This paper examines what is known about professional communication in the field of library and information science, particularly for journal articles. We know a good deal about articles, their authors, their citing behaviour and the topics researched. Two main research methods have been used to study the articles: bibliometrics, which includes citation analysis, and content analysis. There have also been studies of the use made by readers of journal articles.
This paper reviews studies about the articles in library journals. From these studies we know a good deal about the articles, their authors and their citing behaviour, and topics researched.
When we come to studies of use of the journal articles, findings are rather depressing, because diffusion of information to improve library and information services is not so widespread nor internati onal as we would wish. Two research methods, bibliometrics and content analysis, have been widely used to study professional or scholarly communication, and the findings from studies using these met hods will be briefly reviewed. Other methods such as survey questionnaires and interviews are also used.
There have been bibliometric studies of individual journals, for example a study of the first 24 volumes of Journal of Education for Librarianship published from 1960 to 1984. From 1971 articles wer e referred. It was the main journal, published in the USA, for communication for library science academics. The articles published and their authors were examined, and also a bibliometric study mad e of the citations in those articles. Were there changes over time?
The main subjects addressed (using subject indexes) were determined. In the early days there was usually sole authorship, by the early 1980s, one out of three articles had joint authorship. Over 80 % of first authors contributed only 1 article in the 24 years. Other findings were that 7 out of 10 first authors were educators, 9 out of 10 were Americans, and 2 out of 3 were male. As the averag e number of pages in an article increased, so did the number of citations; by the 1980s the average was 17 citations per referenced article. Authors rely on current materials, over half of the cita tions were to material published in the last 5 years, and over 70% to material less than 10 years old. Seventeen journal titles, 6% of those cited, received 60% of all citations to journals; and a l arge number of journal titles received only 1 citation. A few authors received many citations, while many authors received only a few.
One area of bibliometric research has been citation analysis or the study of citations or references made by authors in their writings. Analysis has been made easier by the availability of online ci tation indexes from the Institute of Scientific Information: Citation Index, the Social Sciences Citation Index and the Arts and Humanities Citation Index. They cover mainly journal literature from western countries, in English. Unfortunately library and information science journals are insignificant numerically in the wider sphere of publishing, so only mainstream library journals from the U SA and UK and a few international journals are included. Not one Australian library journal is included in the journal database for the Social Sciences Index, so we have to go back to manual methods .
The norms of scholarly work require authors to cite the publications that they have found useful in carrying out their own work. The citation analyst thus assumes that the citations made by an autho r are an indicator of the value of these writings to his own work (MacRoberts and MacRoberts 1989, p342). There are problems: authors cite themselves frequently, they may neglect to cite important papers, only secondary sources are cited, rather than the original paper.
A recent study has looked at the rate of self citation in library and information science journal articles (Dimitroff and Arlitsch 1995). The 1,058 articles analyzed were for a period of 18 months, 1992 and the first half of 1993, drawn from 28 English language library and information science journals. They were mainly American titles, with some British and international titles included. A se lf citation rate (articles that contain one or more self citations compared to the total number of articles in the sample) of 50% was found, higher than for the sciences and social sciences. When th e percentage of self citations as related to the total number of citations for the articles in the sample, 6.6%, was compared, this was lower than the percentage for the sciences, and higher than tha t, 3.4%, for the social sciences.
Also in citation analysis no account is taken of the affirmative or negative citations; thus there is no account taken of quality of the material cited. Different disciplines have different norms o f citing practice. There is also the problem of giving credit for papers with several authors. (MacRoberts and MacRoberts 1989). All this means there are many possibilities for error in studies us ing citation analysis. As A Neil Yerkey notes: "Citation counts give no clue about whether the cited material was accepted, incorporated, rejected, expanded upon, dismissed, mentioned in passing, o r given in depth analysis" (Yerkey 1993, p167).
