hart.p65 454 College & Research Libraries September 1999 Scholarly Publication by University Librarians: A Study at Penn State Richard L. Hart The amount of scholarly publication that is required of academic librar­ ians is thought to vary among different types of colleges and universi­ ties. The present study looks in detail at the aggregated publishing record of the librarians at Penn State University, an institution that requires publication as a condition of continued employment. Findings indicate that Penn State librarians are quite productive in terms of the number of publications they contribute to the literature, and they are strongly com­ mitted to research. Evidence suggests that increasing demands for pub­ lication have served to influence both the quantity and the quality of librarians’ publications in recent years. umerous studies have de­ scribed various characteristics of academic librarians and their scholarly publications. In general, such studies look at a large group of publications from librarians who are employed at a wide variety of colleges and universities. Another strategy that may prove useful is to make a detailed study of the scholarly publications of li­ brarians at a single institution. This nar­ row focus is particularly appropriate when looking at trends in scholarly pub­ lication because the type of college or university in which a faculty member works has been shown to be a key factor in influencing his or her publication pat­ terns.1 This study attempts to further investi­ gate the subject of librarians as scholarly authors by focusing on the output of the librarians at a single university—Penn State University. Penn State librarians are required to publish in order to be success­ ful in their quest for promotion and ten­ ure, and anecdotal evidence suggests that these librarians believe the expectations for publication have increased dramati­ cally in the past few years. Because pre­ vious studies have shown that most aca­ demic librarians are not required to pub­ lish, Penn State librarians are not seen as being typical of academic librarians, and the findings of this study cannot be gen­ eralized to other colleges or universities. However, a detailed look at the librarians at a single institution should lead to fur­ ther understanding of academic librarians as a whole. Although teaching faculty at American colleges and universities have been pres­ sured to increase their production of scholarly books and articles in recent de­ cades, it is not clear that the same has been true for academic librarians.2 Studies have analyzed and described the library literature, and looked at academic librar­ ians as authors, but there is no clear-cut agreement that librarians face a “publish­ or-perish” scenario. In a study of authors of core journal articles, John M. Budd and Charles A. Seavey downplayed the im- Richard L. Hart is the library director at Penn State Erie, The Behrend College; e-mail: rlh@psulias.psu.edu. 454 mailto:rlh@psulias.psu.edu Scholarly Publication by University Librarians 455 pact of tenure pressure, concluding that “the publishing requirement in academic libraries quite clearly is not as widespread as may be commonly believed.”3 Simi­ larly, Betsy Park and Robert Riggs found that less than 18 percent of all colleges and universities require that librarians pub­ lish as a stipulation for promotion and tenure.4 Even at research and doctorate- granting universities, where publishing demands have traditionally been the most rigorous, it was recently found that only 64 percent of more than 200 campuses surveyed required their librarians to pub­ lish.5 Despite this, however, there is evi­ dence to suggest that librarians are pub­ lishing more, although any direct link to tenure demands has yet to be established. Lois Buttlar concluded in a 1991 study that academic librarians were publishing more than in the past, and she speculated that this was due to tenure requirements.6 More recently, Barbara J. Via noted that the dramatic increase in the number of library journals, which can be traced to the 1960s and 1970s, continues. In fact, in the past fifteen years, there has been a “veritable explosion of new periodicals devoted to ever-narrower sub-topics of li­ brary and information science.”7 Via sug­ gests that this explosion may be due in some part to the ethos of “publish or per­ ish.” Studies have analyzed and described the library literature, and looked at academic librarians as authors, but there is no clear-cut agreement that librarians face a publish-or-perish scenario. This study examines librarians and their scholarly publications at an institu­ tion where the values of publish or per­ ish are accepted. The focus of the study was twofold. First, it attempted to get a sense of the Penn State librarians’ com­ mitment to research and a description of their aggregate record of publication. Sec­ ond, it looked for evidence of ways in which publishing patterns might be changing, with a special emphasis on journal articles. It was hypothesized that, over time, the number and