College and Research Libraries Research Notes The Literature of Academic Libraries: An Analysis John M. Budd Like all literature, the literature of academic libraries exhibits certain charac- teristics with regard to source journals and related subjects. Citations within this literature are distributed according to format of material, age, journals and individuals Cited. The present study uses the ERIC database on compact disc to investigate these characteristics. Among the findings is the realization that the citations are primarily to journals and that both authorship of articles and citations to individuals are quite dispersed. Also, while there is some evidence of interdisciplinarity, this aspect of the literature is not very well developed. • he literature related to aca- demic libraries is diverse. It is as varied as the work of the . . libraries, their collections, their personnel, their technology, and the communities they service. As economic challenges, organizational complexity, and computer applications (designed to enhance access, use, and analysis) com- plicate the world of academic libraries, it might be expected that the literature re- flects the related difficulties and oppor- tunities. The sizable literature illustrates both interest in the subject and possible idiosyncrasies of access to it. The present study examines that por- tion of the literature that can be accessed through ERIC on compact disc. ERIC is selected because it indexes many of the major library and information science journals, but also because its coverage is broader and includes such . areas as higher education. Through use of ERIC, a researcher can construct a database of the literature of academic libraries and study the characteristics of the literature. Database users can identify titles of source journals, authors of articles, · and related descriptors. They can also exam- ine citation characteristics, including format and age dispersion of cited works and most frequently cited authors and titles. Such an examination can deter- mine the scope of the literature, the elements that recur in written communi- cation, and the individuals actively contrib- uting to the literature. The creation and use of such a database permit unobtrusive study of the concerns of any one area of academic librarianship as reflected in its literature. This assumes that the literature discusses or debates John M. Budd is Assistant Professor at the Graduate Library School, University of Arizona, Tucson, Arizona 85719. 290 major issues, and that the questions and problems identified by those interested in the field appear in the literature. THE STUDY This article's investigation begins with a set of questions regarding the pub- lished literature and citations in it: • What journals can be identified as hav- ing articles relevant to academic librari- anship? • What are the most frequently occur- ring subjects of articles in this segment of the field? • What formats of materials are cited in the literature? • What is the age dispersion of cited ma- terials? • What specific titles are most frequently cited in the literature of academic librar- ies? • Which individuals are most frequently cited in this literature? An examination of the literature must first establish a set of parameters defining the subject area and time frame. The use of compact disc products such as ERIC-CD, made available through Silverplatter, fa- cilitates this task. 11 Academic libraries" is an operative descriptor in ERIC, employ- able as the central access point to answer the above questions. The relevant time frame as a search parameter was the five- year period of 1984 through 1988. Of course, materials included in ERIC are as- signed descriptors by a variety of indexers, so some unevenness may be evident. ERIC also has levels of indexing, with articles assigned some major descriptors. Descrip- tors attached to articles indicate a connec- tion, not necessarily strong, with regard to the content of academic librarianship. Major descriptors are intended to indicate a relatively strong connection between the descriptor used and the content of a given item. Another parameter of this study is the limitation of output to journal articles, the most formal means of communication in most disciplines. From 1984 through 1988, 1,165 articles were assigned the descriptor II academic libraries." However, with ERIC's indexing policy, an indexed article need only make some mention of aca- Literature of Academic Libraries 291 deinic libraries to be assigned this de- scriptor. 