College and Research Libraries JOHN N. OLSGAARD AND JANE KINCH OLSGAARD Authorship in Five Library Periodicals Authorship data for five major library science journals covering a ten-year period were reviewed. Information was tabulated to determine if publication trends suggest a bias on the basis of sex, occupation, or geographic location of the author. Results indicate that a higher than expected number of male authors were published and that a higher than average number of authors were located in the Northeast and Midwest regions of the United States. Occupational trends were also investigated, and recommendations for future research are discussed. THE STOCK IN TRADE OF LIBRARIANSHIP is communication and the transfer of informa- tion; yet little is known about the communi- cation of ideas within our own profession. Librarianship is sadly behind the disciplines of economics, psychology, and the sciences in determining the bibliometric nature of the professional literature. This field of in- quiry has become a topic of interest begin- ning with studies by Masse Bloomfield and Paula De Simone Watson concerning the characteristics of publishing librarians. 1 • 2 More recently O'Connor and Van Orden unde'rtook a survey of publishing policies of major hbrary journals, and Kim and Kim followed with a study of methodological changes in the literature. 3 • 4 The purpose of this study is to explore further the foundations of professional communication in the field of librarianship, by examining several aspects of authorship from selected library science journals. Spe- cific areas of inquiry include possible publi- cation bias on the basis of sex, occupation, and geographic location of the author. METHOD Five major journals were selected for this study: College & Research Libraries (C&RL). volumes 29-38, Library Journal (LJ) vol- john N. Olsgaard is documents librarian and archivist, I. D. Weeks Library, and Jane Kinch Olsgaard is head of reference, Lommen Health Sciences Library, both at the University of South Dakota at Vermillion. umes 93-102; Library Quarterly (LQ) vol- umes 38-47; Library Trends (LT) volumes 16-25; and RQ volumes 7-16. To be selected for inclusion in the study, the jour- nal must have been in existence for at least ten years, use an article format, and be rec- ognized as a nationally known journal of li- brary science. The five journals cited above met these qualifications. These journals were .also considered to have a major impact on librarianship and exhibit common trends in publishing. 5 All authored articles in these journals dur- ing the past ten years were used as data entries. Book reviews and letters were excluded. Each author was equivalent to one data entry; thus, multiple authorship ar- ticles were given multiple data entries. Each data entry consisted of the sex of the author, geographic location, and occupation. For the data entry, only the current infor- mation listed in the article byline was used; no additional or contrary information was added. To determine the sex of an author, an analysis of first names was undertaken in conjunction with the following rules: (1) First names that could be of either gender or that were not recognizable as being at- tributable to either gender were listed as indeterminable; and (2) first names of au- thors represented only by initials were listed as indeterminable. In analyzing the occupation of an author, the following guidelines were used: (1) If two occupations were listed as current, both I 49 50 I College & Research Libraries • January 1980 positions received full data value; and (2) for an author to be listed as a professional li- brarian, the individual must be a working professional with a specific title in an area of specialization. (For example, if an author was represented as "working at a librnry," the author was not listed as a librarian since the specific status could not be determined.) Geographic location was entered by indi- vidual state and then subgrouped into the following regions: 1. Northeast: Connecticut, Delaware , Washington, D.C., Maine, Maryland, Mas- sachusetts, New Hampshire, New Jersey, New York, Pennsylvania, Rhode Island, Vermont. 