quality of pub­ lications by librarians have increased as a result of growing tenure pressures. Methodology A common research design of studies that examine publishing patterns among li­ brarians is to look at each article in the published library literature for a certain time period and for a specific group of journals. For instance, Paula D. Watson examined the articles in eleven library science journals over a five-year period, Buttlar collected data from sixteen jour­ nals over a two and a half years, and Budd and Seavey’s larger study investigated authors in thirty-six journal titles over five years.8–10 This approach seems particu­ larly good at providing a description of the state of the core library literature. However, one difficulty with this ap­ proach is that many librarian publications fall outside the journals under examina­ tion. For example, books, book chapters, and conference proceedings are omitted. Also omitted are journal articles in nonlibrary science journals, which have been shown to comprise nearly 25 per­ cent of all librarian journal publications.11 Other studies, rather than taking the jour­ nal literature as their starting point, be­ gin with specific librarians and examine all of their publications. Some studies, such as Pamela S. Bradigan and Carol A. Mularski’s study of academic health sci­ ences librarians, used a survey question­ naire combined with examination of each respondent’s list of publications.12 And others, such as Karen E. Pettigrew and Paul T. Nicholls’s study of library and information science faculty, have identi­ fied the names of their subjects and then collected publication data by means of a series of database searches.13 The population for this study com­ prised all librarians employed at Penn State as of the fall of 1998. Two primary methods of data collection were used. A short survey questionnaire was used to obtain both data on the librarians’ atti­ tudes toward publication and on certain http:searches.13 http:publications.12 http:publications.11 456 College & Research Libraries September 1999 demographic information such as years of experience and tenure status. In addi­ tion to the survey questionnaire, data on each librarian’s publications were col­ lected by examining the publications listed in each respondent’s vita. The de­ cision to use the vita as the primary means of publication information was based on the fact that Penn State librarians are known to keep their vitae, not only up­ to-date, but also on file in the offices of the university libraries. Along with the survey questionnaire, a release form was sent to each librarian. The release form gave the researcher permission to obtain the librarian’s vita from the university li­ brary files. Those librarians who preferred not to sign the release form were asked to send a list of publications when they returned the questionnaire. The Penn State University Libraries has a unique administrative structure in that it is composed of a system of librar­ ies found at twenty-four campus locations throughout the state. By far, the largest library in the system is located at the University Park campus, which is home to more than 40,000 students. It is less widely known that another 38,000 stu­ dents are found at the other twenty-three campuses, at locations such as Altoona, Erie, Hazelton, and Wilkes Barre. Gener­ ally, the libraries at these campuses serve populations ranging in size from 800 to 3,600 students and employ from two to six librarians. Librarians at all locations are subject to the same tenure and pro­ motion criteria, including the require­ ment to publish. Findings In September 1998, a survey was sent to all full-time, tenure-track librarians at Penn State, with a single follow-up letter to those who had not responded within a few weeks. Two librarians were elimi­ nated from the survey population because they were on leave. The response rate of 76 percent represents fifty-nine respon­ dents out of a possible seventy-eight li­ brarians surveyed. There appears to be some extremely modest evidence of re­ sponse bias in terms of campus location and possession of tenure. Librarians at the University Park campus responded at a lower rate (70%) than did those at other campus locations (84%). And tenured li­ brarians were somewhat less likely to re­ spond (73%) than were those without ten­ ure (79%). However, because campus location is strongly related to possession of tenure (74% of University Park cam­ pus librarians have tenure, compared to only 45% of those at other locations), it is unclear which of these factors influenced a librarian’s decision to respond to the survey or, indeed, whether these factors had any influence at all on the decision to respond. There was no difference in re­ sponse rate by gender. In general, it ap­ pears that the respondents were represen­ tative of the survey population. The questionnaire included a number of demographic variables that provided a description of the respondents. It re­ vealed that this group of librarians pos­ sessed considerable professional experi­ ence. When