11 Academic libraries" as a major descriptor was attached to 328 articles. This set of papers forms the basis of the present study. · The 328 articles can be analyzed in a number of ways, but potential idiosyncra- sies of this set need to be stated. Some characteristics are alluded to above; the assignment of descriptors is at the discre- tion of an indexer. Recognition of the sub- ject of academic libraries may or may not be consistent. The extent of coverage (jour- nals indexed and the degree of coverage of each title) is also at the discretion of ERIC. Margaret F. Stieg and Joan L. Atkinson point out discrepancies in the level of coverage of specific titles among the com- pact disc products of ERIC, Library and Information Science Abstracts (LISA), and Library Literature. 1 For instance, the per- centage of coverage of articles in 1984 in College & Research Libraries was 7 4% for ERIC, 72% for LISA, and 95% for Library Literature; in the Journal of Academic Librar- ianship, coverage was 77% for ERIC, 92% for LISA, and 100% for Library Literature. On the other hand, the coverage of Library Quarterly was more complete for ERIC (91 %) than for LISA (81 %) or Library Lit- erature (75%). Of course, coverage of spe,.. cific titles over time is subject to change, as Stieg and Atkinson note. Because ERIC is a widely held and widely used tool and is multidisciplinary, it is the source for the list used here. Analysis focuses on the characteristics outlined in the introduction and set of questions above. In part, this entails anal- ysis of the citations included in the source papers. Sixty of the 328 articles contained no references, and 6 more were not avail- able for analysis; therefore, citations in 262 papers are studied. FINDINGS The first question considered relates to those journals that have published articles relevant to academic libraries. The 328 ar- ticles are dispersed over 50 separate jour- nals. Forty of these are library- or information science-related titles; 10 are from other disciplines-5 from higher ed- ucation, 4 from computing technology, 292 College & Research Libraries TABLE 1 JOURNALS WITH MOST ARTICLES ON ACADEMIC LIBRARIES Title No. College & Research Libraries 76 Journal of Academic Librarianship 44 RQ 21 Library Journal 18 International Library Review 17 Catholic Library World 12 Library Trends 12 New Directions for Teaching and Learning 10 Library & Information Science Research 9 Librar:!J. Quarterl:!J. 9 and 1 from journalism. As might be ex- pected, the 2 top-ranked journals are Col- lege & Research Libraries and Journal of Academic Librarianship. This fact is consis- tent with the findings of Mary K. Sellen in a more limited citation analysis.2 The most heavily represented journals are presented in table 1. The authors of these papers are even more widely dispersed. Three hundred different individuals are represented-282 by a single article each. Of the 18 authors responsible for multiple contributions, only 4 have written more than two arti- cles each. Because of the somewhat less than exhaustive coverage of some jour- TABLE 2 MOST FREQUENTLY OCCURRING DESCRIPTORS Descriptor No. Library services 52 Librarians 49 Library role 42 Library administration 38 Library instruction 37 Library collections 34 Public libraries 28 Research libraries 24 Reference services 22 Library collection development 20 May 1991 nals by ERIC and the limitation to major descriptors, the researcher cannot neces- sarily infer that there are only 18 multiple contributors to this subfield. It is likely that some of the authors have written articles that are at least tangentially related to aca- demic libraries and are not included in this study population. The second basic question of this study focuses on additional subjects of articles dealing with academic libraries. The major descriptor "academic libraries" is cer- tainly not the only one assigned to these articles. In fact, a total of 406 different de- scriptors is attached as index terms. While some of these other descriptors are outside of the library and information science realm (such as "developing nations," "fu- tures of society," and "college faculty''), most are in the library mainstream. The 10 most frequently occurring descriptors reflect this trend; they are presented in table 2. TABLE3 FORMATS OF MATERIALS CITED Format No. % Cum.% Journals 2,088 56.3 56.3 Books 1,064 28.7 85.0 Reports 179 4.8 89.8 Theses and Dissertations 69 1.9 91.7 Proceedings 59 1.6 93.3 ERIC 47 1.3 94.6 Miscellaneous 202 5.4 100.0 As stated above, 262 articles can be ex- amined with regard to their citations. Spe- cifically, the 262 articles include 3,708 citations, or an average of 14.2 citations per article. The third question addresses the formats of these cited items; the intent is to discover the kinds of materials that are incorporated into formal communication in the field. The journal article is the most common format of cited material, with books a . distant second. Other formats, such as reports, theses, and proceedings, appear, but not in great numbers. The dis- persion by type of material appears in table 3. These results