducted, 144 were usable. After design- ing and pretesting the questionnaire, the actual administration of the instrument took place between November 8 and 19, 1982. Days and times for interviewing were selected randomly to include both heavy and light use. All interviews were Student Success 49 conducted within the centralized Refer- ence Department of the library by a mem- ber of the specially trained survey re- search committee. The selection of subjects was randomly conducted to guard against bias. Inter- viewers selected the first person to ap- proach the PSL viewer at the beginning of the interview period. Persons already at the microfiche viewers at the start of this period were disregarded. This process was repeated upon the completion of each interview. Most people approached by the interviewer agreed to participate in the study, with only a handful electing not to answer the questions. The average inter- view time was between five and ten min- utes. This variation in time resulted from whether or not the patron was successful in searching for a periodical listing. To guard against bias and interviewer intru- sion, no help was given to the patron con- cerning the interpretation of the listing until the completion of the questionnaire. At the conclusion of each interview, those patrons who sought to physically retrieve the periodical(s) desired were followed by the interviewer. This was done to ascer- tain the actual success rate of periodical re- trieval within the library. RESULTS Characteristics of the sample are pre- sented in tables 1 and 2. Much to the sur- prise of the interviewers, the largest per- centage of library users (47.2 percent) were seniors in the School of Business (19.4 percent). The class level may be re- flective of the fact that all students must complete a senior project before graduat- ing from Cal Poly, although staff had as- sumed that lower division students en- rolled in formal library instruction classes, or those who experienced point-of-use in- struction by a reference librarian, might constitute a higher percentage of library users in the study. Because previous re- search studies had shown that business undergraduates were not frequent users of the book collection in a university li- brary, it was unexpected that these stu- . dents would be the largest group of peri- odical users. 7 As in the Golden, Golden, and Lenzini 50 College & Research Libraries TABLE 1 CHARACTERISTICS OF THE SAMPLE Level Number Percent Freshman 11 7.6 Sophomore 20 13.9 Junior 32 22.2 Senior 68 47.2 Graduate 7 4.9 Faculty 3 2.1 Other (visitors) 3 2.1 Total 144 100.0 TABLE2 CHARACTERISTICS OF THE SAMPLE School Affiliation Number Percent Afeculture and Natural esources 23 16 Architecture and Environmental Design 7 4.9 Business 28 19.4 Communicative Arts and Humanities 25 17.4 Engineering and Technology 18 12.5 Human Development and Education 20 13.9 Science and Mathematics 20 13.9 No affiliation 3 2.1 Total 144 100.0 studies, inferential and descriptive statis- tics were used to analyze the research results. Fundamental questions to be an- swered concerned patron success rates for matching a citation to a PSL entry, cor- rectly reading the bibliographic holdings statement on the PSL system, and accu- rately retrieving the desired periodical ti- tle from any one of four locations within the 200,000-square-foot library structure. Sources of patron's citations are shown in table 3. Reference from an index or ab- stract listing accounted for more than three-quarters of all the citations searched-78.5 percent. Less than 1 per- cent of all citations originated from a com- puter search printout. This fact could re- flect a lack of knowledge, on the part of the sample population, of the services offered by the library in the area of automated re- trieval, or simply user reliability on searching indexes and abstracts manually. Table 4 indicates that most patrons were successful in their searches. Of the 144 to- tal attempts, 116 (80.6 percent) were suc- cessful, versus only 28 (<19.4 percent) un- January 1985 TABLE 3 SOURCE OF CITATION Cate~o:l Number Percent Class assignment 4 2.8 Reference in index/ abstract 113 78.5 Bibliofaphy in a journal or boo 10 6.9 Footnote in a journal or book 3 2.1 Computer search 1 .7 Recommended by facu1ty or student 5 3.5 Other 8 5.6 Total 144 100.0 TABLE4 TITLE SEARCH ATTEMPTS Number Percent Successful 115 80.00 Unsuccessful 28 19.40 Invalid case 1 .60 Total 144 100.00 successful searches for the periodical title on the PSL system. Of these, an overwhelming 94.8 percent of the patrons were successful in deter- mining the call number, and 78.3 percent were successful in determining the loca- tion of the periodical. This success rate is contradictory to Murfin, who noted that only 50 percent of those who used a serial directory were able to use it correctly, 8 but supportive of Golden, Golden, and Len- zini, who found that 83 percent of those who used a serial card catalog were suc- cessful in locating an entry. 