2. Southeast: Alabama, Florida, Georgia, Kentucky, North Carolina, South Carolina, Tennessee, Virginia, West Virginia. 3. Midwest: Illinois, Indiana, Iowa, Kan- sas, Michigan, Minnesota, Nebraska, North Dakota, Ohio, South Dakota, Wisconsin. 4 . Southwest: Arizona, Arkansas, Louisiana, Mississippi, New Mexico, Ok- lahoma, Texas. 5. West: Alaska, California, Colorado, Hawaii, Idaho, Montana, Nevada, Oregon, Utah, Washington, Wyoming. These regional delineations are based on the designations made by the 1978 ALA Committee on Accreditation of graduate li- brary school programs and are used in order to make the resulting data compatible with other research. Distributions on the basis of sex and occupation were derived using the formula X = a/(N - d), where X equals the repre- sentative percentage of specific data entries, a. equals the number of specific data entries in a journal, N equals the total number of entries in the journal, and d equals the number of entries for which data could not be determined. The formula was slightly al- tered in the cas~ of geographic distribution by in~luding entries that were not from the United 'States in d. PUBLICATION ON THE BASIS OF SEX Several levels of quantitative analysis were employed to derive an accurate dis- tribution. The first level of consideration dealt with the distribution by gender for data entries in a given journal (see table 1). A norm was generated for all librarians in the United States with regard to gender and data then gathered by gender for each of the five journals. The normative distribution for all women publishing should be 84 per- cent to match their share of the general li- brary population. All five journals failed to achieve the percentage, with -RQ having the highest women publishing percentage with 41.3 percent and LQ having the lowest with 21.2 percent. The second level of analysis was distribu- tion of publishing women academic librar- ians (see table 2). The normal level for women among the total population of academic librarians was 61.5 percent. A· similar study seems to verify this value by obtaining a distribution ratio of 61.9 percent for women librarians in colleges and univer- sities.6 Women academic librarians failed to publish up to the normal level in all five journals. C&RL published the highest per- cent with 39.6 percent. LT (24. 7 percent) had the lowest percentage. · For . the third level of inquiry into publi- cation on the basis of sex, only the data en- tries for faculty in schools of library science were used (see table 3). Again the trend for all five journals showed that women library TABLE 1 GENDER FOR ALL ENTRIES Women Men Journal (Percent) (Percent) N= d=* N-d= C&RL 30.5 69.5 495 49 446 L] 37.5 62.5 937 65 872 LQ 21.2 78.8 257 16 241 LT 32.7 67.3 463 41 422 RQ 41.3 58.7 445 21 424 National Averaget 84.0 16.0 *d is the number of entries for which data could not be determined. tRepresents the percentage of librarianship as a whole. U. S. Department of Labor. Bureau -of Labor Statistics, Library Manpower: A Study of Demand and Supply (Washington, D .C .: Govt. Print. Off., 1975), p. l2. Authorship I 51 TABLE 2 GENDER OF PUBLISHING ACADEMIC LIBRARIANS Women Men Journal (Percent) _(Percent) N= d=* N-d= C&RL L] LQ LT RQ National Averaget 39.6 60.4 25.1 74.9 31.1 68.9 24.7 75.3 35.4 64.6 61.5 38.5 249 24 225 169 6 163 48 3 45 107 6 101 158 11 147 *d is the number of entries for which data could not be determined . tRepresents the percentage of all U.S. academic librarians. Association of College and Research Libraries, Salary Structures of Librar- ians in Higher Education for the Academic Year 197~1976 (Chicago: American Library Assn. , 1976), p.6-12 . TABLE 3 GENDER OF PUBLISHING LIBRARY SCIENCE F ACULTI Women Men Journal (Percent) (Percent) N= d =* N-d= C&RL 12.5 87.5 80 8 72 L] 32.3 67.7 175 11 164 LQ 19.2 80.8 77 4 77 LT 29.2 70.8 96 7 89 RQ 33.9 66.1 61 2 59 National Averaget 40.8 59.2 '*d is the number of entries for which data could not be determined . tRepresents the percentage