Nevertheless citation analysis continues to be widely used in library and information studies. Its advantage is that it is an unobtrusive measure and the data is easily collected. One area where ci tation analysis studies are used is to assess the research performance of academic programs or research institutes and individual researchers or faculty members. A recent study in the UK has suggest ed that citation counts for faculty members is a good method for ranking library schools or departments (Oppenheim 1995). All 622 citations received in Social Scisearch for articles published since 1988 by the 217 academics teaching in UK library and information science schools were gathered. They were then used to rank schools and compared to the ratings awarded to the schools by the Universi ties Funding Council Research Assessment Exercise in 1992. There was a good correlation between the total number of citations received by a school, or the average number of citations received by eac h academic in the department, and the rating given the school by the Assessment Exercise.
A ten year citation analysis for the period 1983 to 1992 of the School of Library and Information Science at Indiana University was carried out (Cronin and Overfelt 1994). Cited author searches wer e run for all full time faculty members at the School for 1983 to 1994, using faculty bibliographies of published material, on all three citation indexes. Self citations were not included. A result of number of citations to each faculty member's work over the ten year period was obtained, and different data gathering approaches used. A small number of faculty members accounted for the bulk of the citations. As 92% of citations were found in the Social Sciences Citation Index, the study confirmed earlier studies that library and information science is based in the social sciences. The f aculty were infrequently cited outside their own field. Cronin and Overfelt concluded that citations could be used with other measures of performance to strengthen peer review. They thought that th e low findings from the citation analysis did not equate with a high ranking of the School for research, publication and leadership in perception rankings of American library and information science schools.
Citation studies throw some light on the use of international library literature. A study by Herman (1991) compared UK and USA behaviour in citing foreign literature. Citation data was collected fo r articles published in 29 English language library and information science journals for a year's publication cycle starting in 1987, yielding 478 usable articles containing citations for English lan guage periodical literature published 1980 87. The 2,008 usable citations were then categorized according to the nationalities of the citing author/journal and the cited journal to identify national citation patterns for the UK and USA.
UK and US journals received more than 91% of the master list citations. When the UK and USA citation data were analyzed, it was found that citations for UK articles were 50% to UK published journals , and 42% to USA journals and 8% to other journals. For American articles, citations were 91% to USA journals, 5% to UK journals and 4% to other journals. Journal self citation accounted for 19% of citations in UK articles and 17% in USA articles. (Herman 1991 p43 Table IV). There are problems with the selection of journals for the source list (5 of 9 had international orientation), which ga ve a low concentration of UK authors published in the UK source journals.
Herman (1991,p45) concluded that: "UK journals demonstrated much greater receptivity to foreign literature than did the US journals", and that USA journal literature could be called "parochial." Th e sheer quantity of American library and information science literature did not appear to be the cause for it being cited more. Herman wondered if UK literature was perceived to be of inferior qual ity.
A study of 211 articles published in New Zealand Libraries 1980 1990 showed that New Zealand material was cited 37% of the time, and material published in New Zealand, USA, UK or Australia comprised 85% of the citations. There was no foreign language material cited. (Attwood 1991). Half the citations were to material published in the last five years. The most often cited journal was New Zeal and Libraries itself with 84 citations (out of a total of 952) with Library Journal cited 26 times. There were 126 journals which were rarely cited, once or less a year. Attwood thought the authors weren't scholarly in orientation.
Content analysis consists of extracting and evaluating in a systematic and generally quantitation manner the occurrences of the manifest and latent content of a body of textual or audiovisual material, in order to uncover its k ey symbols and themes and to compare them to one another. It can use different types of analysis among others, classificatory, lexical, and propositional.
The Section on Library Theory and Research has been showcasing a series of studies about journal articles using content analysis at its open meetings, and by some project support. The aim is to stud y research topics and research methods by use of content analysis of a country's library and information science literature. The methodology used is that employed by two Finnish researchers, Kalervo Jarvelin and Pertti Vakkari for a content analysis of 833 international research articles in 1985 in library and information science. (Jarvelin and Vakkari 1990, pp395 421). Jarvelin and Vakkari f ound that 54% of the articles were research articles. The topics of library and information service activities, and information storage and retrieval were the focus of 30% each of research articles, with only 6% on information seeking and 7% on scientific communication. The survey research method was used in 22% of studies; it was the most popular.
