Time to take another Bath? A preliminary report on the feasibility of repeating the INFROSS study
Social Science Subject Consultant
Oxford University Library Services
The post-INFROSS decline of information needs and uses studies
One of the initial tasks of this feasibility study was to report on studies of information needs and uses in the social sciences that had been conducted since INFROSS. As Hobohm (1999) discovered, though, aside from one or two specialist studies of applied social science fields, and a number of studies of social science information within the political context of developing nations, there has in fact been very little research conducted on this theme since the early 1970s. Certainly, nothing of the scale or ambition of INFROSS has been attempted for the social sciences since its completion and, although the study itself was cited extensively at the time, neither its findings nor the recommendations of its authors have had any discernible long-term impact upon the development of information services or systems within these disciplines.
The swift decline of interest in large-scale, discipline-based studies of information need and use has not been restricted to the social sciences. Similar endeavours in the natural sciences (Menzel, 1966; Garvey, 1979) also petered towards the end of the 1970s. Undoubtedly, one of the major factors in this decline was the growth of computer technologies during the decade, which in turn encouraged information system designers into believing that many of the existing information problems addressed by these studies would simply disappear. Another reason was that the practical premise of the studies – to discover information about need and use in disciplines that could be used in the design of information systems – was essentially contradicted by the finding that scientists (and particularly social scientists) do not favour formal information systems when finding information but instead make use of informal communication channels (Garvey, 1979; Brittain, 1982). This finding generated a paradox and potential conflict of interest for the information needs and uses researcher who was in principle reliant upon these methods (such as personal contacts, so called ‘citation chaining’, or simply browsing collections) being woefully inadequate for the purposes of scientific research in order to justify the purpose of the study.
It was this combination of contradictions and perceived failings that led Martyn in his review of information needs and uses studies to declare (just four years after the completion of INFROSS) that large-scale studies of information need and use were now consigned to the "age of the dinosaurs" (Martyn, 1974). It would be some 10 years before any reassessment of this genre of study was again felt necessary, partly as a consequence of the realisation that information retrieval problems in the sciences had not simply disappeared with technological advancement as expected. Subsequent calls for a reassessment of the field and for new studies, though, did not thereby prefigure a return to the disciplinary model. On the contrary, new approaches have taken their lead in large part from cognitive information retrieval research and in some cases have explicitly argued against any consideration of the field or discipline to which users belong (Dervin, 1986).
A return to the discipline-based approach?
So why now, some 20 years after widespread disenchantment with disciplinary studies in the field of information research, should a new and updated version of the INFROSS study be proposed? The simplest answer is that we (i.e., librarians, information scientists, government organisations, publishers etc.) still do not posses adequate levels of knowledge about social science information needs in order to provide effective services, effective collection development, information products that are actually needed by social scientists, or to fund suitable projects aimed at improving social science communication systems.
When Maurice Line reflected on the possibility of repeating and improving upon the original INFROSS studies he was keenly aware that one shortcoming which needed addressing was its failure to properly understand information needs in the social sciences. "Uses", he wrote, "are not difficult to ascertain, but needs are another matter" (Line, 1999, p. 134). Similarly, Brittain (1982) has pointed out that "the majority of user studies have been library and document oriented, rather than consumer oriented, or work oriented" (p. 142). He argues that other studies are now looking more closely at the working environment and professional aims of those in need of information, and suggests this as a new focus for calculating "the information needed to successfully and efficiently carry out work tasks, and solve scientific, social, economic, political, and practical problems" (p. 142).
A new study based upon INFROSS would certainly need to take these recommendations on board. Not only would it have to address the many technological changes to scholarly communication that have taken place since the 1970s, it would also have to advance upon the groundwork already laid by INFROSS by investigating information needs as well as uses. Line hints that this will require more than just a quantitative approach, suggesting focus groups and other "softer methodologies". It will also demand a renewed interest in the professional and task-oriented contexts that generate information need and use in the social sciences, and this ultimately means understanding research disciplines and practitioner environments as primary influences upon the information needs of individuals.
It is not hard to perceive, from the numerous reviews of the field of information needs and uses research, a general dissatisfaction on the part of commentators with the level of methodological sophistication that tends to be employed in user studies. As reviews of the research and professional literature have demonstrated (Julien, 2000), there is still much debate about the best way to proceed in understanding user issues but very little in the way of comprehensive studies. When studies are conducted, questionnaires are still by far the most popular method of obtaining information about user behaviour and offer an efficient method for scoping large numbers of users. Nevertheless, deployment of this method alone does not lead to a sophisticated understanding of information use and neither does it uncover any useable knowledge about information need.