of all U. S. library science faculty . Russell E. Bidlack, " Faculty Salaries of 62 Library Schools, 1977-78," journal of Education for Librarianship 18:260 (Spring 1978). science faculty did not publish up to their normal level of the population (40.8 per- cent). The leader among the journals for publishing articles by women library science faculty was RQ with 33.9 percent. PUBUCATION BY OCCUPATION An objective of this study was to learn the occupation of the contributors to the five - journals (see table 4). Although each journal has a unique makeup, several · trends can be discerned by · comparison. All of the jour- nals, with the exception of LQ , had a sub- stantial number of articles by practicing li- brarians. The most notable example was RQ where 62.9 percent of all authors were pro- fessional librarians . Also of importance was the strength of the distribution of library science faculty , which ranged from 16.6 percent in C&RL to 30.4 percent in LQ. The prolific publishing rate of library science faculty is more sig- nificant when examined in light of various numeric relationships. A study by Russell E. Bidlack found 609 library science profes- sors in ALA-accredited United States library schools, 7 while the ACRL 1976 salary study recorded more than 13,000 academic librar- ians. 8 Although the numeric ratio of academic librarians to library science faculty TABLE 4 OCCUPATIONS OF AUTHORS Library Library Academic Public Other Science Science Other Nonlibrarian Librarian Librarian Librarian Faculty Student Faculty Nonacademic Journal (Percent) (Percent) (Percent) (Percent) (Percent) (Percent) (Percent) N= d=* N-d= C&RL 51.6 1.7 2.5 16.6 2.9 8.3 16.6 495 12 483 L] 18.9 16.3 8.7 19.6 3.2 4.4 28.9 937 43 894 LQ 19.0 4.4 4.0 30.4 4.0 17.0 21.3 257 4 253 LT 23.9 10.1 9.4 21.5 0.5 7.8 26.9 463 16 447 RQ 44.8 9.6 8.5 17.3 4 .0 4.5 11.3 445 92 353 *dis the number of entries for which data could not be determined. 52 I College & Research Libraries • ]anua-A} 1980 is approximately twenty-one to one, the publication rates for both groups are at the highest with four to one in C&RL and at the lowest with a ratio of one to two favor- ing library science faculty in LQ . In all cases, library science faculty carried a much higher percentage of publication than their population would indicate. A third area of importance deals with a significantly large number of authors who do not work in academia or any type of library setting. LJ published the highest percentage of these authors with 28.9 percent. and C&RL the lowest with 16.6 percent. PUBLICATION BY GEOGRAPHIC DISTRIBUTION To put the geographic data into proper perspective, a three-level approach was again used. The first level of consideration was to plot the geographic distribution for all data entries (see table 5). The trend in publication seemed to show the Northeast and Midwest regions of the country with a positive ratio of contributors in relation to their national norm. The Southeast and Southwest regions showed a negative ratio, while the West approximately matched the national average of publication for that re- gion. The second level of analysis, the geo- graphic distribution for academic librarians, again showed a positive ratio of contributors in the Northeast and Midwest regions (see table 6). The West showed a positive ratio of contributors for three of the journals and almost a matching ratio for the other two. The Southeast had a negative ratio for all five journals. The Southwest matched for one journal, RQ, and had a very negative ratio for the other four journals. The same ratio correlation holds true for the geographic distribution of library sci- ence faculty as it did for academic librarians (see table 7). SUGGESTIONS FOR FUTURE RESEARCH Journal literature· makes up 71 percent of the intradisciplinary communication in li- brarianship according to Nicholas and Ritchie. 