asked how many years they had worked full-time as a librarian, the answers ranged from one to forty-three, which averaged out to sixteen years of full-time experience. Most of the respon­ dents had worked at another library ear­ lier in their career, and twenty-seven of them reported having spent more than three years at another institution. How­ ever, as a group, most of their careers have been spent at Penn State, which accounted for nearly 70 percent of their total years of professional experience. This is impor­ tant because it was an assumption of this study that the publication patterns of these librarians reflected Penn State’s pro­ motion and tenure expectations. The fact that 70 percent of their professional ex­ perience was at Penn State lends support to this assumption. Commitment to Research The survey questionnaire included sev­ eral questions designed to get a sense of the librarians’ commitment to research. One question asked whether the respon­ dent was “currently engaged in scholarly Scholarly Publication by University Librarians 457 work which you expect to lead to a pub­ lication.” Another question asked how many hours per month the respondent spends “on work related to research or scholarly activity.” Eighty-three percent of the respondents indicated they were currently engaged in a scholarly project that they expected to lead to publica­ tion. The amount of time these librar­ ians reported spending on research was considerable. The results of this ques­ tion are shown in table 1. On average, the librarians spend 19.8 hours per month on their research, and fully 12 percent re­ ported spending more than thirty hours per month. Another survey question asked re­ spondents the extent to which they agreed with the statement that “the amount of research and publishing expected of Penn State librarians has increased over the past fifteen or twenty years.” (The question­ naire encouraged younger librarians to respond to this question “based on their understanding of the history of these types of expectations.”) A resounding 95 percent agreed (73% strongly agreed, 22% moderately agreed) that publishing ex­ pectations have increased over the past fifteen to twenty years. This is viewed in part as a measure of the individual librarian’s commitment to research be­ cause increased tenure pressures must be accepted and met if one is to remain em­ ployed at Penn State. Thus, these three indicators—the percentage of respon­ dents currently working on a research project, the number of hours spent on re­ search, and the nearly unanimous agree­ ment on increased publishing pressures— serve to characterize Penn State librarians as a group that makes a considerable com­ mitment to scholarly pursuits. Publishing Output Successful data collection for this study required not only the return of the ques­ tionnaire, but also examination of each respondent’s list of publications. A list of publications was obtained for fifty-seven of the fifty-nine respondents. Most re­ spondents signed a release form, and their TABLE 1 Librarians' Time Spent on Research Hours per Month Percentage of Librarians 2-10 32 11-20 35 21-30 21 more than 30 12 Total 100 vita was then obtained from university files. Some provided a copy of their cur­ rent vita when they returned the survey questionnaire. Two librarians responded to the survey but did not provide a list of publications. Several databases were searched in order to identify the publica­ tions of these two librarians, and the re­ sults have been included with those of the other respondents. This study did not include book re­ views as publications. Also omitted from the tally of publications were encyclope­ dia articles, entries in biographical dictio­ naries and similar brief reference contri­ butions, articles that appeared as a regu­ lar column or editorial in a journal, and translations. Although these types of pub­ lications were well represented in the re­ spondents’ vitae, the focus of the study was on those types of publications that are thought to be most commonly ac­ cepted for promotion and tenure: refer­ eed and nonrefereed journal articles, book chapters, conference proceedings, and four categories of book-length publica­ tions, including authored books, edited books, bibliographies, and technical re­ ports. The imprecise definition of what constitutes a “refereed” journal has been noted in other studies.14 In the present study, the respondents defined what were “refereed” or “nonrefereed” journals. Penn State librarians use a standard for­ mat for their vitae that requires them to distinguish between these two types of publications. Review of the titles the li­ brarians listed as either refereed or nonrefereed suggests that there is broad agreement among them as to these defi­ nitions. http:studies.14 458 College & Research Libraries September 1999 TABLE 2 Librarian Output by Type of Publication Librarians with % of Total Per Capita at least One Type of Publication Number Publications Publications Publication Refereed articles 203 38 3.4 85% Nonrefereed articles 128 24 2.2 