The Social Context of Information Needs:
Developing a Research Agenda

How does social context influence information needs and behavior? To what extent, and in what manner, must digital library features be tailored to fit different social contexts?

In developing a research agenda to address these questions, we might begin with the assumption that not only each information need, but each document and library is socially constructed. Each arises from and bears the (implicit and explicit) impress of a complex social context. On the other hand, documents, libraries, and people all have multiple identities--the social markings are to some extent negotiable as they are brought together in a particular social context in each information-seeking and use situation.

In exploring the social context of information needs, I would start with three overarching research questions. First, what aspects of social context are important to understanding digital library use? Second, how do we go about collecting and representing data to identify or study those aspects? And third, how do we apply what we have learned to improve our understanding and our digital library systems? These questions are of course interrelated; answers formulated for each should in turn inform and reformulate the others.

Aspects of social context that seem fruitful for study include:

1. The nature of social interactions--both in and outside the library--that occur as people become aware of information needs and seek to resolve them (Reich; Twidale).

2. Work and power transformations accompanying the introduction of information systems, for both library workers and the end users of systems (Markussen; Suchman).

3. Affective responses to information systems: what kinds of fears, discomforts, and pleasures are evident when people confront systems? why do they arise and what are their implications for digital library use?

4. The influence of social context (e.g., social pressures, norms, practices) on information needs and behaviors, where social context may incorporate several layers (Pejtersen; Van House):
   • General cultural or national context
   • Professional context
   • Organizational context (Elliot; Kling) --Task context
   • Situational context

5. The prevailing social context in which documents are created, shared, brought together in a collection, and used, including the nature of prevailing institutions, norms, and genres associated with document creation and dissemination (Levy) and the manner in which people move between public and private document spaces (Bishop).

6. The nature of cultural assumptions embedded in system design: What social values and beliefs (e.g., people should be diligent and rational in using the system; organization is "naturally" hierarchical), as well as expectations (e.g., people using the system speak English) are inherent in the design of information systems? And how are these assumptions encoded in the social characteristics of the system, i.e., system mechanisms for interacting with users?

7. The socialization process related to information needs and behaviors: How do people learn about the norms, constraints, and practices governing information needs, behaviors, and systems in their discipline, their organization, their daily life?

8. The manner in which new technologies are realized in unique ways by individuals, who are seen as active constructors, not passive recipients, of the system (Bruce).

How can libraries deal with the loss of face-to-face social contact in those cases where all interactions occur online?

How can libraries deal with the explosion in scenarios of use, in those cases where the diversity of its audience has in fact expanded greatly?

These problems suggest several specific avenues of research. Aside from the question of the how system content and features may need to be tailored to fit different social contexts, we can also try to determine what the library has to know about the person using the library, as well as what the user has to know about the library and its collection, in order to facilitate a smooth fit between the user and the system. Research on the social context of information needs and behaviors could also be applied to improving user support and training, which, like library content and features, might be tailored to diverse social contexts.

The second overarching question posed above is: how do we go about collecting and representing data to identify and study the social context of information needs and behaviors?

Research approaches which I think are especially needed at this juncture include:

1. Ethnographic observations of the social context in which information needs arise.
2. Observations and interviews to capture the actual questions and document clues that people bring to information systems
3. Observations of the kinds of formal and informal social interactions that occur as people use digital libraries
4. System instrumentation to capture not only individual information behaviors, but the relationships between people and documents, people and systems, people and people (Borgman; Buttenfield; Gay and Mead)?

Finally, "cross-cultural" system studies represent a class of research design that can be used to both collect data and consider how knowledge gained can be applied to tailoring system features to different social contexts. For example:

1. The information needs and behaviors of students from different cultures and backgrounds who are using a particular library system could be compared.

2. The manner in which a digital library whose use naturally spans several application domains is, in fact, used by representatives of those different domains, could be studied (e.g., How is a GIS used by students, environmental planners, real estate agents? How is a medical information system used by patients, doctors, nurses, the general public (Lambert)? How is an online collection of museum artefacts used by researchers, artists, and teachers (Sandore)?

3. The use of different versions of system features (such as retrieval mechanisms, icons, and document representations) associated with a single digital library could be compared within and among different social groups to investigate the generalizability of features.

   Implicit in any system is its intended user, an ideal fiction of perfect fitness. One step in better adapting systems to their intended users is to make the details of that embedded fiction more explicit. We already do this, if indirectly, when we conduct user studies: we study the user to discover where that implicit fiction and the concrete subject fail to merge. Yet it must be acknowledged that each study of "the user," if conducted with sufficient granularity, yields or reconstructs a new identity, particularly as we move across knowledge domains, languages, socioeconomic statuses, and national cultures. We may begin to ask how many implicit identities are necessary for optimal function. Certainly a limit is reached, at some point, beyond which the user may have no choice but to adapt to a recalcitrant technology which in some measure fails to fully acknowledge the claims of human singularity. In this, however, our technology hardly differs from our institutions.

4. Many issues related to the social context of digital library use were addressed by participants in the 37th Allerton Institute on "How We Do User-Centered Design and Analysis of Digital Libraries: A Methodological Forum." References refer to the authors of relevant discussion documents submitted for the Institute, which may be accessed at allerton.