A variety of methods, such as observation and focus group research, have been proposed as a replacement for, or supplement to, the questionnaire approach in order to achieve deeper levels of understanding. An overriding factor, though, is the intellectual context in which these deeper level studies are then interpreted and presented to their audience. As Dervin has stated in her review of the field in the Annual review of information science and technology "a concern for conceptual impoverishment in the information needs and uses literature has run through past ARIST chapters like a thin but obvious thread of many colours" (1986, p. 3). Even if research methods are agreed and used widely in a variety of studies, continued conceptual impoverishment will always remain problematic in establishing and relating findings back to a steady state of knowledge.
Information science is principally a borrowing discipline. It takes from other disciplines the methods and concepts it feels it needs and adapts these to its own ends in analysing and solving information research problems (McKechnie, 2002). Difficulties arise when the variety of these borrowings means that the concepts and findings resulting from different studies are not related properly to one another. Computer science, cognitive science, sociology, ethnography, and grounded research represent just some of the disciplinary borrowings used in recent user and information retrieval research. When Line and his colleagues conducted the studies for INFROSS their methodological borrowings were limited to basic social science research methods. To develop our understanding of social science research information needs and uses further will require a different, more advanced social research perspective. Nevertheless, it is vitally important that this is seen both as an update and extension of INFROSS (rather than as yet another new beginning for information needs and uses research) and relates comprehensively to other information science user research approaches, such as information retrieval research.
The focus upon information need as the basis for both actual and potential use takes us into the realm of the working environment of the social scientist. As part of the context of academic research in the social sciences, this should align the methodology of the study much more closely with the discipline-based approach of a sociology of knowledge than with more popular cognitive approaches to the study of information retrieval. As part of the context of applied social science, this sociological approach will necessarily extend further into the environment and structure of working life in the applied social sciences.
With regard to academic social science, it is already known (from social studies of science) that scientists act and think within the peer group of their particular field of study (Price, 1963; Crane, 1979). Recent research has also revealed the way in which disciplines use peer review and formal communication via journals to establish hierarchies and set research agendas and funding priorities (Whitely, 2000). The necessity of taking part in these social structures (by producing and attempting to have validated one’s own research results) in turn creates information needs. How precisely this works in the social sciences, generally regarded as less structured than the natural sciences, and what information needs the various stages of research and participation in a research field generates, should be a primary object of investigation. Both conceptually and methodologically this will require a sociological study of social science knowledge with a deliberate focus upon the transmission of that knowledge.
Learning from the lessons of the past
A return to INFROSS should learn from and build upon the lessons of the past. As Line and his colleagues stated in their introduction, INFROSS was but a preliminary investigation. The intentions were modest but this study was nevertheless unfortunate in occurring at time when a post-war push for widespread reform of scholarly communication systems had reached feverish levels because of the availability of new technologies to implement these reforms. We now know that blind faith in the transformative power of technology was misplaced, especially without the willing participation of scientists, and that the expectations it placed upon information needs and uses studies to provide the ground for reforms was largely excessive as a consequence.
Recently, a similar scenario has arisen in which dissatisfaction with the costs and reactionary nature of formal scholarly communication systems has been fuelled by the availability of Web publishing. Expectations, at least in certain quarters, are once again high for widespread scholarly communication reform. A return to the ground covered by INFROSS should avoid the same mistake of raising expectations about useable findings for sweeping scholarly communication reform. Such a study is unlikely to contradict the original finding that social scientists do not care sufficiently about the mechanics of publishing and information systems design to completely revolutionise their existing practices.
The expectations raised by a return to INFROSS should be more pragmatic and relevant to the daily decision making of information service providers and individual product development teams than before. Questions asked by information professionals (such as: What coverage of electronic journals should we provide? What is the best method of providing access to statistical data? Do social scientists in all fields read monographs?) are a prelude to decision making about purchases, services, and product development for the social science community, and should be answered on the basis of as much relevant information about social scientists’ working habits and needs as can be discovered. Currently, much of the information required to make informed decisions is lacking and it should be a primary aim of any new study to fulfil this need.