9 Why were the Cal Poly users more suc- cessful than others in locating a periodical entry, call number, and journal in the li- brary? Perhaps frequency of use was a fac- tor. To test the notion that a patron who frequently uses the PSL is more proficient in locating call number and title entries than one who uses it less often, a paramet- ric test of difference was employed with the analytic assistance of the Statistical Package of the Social Sciences (SPSS). At-test, a common tool for detecting sig- nificant differences between group means, was used; Table 5 summarizes the results. Group one was the group of suc- cessful patrons, and group two the unsuc- cessful patrons. The -mean (X) number of times the PSL had been previously used TABLE 5 PSL USAGE Number of Times Used X Successful patrons Unsuccessful patrons t-value = .25, df = 117 97 22 16 15 between the two groups was similar. A value of .25 indicates that the number of times a patron uses the PSL is not a deter- minant of whether -or not the search would be successful, which corroborates the finding of Golden, Golden, and Len- • • 10 Zffil. Looking at this hypothesis in a slightly different way, it might be expected that a patron who has never used the PSL sys- tem would have a higher rate of failure than one who had previously searched for title entries. However, when tested, there was no relationship between prior usage of the PSL and a successful search. This could reflect the fact that the PSL system is fairly easy to understand and functional even to the novice user. Golden, Golden, and Lenzini note that another factor that might influence the success of any search is the presence of a written citation. It could be assumed that a patron who simply · approaches a serials holdings list, and uses memory to recall a call number or location for a periodical ti- tle, will not be as likely to find a periodical in the library as one who has written down the citation. However, as in the Golden, Golden, and Lenzini study, no relation- ship existed between a written citation and the successful completion of the _ search. If a patron had been referred to the PSL from another source, it could be expected that he or she might have more accurate or complete information, which could im- prove the chances for a successful search. When tested, it was determined that refer- rals had no effect on the success of a pa- tron's search, whic~ again supports the results of the Golden, Golden, and Len- zini study. However, if a patron were looking for a periodical that had been previously used, it could be assumed that the chances for a successful search would be greater than in Student Success 51 a search containing a periodical never used before. The chi-square value (9.72) indicates that a definite relationship does exist between previous uses of a periodical and a successful search. This information is displayed in table 6. TABLE6 PRIOR USE AND SUCCESS OF SEARCH Entry Found Entry Not Found Total Prior use of periodical 52 (94.5%) 64 (71.9%) 116 No prior use 3 ( 5.5%) 25 (28.1%) 28 Total 55 89 144 x?- = 9.72, df = 1, sig. = .0018, phi = .278 One last factor examined was the influ- ence of prior instruction on the use of the PSL system. Of a total of 119 patrons who had used the PSL before, 81 stated that they had received some instruction, rang- ing from an explanation in a formal library class to simply a friend's assistance at the moment of the search. It was assumed that students with instruction would be more successful than students without in- struction. However, when tested, no rela- tionship existed between instruction and search success. This refutes the findings of Golden, Golden, and Lenzini, who found some relationship between these vari- ables, albeit a weak one (sig. = .07, phi = .09). 11 To learn more about the causes for suc- cessful searches, six factors were tested. These were: (1) frequency of PSL use, (2) prior use of the PSL system, (3) written ci- tation, (4) referral to the PSL, (5) prior use of the periodical, and (6) previous instruc- tion in the use of the PSL Only one of these factors, prior use of the periodical, contributed significantly to the successful search. However, it is possible that due to the interrelationships of all six factors, they could, taken together, influence the success of a given search more than any one single factor. LOCATING THE PERIODICAL Although 114 patrons were ·successful in locating the title of a periodical, only 91 chose to immediately find the periodical in the library. These individuals were fol- 52 College & Research Libraries lowed to one of several journal locations by the interviewer to determine if they could actually find the periodical without any assistance. The majority of search inquiries were successful-that is to say, when patrons went to the shelves for bound volumes, to the Microforms Room for microfilm reels, or to the Current Periodicals Department for unbound issues, the interviewer wit- nessed the successful retrieval of the jour- nal. For those searches that were unsuccess- ful, various reasons emerged. They in- cluded inaccurate location information on the PSL (43 percent), missing periodical is- sues (29 percent), and user errors (28 per- cent). FAILED PSL