69% Book chapters 75 14 1.3 49% Conference proceedings 75 14 1.3 44% Authored books 22 4 .4 20% Bibliography (books) 18 3 .3 18% Edited books 8 1.5 .1 8% Technical report 8 1.5 .1 12% Total 537 100.0 9.1 Table 2 provides a summary of the pub­ lications of the Penn State librarians. The table shows not only the total number and percentage of publications for each cat­ egory, but also the percentage of respon­ dents who had at least one publication of that type and the per capita publications per category. As would be expected, the most common type of publication is the journal article. The 203 refereed journal articles and 128 nonrefereed journal ar­ ticles combine to make up 62 percent of the total publications. This is remarkably similar to results of a study of academic health sciences librarians, which found that 68 percent of their publications were journal articles.15 Four categories of book- length publications combine for 10 per­ cent of the total. Book chapters and con­ ference proceedings, produced at an iden­ tical rate, combine for the remaining 28 percent. On average, the Penn State librarians reported publishing slightly more than nine publications each. This large per capita publication rate is substantially higher than what has been reported in previous studies. Aubrey Kendrick’s study of academic business librarians re­ ported an average of 1.4 items (exclud­ ing book reviews) per librarian over their careers.16 Bradigan and Mularksi found that health sciences librarians averaged 2.7 publications over a ten-year period.17 With an average of sixteen years of expe­ rience and slightly more than nine publi­ cations, Penn State librarians have a ca­ reer average of .56 publications per year. Changing Patterns Are patterns in publication changing? As noted above, in response to a survey ques­ tion, 95 percent of the librarians agreed that publishing expectations have in­ creased at Penn State in recent years. If that is indeed the case, further analysis of their publishing output may show that publication patterns are changing as a result of the institution’s rising expecta­ tions. It was hypothesized that not only would the number of publications in­ crease over time, but also that the quality of the publications, as defined by the number and percentage of refereed jour­ nal articles and by the number and per­ centage of journal articles found in the “core” library literature, would increase to meet rising tenure and promotion ex­ pectations. To analyze publishing output, all pub­ lications were placed into one of two cat­ egories: those published before 1990, and those published between 1990 and 1998. These categories were determined after the researcher had identified the year that represents the midpoint of this group’s aggregate years of experience as full-time li­ brarians. As a group, the librarians re­ ported having 943 years of combined ex­ perience. As noted above, the average http:period.17 http:careers.16 http:articles.15 Scholarly Publication by University Librarians 459 TABLE 3 A Com[arison of Earlier and Later Publications Type of Publication Before 1990 % of Total N Publications 1990 to date % of Total N Publications Change (%) Refereed articles Nonrefereed articles Book chapters Conference proceedings Authored books Bibliography (books) Edited books Technical reports 44 47 20 13 10 4 2 3 31 33 14 9 7 3 1 2 159 81 55 62 12 14 6 5 40 20 14 16 3 4 2 1 +261 +72 +175 +376 + 20 +250 +200 + 66 Total 143 100 394 100 +175 number of years of experience in 1998 is nearly sixteen years. The midpoint of this group’s experience (to the nearest year) is 1990; 464 years of their aggregate work­ ing lives as librarians fell before 1990, and 479 years fell between 1990 and 1998. This analysis makes the assumption that if the librarians’ publishing attitudes and be­ haviors had remained constant over time, the two categories of publications should look similar. However, it was hypoth­ esized that the two groups would not be similar, reflecting the changes in tenure pressure and the librarians’ increased scholarly productivity over time in re­ sponse to that pressure. Table 3 shows the various categories of publication divided by date. The per­ centage increase in each category is sub­ stantial, and the results provide rather dramatic evidence of the way in which publishing output has increased. Particu­ larly large percentage increases are noted for refereed articles, book chapters, con­ ference proceedings, bibliographies, and edited books. It is particularly interesting to note the change in em­ phasis placed on the im­ portance of refereed ar­ ticles as opposed to nonrefereed articles. Be­ fore 1990, nonrefereed articles made up slightly more of the total publi­ cations than refereed ar­ ticles (33% and 31%, respectively). How­ ever, after 1990, refereed articles outnum­ ber nonrefereed articles by a ratio of two to one (40% and 20%, respectively). The increase in refereed journal ar­ ticles, in terms of both the total number of articles and their proportion of total publications, serves as evidence not