agree with those of Sellen. She found that 56% of the references in College & Research Libraries and 54% of those in Journal of Academic Librarianship were to journals (for an overall percentage of 56%).3 She also reported that 39% of College & Research Libraries references and 36% of those in Journal of Academic Librari- anship (38% overall) were to monographs. This percentage is higher than that for books in the present study, but the differ- ence may be due, at least in part, to collaps- ing categories. For instance, Sellen may have considered proceedings or reports as monographs. The journal article is the most com- mon format of cited material, with books a distant second. The dispersion by format of these cita- tions is a bit different from that in other social science or humanities fields. For instance, in American studies, sociology, American literature, and English litera- ture there is greater reliance on books than on periodicals. 4 In the sciences, though, there is a much higher propor- tion of periodical literature among cited works. 5 This does not imply that writ- ings on academic libraries are more sci- entific than are those on sociology, but it does indicate the formats likely to find their way into formal communication channels in the area under scrutiny here. TABLE4 AGE DISPERSION OF CITED WORKS Age (Years) No. % Cum.% 0-5 1,900 51.6 51.6 6-10 935 25.4 77.0 11-15 359 9.8 86.8 16-20 189 5.1 91.9 21-25 77 2.1 94.0 26-30 48 1.3 95.3 31-35 33 0.9 96.2 36-40 19 0.5 96.7 41-45 15 0.4 97.1 46-50 26 0.7 97.8 51+ 81 2.2 100.0 Literature of Academic Libraries 293 TABLE 5 MOST FREQUENTLY CITED JOURNALS Title No. College & Research Libraries 291 Library Journal 132 Journal of Academic Librarianship 131 RQ ~ Library Quarterly 65 Library Trends 60 American Libraries 55 Library Resources & Technical Services 46 College & Research Libraries News 38 Special Libraries 35 Library & [nformation Science Research 33 Collection Management 27 American Archivist 26 Chronicle of Higher Education 25 Wilson Library Bulletin 25 Journal of the American Society for Information Science 24 Drexel Library Quarterly 23 Advances in Librarianship 20 Information Technology & Libraries 20 Canadian Library Journal 17 Journal of Documentation 17 " Includes citations to the title Library Research. Examination of the age dispersion of these cited works, the focus of the fourth question, also helps define the kinds of materials incorporated by authors. The majority of citations are to recent mate- rials. Seventy-seven percent (2,835) of them are to works 10 years old or less. Table 4 illustrates the dependence on re- cent, as opposed to older, materials. The dependence on recent materials is slightly more pronounced within cita- tions to journals. A total of 54.9% of the citations are to journal issues 0 to 5 years old. Thereafter, there is little diversion from the overall age dispersion. Data gathered by Sellen also demonstrate re- liance on recently published materials.6 With regard to age dispersion, the litera- ture on academic libraries is somewhat 294 College & Research Libraries similar to that on sociology (where nearly 70% of the citations are to items 10 years old or less) and to that on micro- biology (where just over 82% of the cita- tions are to items 15 years old or less). 7 Which titles are most frequently cited? Not surprisingly, College & Research Li- braries and Journal of Academic Librarian- ship receive substantial numbers of citations. These titles rank first and third, respectively, with Library Journal ranking second. The most frequently cited jour- nals appear in table 5. The results of this analysis can be com- pared to a previous study by Robert Swisher and Peggy C. Smith. 8 They sur- veyed ACRL members in 1978 and de- rived a list of the 26 most frequently read journals. Fourteen titles are common to both lists, including 13 of the top 15 iden- tified by Swisher and Smith. TABLE 6 MOST FREQUENTLY CITED INDIVIDUALS Name No . Patricia Battin 19 F. W. Lancaster 16 William E. McGrath 15 Barbara B. Moran 14 Evan Ira Farber 13 Kathleen M. Heim 13 Maurice P. Marchant 13 Charles R. McClure 13 Herbert S. White 13 Maurice B. Line 12 Anita R. Schiller 12 Richard DeGennaro 11 Peter Hernon 10 Edward G. Holley 10 Ching-Chih Cheng 9 Elizabeth Frick 9 Allen Kent 9 Anne K. Beaubien 8 Richard M. Dougherty 8 Mary Jo Lynch 8 John N aisbitt 8 Iohn N. Olsgaard 8 May 1991 The list in table 5 exhibits a preference on the part of authors for citations to journals in library and information sci- ence. Only two titles, American Archivist and Chronicle of Higher Education, can be considered outside the field, but their subjects are related to librarianship, es- pecially academic librarianship. While the most frequently cited titles tend to be within the field, some journals in other disciplines receive