9 The ability to publish in them is one criterion for professional advancement, and so a study of publishing patterns in li- brary science journals is merited. While this study attempted to describe TABLE 5 GEOGRAPHIC DISTRIBUTION: ALL ENTRIES Northeast Southeast Midwest Southwest West Journal (Percent) (Percent) (Percent) (Percent) (Percent) N= d=· N-d= C&RL 28.9 10.4 34.5 7.4 18.6 486 54 432 L] 49.3 11.3 20.0 3.0 16.6 937 200 737 LQ 27.6 4.3 50.4 4.7 12.9 257 25 232 LT 38.0 9.4 32.8 5.7 14.1 464 -59 405 RQ 37.5 12.4 28.8 8.2 13.2 445 42 403 Regional Averaget 26.9 17.9 28.1 11.9 15.2 •d is the number of entries for which data could not be determined . tRepresents percentages of general U.S . pupulation . Data derived from U. S. Bureau of the Census, Statistical Abstract , 99th ed. (Washington , D .C .: Govt. Pri nt. Off., 1978), p.34. TABLE 6 GEOGRAPHIC DISTRIBUTION: ACADEMIC LIBRARIANS Northeast Southeast Midwest Southwest West Journal (Percent) (Percent) (Percent) (Percent) (Percent) N= d=· N-d= C&RL 23 .8 11.7 32.5 8.3 23.8 249 9 240 L] 37.3 14.3 22.4 2.5 23.6 169 8 161 LQ 26.7 4.5 53.3 2.2 13.3 -48 3 45 LT 35.0 9.0 26.0 7.0 23.0 107 7 100 RQ 3S.5 11.8 27.0 11.2 14.5 158 6 152 Regional Averaget 31.3 17.3 24.7 11.1 15.6 •d is the number of entries for which data could not be determined. tRepresents percentages of all U.S. academic librarians . Data derived from U. S. Department of Health , Education and Welfare. Na- tional Center for Education Statistics, Library Statistics of Colleges and Universities, Fa/11975: Institutional Data (Washington, D .C .: Govt. Print. Off., 1977), p.221-78. Authorship I 53 TABLE 7 GEOGRAPHIC DISTRIBUTION: LIBRARY SCIENCE FACULTY Journal C&RL L] LQ LT RQ Regional Averaget Northeast (Percent) 24.7 60.1 22.2 26.7 41.1 33.8 Southeast Midwest (Percent) (Percent) 13.7 41.1 15.0 19.0 2.8 58.3 11.1 43.3 12.1 31.0 17.7 25.9 Southwest West (Percent) (Percent) N= d=' N-d = 4.1 16.4 80 7 73 0.7 5.2 175 22 153 0.0 16.7 77 5 72 6.7 12.2 96 6 90 3.5 12.4 61 3 58 7.9 14.6 *d is the number of entries for which data could not be determined. tRepresents percentages of all U.S. library science faculty . Russell E. Bidlack, "Faculty Salaries of 62 Library Schools, 1977-78," ]ourool of Education for Librarianship 18:263 {Spring 1978). the pattern of authorship in five journals, it is beyond its scope to speculate on the rea- sons why these patterns exist. Future re- searchers may wish to go to the journals and determine if the distribution ratios hold true for the collective body of submitted manu- scripts, although, as O'Connor and Van Orden describe it, precious little informa- tion is currently available in this area. 10 If we accept the truism that knowledge of the communication process itself will partially constitute the value we place on the infor- mation being conveyed, then research in bibliometrics is indispensable. REFERENCES l. Masse Bloomfield, "The Writing Habits of Librarians," College & Research Libraries 27:1~19 (March 1966). 2. Paula De Simone Watson, "Publication Activ- ity among Academic Librarians," College & Research Libraries 38:375--84 (Sept. 1977). 3. Daniel O'Connor and Phyllis Van Orden, "Getting into Print," College & Research Li- braries 39:38~96 (Sept. 1978). 4. Soon D. Kim and Mary T. Kim, "Academic Library Research: A Twenty Year Perspec- tive," in Robert D. Stueart and Richard D. Johnson, eds., New Horizons for Academic Libraries (New York: K. G. Saur, 1979), p.375-83. 5. It should be noted that Library Quarterly conference issues consist of commissioned ar- ticles only. 6. Data derived from U.S . Department of Health, Education and Welfare. National Center for Education Statistics, Library Statistics of Colleges and Universities, Fall 1975: Institutional Data (Washington, D.C. : Govt. Print. Off., 1977), p.221-78. 7. Russell E. Bidlack, "Faculty Salaries of 62 Library Schools, 1977-78," journal of Educa- tion for Librarianship 18:263 (Spring 1978). 8. Association of College and Research Librar- ies, Salary Structures of Librarians in Higher Education for the Academic Year 1975-1976 (Chicago: American Library Assn., 1976), p.~12. 9. David Nicholas and Maureen Ritchie, Litera- ture and Bibliometrics (Hamden, Conn.: Linnet Books, 1978), p.62. 10. O ' Connor and Van Orden , "Getting Into Print," p.38~96 passim.