SEARCHES In examining the unsuccessful searches (19.4 percent}, four possible reasons emerged: (1) the library did not own the periodical; (2) user error, in which the pa- tron had an entry that was listed on the PSL but had failed to find it; (3) user error in writing the citation; and (4) PSL error, in which the needed material was owned by the library, but was either not listed on the PSL or was listed incorrectly. USER ATTITUDES In an attempt to learn more about user attitudes toward the PSL systems, two Likert-type, summated, even-integral rat- ing scales were developed. On the first scale, the patron was asked to rate four at- tributes of the system, but only if he or she had previously used it. These attributes were (1) easy to use, (2) accuracy of infor- mation, (3) readability of microfiche, and (4) accessibility of viewers. The results of the first rating scale required no direct in- terpolation; however, five important points do need to be underscored: (1) no one in the sample indicated a response of "very poor," (2) there was no significant difference between successful and unsuc- cessful patrons, (3) no neutral position was offered as a response choice, (4) most patrons could not make a comparison be- tween the present PSL system and any al- ternate choices, and (5) because some pa- trons had not used the PSL before, they January 1985 chose not to respond to some of the inter- viewer's questions. Results of the first scale show that 97 percent of all patrons who had used the PSL before found it easy to use and con- taining accurate information. Moreover, over three-fourths of the respondents found the information to be readable (80 percent), and almost all (98 percent) felt that the microfiche viewers were accessi- ble. The second Likert scale asked patrons to rate the present location of periodicals, if centralization would be an improvement over the current arrangement, and if a course in bibliographic instruction would be useful. This attitude scale was offered as a check on the actual usage to ascertain if any variations in responses occurred. It is interesting to note that, although 82 percent of the respondents found periodi- cals in the present location to be easy to find and accessible, 86 percent felt that centralization would be an improvement over the present four locations. In addi- tion, 88 percent felt that a library course in teaching the location of periodicals would be useful, although such a course now ex- ists. This information would seem to indi- cate that, although respondents are satis- fied with the current arrangement of periodicals, they still feel that there is room for location improvement. STAFF REACTIONS TO THE RESULTS Although the study was scientifically designed, and the methodology em- ployed was sound according to the con- ventions of survey research, members of the User Survey Committee had various reactions to the research results. The com- mittee was surprised at the high user suc- cess rate for finding title entries on the PSL. Upon checking the actual citations written on the survey instrument by the interviewer, it was learned that a majority of desired issues were. periodicals indexed in The Reader's Guide to Periodical Literature. These entries were fairly straightforward, and therefore, did not include compli- cated forms of entry often associated with · annuals, proceedings, transactions, and · other serial titles. However, when the to- tal number of periodical titles was tallied, it was learned that titles not listed in Read- er's Guide slightly outnumbered all others, despite the majority of issues emanating from the Guide. The high success rate was also questioned in light of the number of interlibrary loan (ILL) requests generated for titles and issues already owned by the library (203 out of 2,000 requests received over the past six months). In other words, if students could successfully use the PSL, why were they asking for titles already available in the library? After analyzing ILL requests, it was determined that one- third were problematic and concerned (1) association publications with no cross- references from the title of the publication to the entry under association, (2) incom- plete or incorrect holdings statements, (3) acronym titles filed at the beginning of each alphabetical sequence with no cross- reference, and (4) government publica- tions not listed on the PSL. Therefore, it was understandable that users could not. successfully locate these titles. However, the staff still wondered why two-thirds of the users who had searched the PSL still could not find the desired periodical. Per- haps the users simply did not search the holdings list or, in the realm of the thou- sands of times the PSL had been used over the past several months, the 136 titles not found should not be viewed as a large number. The Reference Department was sur- prised that no relationship was found be- tween prior instruction in the use of the PSL and success rate for firiding a title in the library. This phenomenon can be at- tributed to the fact that the Current Peri- odicals Department is located in a