citations. In fact, of the 401 different journals cited, 220 (54%) lie outside of library and informa- tion science. The most frequently repre- sented subjects are general education, management, social science research, and higher education. Although anum- ber of titles are cited, these nonlibrary journals account for only 22% (471) of the total number of citations to journals. The degree of interdisciplinarity in the literature of academic libraries can be interpreted as broad, but not necessarily deep. One other area for analysis is that of most frequently cited authors. Because the most frequently cited journals are primarily in the field of library and in- formation science, it might be expected that the most frequently cited individu- als work in this field as well. The list of such individuals in table 6 reveals that only trend analyst John Naisbitt works outside the field. The data set is too narrowly focused and the coverage of journals on the part of ERIC not extensive enough to allow exhaustive analysis, such as adherence to Lotka' s law. SUMMARY The analysis presented here illustrates certain characteristics of the literature pertaining to academic libraries and of the materials incorporated into the liter- ature by the author. For instance, the dispersion of authorship of articles on the subject is very broad, with few indi- viduals making multiple contributions. The related descriptors point out that subjects of the published works are sol- idly in the field of library and informa- tion science. Citation analysis reveals the heaviest use of the periodical literature for documentary support and a reliance on the internal literature of the field. Au- thors draw predominantly from recently published materials, with 77% of cited works being 10 years old or less. This brings up two questions for further in- vestigation: Is the literature of this field cumulative? Is the bulk of the body of knowledge in the discipline incorpo- rated into the current literature? The most frequently cited journal titles also show a dependency on the literature of librarianship. While the interests of academic librar- ians may, perforce, be broad, analysis of Literature of Academic Libraries 295 the literature exhibits a vision that is less interdisciplinary than might be ex- pected. This is not to imply that the liter- ature is failing the profession, but discussions of inter- or multidisciplinar- ity are not reflected to any great extent in the cited works in the literature of aca- demic libraries. Such areas as systems and computer science, organization the- ory, sociology, and even information sci- ence are not heavily represented. The question remains: Should they be? A more pertinent question may be: To what extent should they be represented? Seek- ing an answer to that question could be fertile ground for future research. REFERENCES AND NOTES 1. Margaret F. Stieg and Joan L. Atkinson, "Librarianship Online: Old Problems, No New Solutions," Library Journal113:54 (Oct. 1, 1988) . 2. Mary K. Sellen, "Bibliometrics in Information Science: A Citation Analysis of Two Academic Library Journals," College & Research Libraries 45:129-32 (Mar. 1984). · 3. Ibid., p.129. 4. Charles Avery Bolles, "Characteristics of the Literature of American Studies as Indi- cated by Bibliographic Citations" (Ph.D. diss., Univ. of Minnesota, 1975); James C. Baughman, "A Structural Analysis of the Literature of Sociology," Library Quarterly 44:296 (Oct. 1974); John M. Budd, "Characteristics of Written Scholarship in American Literature: A Citation Study," Library & Information Science Research 8:203 (Apr.-June 1986); Richard Heinzkill, "Characteristics of References in Selected Scholarly English Literary Journals," Library Quarterly 50:352-65 (July 1980). 5. Herman H. Fussier, "Characteristics of the Research Literature Used by Chemists and Physicists in the United States," Library Quarterly 19:19-35 (Jan. 1949); Herman H. Fussler,"Characteristics of the Research Literature Used by Chemists and Physicists in the United States. Part II," Library Quarterly 19:119-43 (Apr. 1949); J. E. G. Craig, Jr., "Characteristics of Use of Geology Literature," College & Research Libraries 30:230-36 (May 1969); and James E. Kanasy, "Citation Characteristics and Bibliographic Control of the Literature of Microbiology," (Ph.D. diss., University of Pittsburgh, 1971). 6. Sellen, "Bibliometrics in Information Science," p.130-31. 7. Robert N. Broadus, "Analysis of Literature Cited in the American Sociological Review," American Sociological Review 17:355-57 (June 1952); Robert N. Broadus, "A Citation Study for Sociology," American Sociologist 2:19-20 (Feb. 1967); Kanasy, "Citation Char- acteristics." 8. Robert Swisher and Peggy C. Smith, "Journals Read by ACRL Academic Librarians, 1973 and 1978," College & Research Libraries 43:56 (Jan. 1982).