only of increasing publishing demands, but also of an improvement in the quality of the respondents’ publications over time. Further evidence of the quality of the jour­ nal publications can be gauged by the degree to which articles are found in the “core” library literature. Several authors have defined a core of library journals in efforts to analyze the library literature.18 For this study, Budd and Seavey’s list of thirty-six journal titles is used to define a core.19 It should be noted that most, but not all, of the thirty-six core titles are ref­ ereed journals. Table 4 summarizes Penn State librarians’ contributions to the core library literature for earlier and later pub­ lications. Their contributions to core jour­ nals have increased in terms of both their TABLE 4 Total Articles and Core Articles by Time Period Time Period Total Articles Core Articles (%) Published before 1990 91 24 (26.4) Published between 1990 and 1998 240 97 (40.4) http:literature.18 460 College & Research Libraries September 1999 total contributions and the percent of their articles that are part of the core. This lends additional support to the hypothesis that the quality of publications has improved over time. "Inside" versus "Outside" the Library Literature It was not the original intent of this study to look at the articles “inside” or “outside” the library literature, nor was it expected that nearly 60 percent of all articles pub­ lished between 1990 and 1998 would fall outside the core of library literature as de­ fined in this study. It had been assumed that with strong publishing expectations at Penn State, a higher percentage of jour­ nal articles would be found in the core. Where are the articles to be found if not in the core library literature? To answer this question, both refereed and nonrefereed articles were examined. The nonrefereed articles of the Penn State librarians, although outside the core as defined by thirty-six library science titles, are found almost exclusively in li­ brary-related publications. The journal most heavily represented in this category is College & Research Libraries News, with a total of twenty-one articles. Nine ar­ ticles, the second highest total for a jour­ nal title, were reported in the Pennsylva­ nia Library Association Bulletin. Similarly, many of the remaining articles are found in newsletters of various library organi­ zations representing specialized areas of librarianship, such as archives, art, and business. The refereed journal articles that fall outside the core represent a wide mix of publications. Fifty-five journal titles are represented. Of these, thirty-two titles are not indexed in Library Literature, indicat­ ing that they fall outside the subject areas of library and information science. How­ ever, the total number of articles found in these nonlibrary science journals is only thirty-seven since multiple contributions are made only infrequently to these out­ side journals. The subject areas covered by these journals are diverse. Examples of titles include American Entomologist, Civil Engineering Education, Hospitality and Tourism Education, Irish Slavonic Studies, Journal of Higher Education, Journal of the Air and Waste Management Association, and Teaching Political Science. However, most of the refereed articles outside the core are found in journals that are indexed in Li­ brary Literature; fifty-seven articles are found in twenty-three titles, including Acquisitions Librarian, American Archivist, Information Technology and Libraries, Medi­ cal Reference Services Quarterly, Serials Re­ view, and World Patent Information. Al­ though these articles lie outside the core, they are in refereed journals devoted to library and information science. Perhaps this should not be surprising, consider­ ing Via’s observation about the prolifera­ tion of library science journals with an ever-narrower focus. The picture that emerges from this study shows that Penn State librar­ ians must be viewed as a highly active group in terms of research and scholarly publication. It seems clear that there is a shift away from publishing in journals outside li­ brary science. Overall, 18 percent of the librarians’ refereed journal articles are found in journals that are not indexed in Library Literature. However, when divided by date, these comprise fully 36 percent of pre-1990 refereed articles, but only 13 percent of the total between 1990 and 1998. Conclusion The purpose of this study was to take a closer look at the characteristics of librar­ ians and their publications at an institu­ tion where publishing expectations are strong. Although the results may not be used to draw generalizations about other academic libraries, it is quite possible that similar findings would be observed at other colleges and universities where ten­ ure pressures have strengthened in recent years. The picture that emerges from this study shows that Penn State librarians Scholarly Publication by University Librarians 461 must be viewed as a highly active group in terms of research and scholarly publi­ cation. On average, each