We cannot assume that most library and information science literature will be published within the field. Content analysis of document surrogates, that is 855 library and information science related references retrieved from 32 databases, was carried out by A. Neil Yerkey (1993). He wanted to find out if the library and information science literature was written by people working in the librar y and information science field or outside it, and whether they published inside or outside the LIS literature. The findings showed that practising librarians published 33% outside the field, and th at 46% of writings by information professionals were guides to information sources, collection descriptions and bibliographies, all aimed at enhancing information access. Of the documents published outside the LIS literature 61% were descriptions of services and sources. Of the 855 documents, 497 were published in library literature sources and 358 outside. Searches of only Library Literature or Library and Information Science Abstracts will miss important items.
Both bibliometric measures and content analysis can be profitably used together for studies of scholarly and professional communication.
Use studies of library journals We have a few studies available on the use of library journals. Edward Dudley (1992) noted three market research studies that had been carried out over recent years i n the UK. It is difficult to contact all readers by means of user questionnaire surveys, when many subscriptions may go to libraries. The Library Association Record which goes to members of the Lib rary Association, used a telephone interview of a structured sample of members to find out more about readership of the journal. Most journals seem to run regular feedback surveys.
The results of reader studies we do have show that the journal articles are not always being read by their targetted audience. L. O. Aina reported at the Round Table of Editors of Library Journals o pen meeting at New Delhi in 1992 the results of a survey carried out by Nwafor (1987) of 39 Nigerian librarians. No Nigerian library journal, nor any African library journal, was included in the top nine library journals reported as regularly used. Nigerians were reading overseas published journals. An American study by Renee Tjoumas (1991) asked "What are the professional periodical titles c urrently being utilized by [public library] directors in seeking information which is applicable to their work situation? The topic of reading habits of public librarians had not been studied in the previous thirty years. A total of 172 public library managers of larger libraries was identified and a list of 56 journal titles compiled. The managers were asked to scale the titles for importanc e, with a 74% response rate. Nine titles received high scores, including 4 general ones, Library Journal, American Libraries, Wilson Library Bulletin and Unabashed Librarian; 2 targeted at public li brarians, Public Libraries and Public Library Quarterly; and 3 that were research oriented, Library Trends, Library Quarterly and RQ. Other titles considered important included specialist ones for c hildren and young adult services and ones on technical services and application of technology. Tjoumas noted "that a variety of periodicals considered core materials or essential for communicating r esearch, are either not read or not known by public library managers" (Tjoumas, 1991, p11).
There have also been studies to determine the ranking of professional journals for prestige. One carried out by Virgil Blake (1991) surveyed all 67 (52% usable responses) library and information sci ence faculty with a specialization in school library media centres and also district level school library media coordinators in the 120 (68% usable responses) largest school districts in the USA. Th e methodology was similar to the Tjoumas study, with a list of 55 journals circulated and respondents asked to rank them on a scale of importance, in the case of faculty members, in terms of publishi ng for tenure and/or promotion. The school library media coordinators were asked to assess each journal title in terms of its value for accomplishing professional duties.
The results give only one journal, School Library Media Quarterly, that received a high ranking by faculty and practitioners. Practitioners ranked American Libraries, Library Journal and Wilson Libr ary Bulletin, general interest publications; and School Library Journal, a general specialist publication highly; and only one scholarly journal, School Library Media Quarterly. The two groups dif fered significantly in their ratings of 81.8% of the journals. Blake concluded that the journals where the faculty were likely to place articles were not those that practitioners were likely to read . If research findings "are not communicated to practicing library/information science professionals as a help to improving library/information services, research becomes a self serving and sterile exercise" (Blake 1991, p146).