well- trafficked, clearly marked, and visible location in the library, and that users intui- tively assumed that recent titles could be found there by virtue of the descriptive name of the department. Moreover, al- though respondents were successful in finding back issues of subscriptions, it seemed that most users had a misconcep- tion that all back issues could be found on microfilm, due to the proximity of the Mi- croforms Room to the Reference Desk and to the index and abstract volumes. There- fore, it is perhaps through other factors Student Success 53 that users were successful in locating both current and microfilm issues of desired ti- tles. The need to examine more rigorously these factors, or what is called the SORE (subliminal osmotic random exposure) syndrome-the ability of students to con- duct successful searches by misconcep- tion and chance, enhanced by the location of relevant sources and service points- caused the committee to further inquire into the predominant form of those jour- nals that were found. Table 7 reports these results. By a margin of almost three to one, the most heavily used form of periodicals was microfilm. This fact does not support Murfin, who noted that the separation of microform and bound volumes resulted in user inability to locate desired volumes. u Moreover, in the present study, of those users who did find titles on microfilm, a majority had neither previous instruction in library use nor previous experience us- ing the PSL. Therefore, one can conclude that microfilm is not difficult to find in the library if a sound floor plan (with close proximity between abstracts and indexes, the PSL, and the Reference Service Desk) coexists with a highly visible Current Peri- odicals Department. TABLE7 FORMAT OF PERIODICALS FOUND Number Percent Current periodicals 15 21 Microfi1rit 42 60 Bound 13 19 Total 70 100 CONCLUSION Of the 144 usable interviews conducted, 80.6 percent of the patrons sampled were successful in their searches for informa- tion using the microfiche directory, the PSL. In addition, 94.8 p~rcent of those successful patrons were able to correctly determine the Library of Congress call number of the periodical, and 78.3 percent were able to determine correctly the loca- tion of the desired journal from one of four separate places in the library. Only one factor, prior usage of a periodical, proved to be a significant indicator of a successful ~------------------------------------------------------------------------ 54 College & Research Libraries search. The other factors-frequency of PSL use, prior use of the PSL system, a written citation, referral to the PSL from another area in the library, and prior in- struction in use of the PSL-were not sig- nificant by themselves. However, it is pos- sible that collectively they could influence the success of a search. Of those respondents who had previ- ously used the PSL, 97 percent found it to be easily used and to contain accurate in- formation, with 83 percent finding that periodicals were easy to find, although lo- cated in four separate areas within the centralized library structure. When the in- January 1985 terviewers followed the respondents to several periodical locations to witness the actual retrieval, 79 percent of those pa- trons were successful in their pursuits. Given these findings, it appears that the PSL is an effective and accurate tool for lo- cating periodicals, especially when the main Reference Desk is arranged near the Microforms Room, abstract and index vol- umes, and multiple copies of this micro- fiche periodicals holdings list. Therefore, the present arrangement of periodicals in four separate locations within the library will not be altered at this time. REFERENCES 1. Julia L. Long, "Journal Use-study in a VA hospital," in Ching-Chih Chin, Quantitative Measure- ment and Dynamic Library Service (Phoenix, Ariz.: Oryx Pr., 1978), p.95-102; Ruth Schwartz, "A Periodical Use Study," Illinois Libraries 60:106-109 (Feb. 1978); W. M. Shaw, Jr.," A Practical Jour- nal Usage Technique," College & Research Libraries 39:470-84 (Nov. 1978); Sheila Newlands Testa, "Journal Use in a Community Hospital," in Quantitative Measurement and Dynamic Library Service, p.129-35 . 2. Roger Flynn, "Use of Journals," in Use of Library Materials : The University of Pittsburgh Study (New York: Dekker, 1979), p.57-104. 3. Gary A. Golden, Susan U. Golden, and Rebecca T. Lenzini, "Patron Approaches to Serials: A User Study," College & Research Libraries 43:22-30 Oan. 1982). 4. Marjorie E. Murfin, "The Myth of Accessibility: Frustration and Failure in Retrieving Periodi- cals," Journal of Academic Librarianship 6:16-19 (Mar. 1980). 5 . Charles H. Backstrom and Gerald Hursh-Cesar, Survey Research, 2d ed. (New York : Wiley, 1981). 6. F. W. Lancaster, The Measurement and Evaluation of Library Services (Washington, D. C.: Information Resources, 1977), p.20 . 7. Jo Bell Whitlach, ''Library Use Patterns Among Full- and Part-time Faculty and Students,'' College & Research Libraries 44:146 (Mar. 1983). 8. Murfin, p .18. 9. Golden, p.24. 10. Ibid ., p .25. 11. Ibid., p.26. 12. Murfin, p .18.