has produced slightly more than nine publications, in­ cluding articles, book chapters, confer­ ence proceedings, and books. Consider­ ing the level of scholarly output, it is not surprising that Penn State librarians re­ port spending an average of nearly twenty hours per month on research. Nor is it surprising that more than 80 percent report that they are currently at work on a project they expect to lead to publica­ tion. Moreover, it is not surprising that the librarians in this study have authored more publications than other studies have reported. This study focuses on librarians at a single institution where expectations are known to be strong; other studies usu­ ally include librarians at a wide variety of colleges and universities with a wide variety of expectations. This study helps to confirm the notion that scholarly pro­ ductivity is most strongly influenced by the context of one’s employment. This has been shown to be true for teaching fac­ ulty,20 and it has been suggested in the library literature that academic librarians publish more at institutions that require them to do so.21 At Penn State, there are strong expectations for publication and the substantial amount of publishing out­ put that this study documents is not viewed as an unexpected result. Penn State librarians present rather strong evidence that “the bar has been raised.” There was nearly unanimous agreement among these librarians that publishing expectations at the university have increased substantially in the past two decades. Examination of their publi­ cations has confirmed that a dramatic in­ crease in publication quantity and qual­ ity can be documented. All types of pub­ lications, including refereed and nonrefereed journal articles, book chap­ ters, conference proceedings, and books, are being produced at a greater rate. There has been a noticeable shift to refereed journals and away from nonrefereed jour­ nals, and a considerably higher percent­ age of articles are being found in the core library literature. Notes 1. J. Scott Long and Robert McGinnis, “Organizational Context and Scientific Productivity,” American Sociological Review 46 (1981): 422–442. 2. Jack H. Schuster and Howard R. Bowen, “The Faculty at Risk,” Change 17 (Sept./Oct. 1985): 13–21. 3. John M. Budd and Charles A. Seavey, “Characteristics of Journal Authorship by Academic Librarians,” College and Research Libraries 51 (Sept. 1990): 469. 4. Betsy Park and Robert Riggs, “Tenure and Promotion: A Study of Practices by Institu­ tional Type,” Journal of Academic Librarianship 19 (May 1993): 72–77. 5. Pamela S. Bradigan and Carol A. Mularski, “Evaluation of Academic Librarians’ Publica­ tions for Tenure and Promotion,” Journal of Academic Librarianship 22 (Sept. 1996): 360–365. 6. Lois Buttlar, “Analyzing the Library Periodical Literature: Content and Authorship,” Col­ lege and Research Libraries 52 (Jan. 1991): 38–53. 7. Barbara J. Via, “Publishing in the Journal Literature of Library and Information Science: A Survey of Manuscript Review Processes and Acceptance,” College and Research Libraries 57 (July 1996): 366. 8. Paula D. Watson, “Production of Scholarly Articles by Academic Librarians and Library School Faculty,” College and Research Libraries 46 (July 1985): 334–342. 9. Buttlar, “Analyzing the Library Periodical Literature: Content and Authorship,” 38–53. 10. Budd and Seavey, “Characteristics of Journal Authorship by Academic Librarians, 463– 470. 11. A. Neil Yerkey, “Publishing in Library and Information Science: Audience, Subjects, Af­ filiation, Source, and Format,” Library and Information Science Research 15 (1993): 165–184. 12. Pamela S. Bradigan and Carol A. Mularski, “Authorship Outlets of Academic Health Sci­ ences Librarians,” Bulletin of the Medical Library Association 80 (April 1992): 188–191. 13. Karen E. Pettigrew and Paul T. Nicholls, “Publication Patterns of LIS Faculty from 1982– 1992: Effects of Doctoral Programs,” Library and Information Science Research 16 (1994): 139–156. 462 College & Research Libraries September 1999 14. Via, “Publishing in the Journal Literature of Library and Information Science,” 367–370; Pettigrew and Nicholls, “Publication Patterns of LIS Faculty from 1982–1992,” 144. 15. Bradigan and Mularski, “Authorship Outlets of Academic Health Sciences Librarians,” 189. 16. Aubrey Kendrick, “A Comparison of Publication Output for Academic Business Librar­ ians with and without Faculty Rank,” Journal of Academic Librarianship 17 (July 1991): 147. 17. Bradigan and Mularski, “Authorship Outlets of Academic Health Sciences Librarians,” 189. 18. Buttlar, “Analyzing the Library Periodical Literature,” 39–40; Budd and Seavey, “Charac­ teristics of Journal Authorship by Academic Librarians,” 464; Stuart Glogoff, “Reviewing the Gatekeepers: A Survey of Referees of Library Journals,” Journal of the American Society for Informa­ tion Science 39 (Nov. 1988): 401. 19. Budd and Seavey, “Characteristics of Journal Authorship by Academic Librarians,” 470. 20. Long and McGinnis, “Organizational Context and Scientific Productivity,” 439–441.