Another lesson to be learnt from both INFROSS and the research literature of social studies of science is that the social sciences should not be regarded as a unified entity. The original study revealed numerous differences between disciplines and Line (1999) has since reflected upon the differences between the social sciences and natural sciences in their looser use of terminology and less stringent penalties for duplicated research. In addition to this, Whitley (2000) has demonstrated how scientific disciplines across the board posses very different social structures and levels of hierarchical development from one another, and more recently Kling and McKim (2000) have argued for greater consideration of field differences in predicting the impact of technological development upon the communication structures of those fields. Greater depth of study into individual fields, accompanied by more comparative study between different disciplines, will also be required. A key element of this comparative work will be an examination of the way in which language and national perspectives shape individuals’ engagement with these academic and professional structures.
Finally, there are a number of more basic pitfalls to be avoided in the organisation of a new study of social science information needs and uses. Many of these are listed by Brittain (1982) and Wilson (1980) and include methodological issues such as avoiding the unnecessary use of a library perspective when questioning social scientists about their information needs, to practical issues such as proper training for interviewers and sufficient funding for the project as a whole. The methodology of study itself should be guided by the requirements of academic research. Essentially, it will be a piece of social research, conducted and organised by researchers of similar background and interests to those who will be studied. Nevertheless, the pragmatic needs of the potential audience for this research should remain in the mind of the researchers at all times. Supportive agents in the communication chain (librarians, publishers, funding bodies etc.) must also feel ownership and fully appreciate the purpose of the study and the knowledge it will provide.
Summary of recommendations
- It is recommended that a study, of similar intention and scope as the original INFROSS study, be conducted to examine the information needs and uses of social scientists.
- The purpose of this new study will be to address the problem of our still insufficient knowledge about these aspects of social science work, to fill the knowledge gaps that remain in the daily decision making of service and information providers, and to extend and follow up the research agenda set by the original study.
- The study should take as its conceptual and methodological basis the social study of academic and professional life, as demonstrated by the field of social studies of science and some recent contextual studies of information use in the field of information science.
- Comparative work on field differences, applied professions, and the influence of national and linguistic contexts should form a significant part of the study.
- The study should avoid the pitfalls of being seen as the basis for a revolutionary transformation of existing systems, portraying itself instead as the attempt to create a comprehensive knowledge base for decision making on the part of information providers and intermediaries. As part of this process, results will need to be properly related to and understood as an extension of the original study.
Brittain, J.M. (1982). ‘The pitfalls of user research, and some neglected areas’. In Social science information studies. 2 (3), pp. 139-148.
Dervin, B. and Nilan, M. (1986). ‘Information needs and uses’. In Cuadra, C.A. (ed.). Annual review of information science and technology. Vol. 21. Washington, DC: American Society for Information Science. pp. 3-33.
Garvey, W.D. (1979). Communication: the essence of science. Oxford: Pergamon Press.
Hobohm, H.C. (1999). ‘Social science information and documentation – time for a state of the art?’. In INSPEL, 1999 (3), pp. 123-130.
Julien, H. and Duggan, L.J. (2000). ‘A longitudinal analysis of the information needs and uses literature’. In Library & information science research, 22 (3), pp. 291-309.
Kling, R. and McKim, G. (2000). ‘Not just a matter of time: field differences and the shaping of electronic media in supporting scientific communication’. Journal of the American Society for Information Science, 51(14), 1306-1320.
Line, M.B. (1999). ‘Social science information – the poor relation’, in INSPEL 33 (3), pp. 131-136.
Martyn, J. (1974). ‘Information needs and uses’. In Cuadra, C.A. (ed.). Annual review of information science and technology. Vol. 9. Washington, DC: American Society for Information Science. pp. 1-23.
McKechnie, L.E.F. and Pettigrew, K.E. (2002). ‘Surveying the use of theory in library and information science research: a disciplinary perspective’. In Library trends, 50 (3) pp. 406-17.
Menzel, H. (1966). ‘Scientific communication: five themes from social science research’, American psychologist, 21, pp. 999-1004.
Price, Derek J. De Solla (1963) Little science, big science. New York: Columbia University Press.
Whitley, R. (2000). The Intellectual and Social Organisation of the Sciences, 2nd. ed. Oxford: Oxford University Press.
Wilson., T.D. (1980) ‘On information science and the social sciences’. In Social science information studies, (1), pp. 5-12.