As journal articles seem to be the most important method of disseminating information about research, S. Nazimdli (1985) looked at the use and attitude of practitioners in the United Kingdom and the USA to the various channels for dissemination of library and information science research. A questionnaire was sent out to chief librarians in public, academic and special libraries, the response ra te being 187 responses, 51.4%, for the UK and 210, 42%, for the USA. Journal articles were the major source for acquiring information about current research for both countries. Personal contacts we re also useful. The popular journals such as Library Association Record and Library Journal were scanned regularly, so researchers should present short popular reports in such journals, as well as s cholarly reports in the research oriented journals, if they want to get their research findings adopted. Edward Dudley asked who reads which library journals and to what effect, what is the effect o f reading library journals? a complex matter. We have only a little information available about this, as noted above. The studies being carried out by Paulette Bernhard and L. Lambert and reporte d at IFLA conferences on how the research results in information studies are communicated, and how such results are perceived and used by information professionals in Quebec and Canada, and one sched uled for this conference will throw some more light on this (Bernhard 1994).
We must also take into account how easy articles in the field are to read; studies in both the UK and USA have shown how difficult most library and information science journal articles are to read long words, long sentences, jargon, etc. This is especially so for research oriented articles. Let us aim for well written articles that communicate to as many readers as possible.
Ali, S.Nazim 1985 Library science research: some results of its dissemination and utilization Libri 35, 151 162.
Attwood, M. 1991 A Mirror to ourselves: a citation analysis of New Zealand Libraries, 1980 1990. Wellington, Department of Library and Information Studies, Victoria University of Wellington.
Bernhard, P. 1993 A la recherche des methodes de recherche utilisees en sciences de l'information (In search of research methods used in information science) Canadian Journal of Information and Libr ary Science 18, 3, 1 35.
Bernhard, P. 1994 Dissemination of results of research in information studies in three journals published in Quebec. Paper given at the open meeting of the Section on Library Theory and Research, I FLA Conference, Havana, Cuba, August 1994.
Blake, V.L.P. 1991 In the eyes of the beholder: perceptions of professional journals by library/information science educators and district school library media center coordinators Collection Manage ment 14 (3/4) 101 148.
Borgman, C.L. ed. 1990 Scholarly communication and bibliometrics Newbury Park, Sage.
Broadus, R.M. 1987 Toward a definition of "bibliometrics." Scientometrics 12 (5 6) 373 379.
Cronin, Blaise and Overfelt, Kara 1994 Citation based auditing of academic performance Journal of the American Society for Information Science 45, 2, 61 72.
Dimitroff, Alexandra and Arlitsch, Kenning 1995 Self citations in the library and information science literature Journal of Documentation 51, 44 56.
Dudley, E. 1992 Who reads which library journals and to what effect? Booklet 7, Division Education and Research, 58th IFLA General Conference, New Delhi, India, Aug.30 Sept.5, 20 21.
Herman, I.L. 1991 Receptivity to foreign literature: a comparison of UK and US citing behavior in librarianship and information science Library and Information Science Research, 13 37 47.
Jarvelin, K. and Vakkari, P. 1990 Content analysis of research articles in library and information science Library and Information Science Research 12, 395 421.
MacRoberts, M.H. and MacRoberts, B.R. 1989 Problems of citation analysis: a critical review Journal of the American Society for Information Science 40, 5, 342 349.
Nwafor, B.U. 1987 Nigerian librarians and publishing Nigerian Libraries 23 (1&2) 50 66.
Oppenheim, Charles 1995 The correlation between citation counts and the 1992 research assessment exercise ratings for British library and information science university departments Journal of Docume ntation 51, 18 27.
Schrader, Alvin M. 1984 A bibliometric study of the JEL, 1960 1984 Journal of Education for Library and Information Science 25, 279 300.
Tjoumas, R. 1991 Professional journal utilization by public library directors Serials Librarian 20 (2/3) 1 16.
Yerkey, A. Neil 1993 Publishing in library and information science: audience, subjects, affiliation, source, and format Library and Information Science Research 15 (2) 165 184.