WPbanner5.jpg (10176 bytes)
No. WP- 02-11
From Users to Social Actors:
Reconceptualizing Socially Rich Interaction Through Information and Communication Technology

Roberta Lamb
University of Hawaii, Manoa
College of Business Administration
Information Technology Management
2404 Maile Way, E601-C, Honolulu, HI  96822
Email: lamb@cba.hawaii.edu
Phone: 1(808)956-7368 FAX: 1(808)956-9889

Rob Kling
Indiana University
SLIS, Center for Social Informatics
10th & Jordan, Room 005C, Bloomington, IN 47405
Email: kling@indiana.edu
Phone: 1(812)855-9763 FAX: 1(812)855-6166

Draft 6.4
Copyright, 2002.  Roberta Lamb

A concept of "the user" is fundamental to much of the research and practice of information systems design, development and evaluation. User-centered information studies have relied on individualistic cognitive models to carefully examine the criteria that influence people’s selections of information and communication technologies (ICTs).  In many ways, these studies have improved our understanding of how a good information resource fits the people who use it.  However, research approaches based on an individualistic “user” concept are limited.

In this paper, we examine the theoretical constructs that shape this “user” concept and contrast these with alternative views that help to reconceptualize "the user" as a social actor. Despite pervasive ICT use, social actors are not primarily “users” of ICTs.  Moreover, such socially thin and somewhat pejorative conceptualizations limit our understanding of information selection, manipulation, communication and exchange within complex social contexts. Using analyses from a recent study of online information service use, we develop an institutionalist concept of a social actor whose everyday interactions are infused with ICT use. We then encourage a shift from "the user" concept to a concept of the social actor in IS research. We suggest that such a shift will sharpen perceptions of how organizational contexts shape ICT-related practices, and at the same time will help researchers more accurately portray the complex and multiple roles that people fulfill while adopting, adapting and using information systems.

1.0 Introduction

By relying on individualistic models in information and communication technology (ICT) studies, researchers have constructed a concept of  “the user” as an atomic individual with well-articulated preferences and the ability to exercise discretion in ICT choice and use, within certain cognitive limits. “User studies” informed by this concept have taken researchers a long way toward understanding what makes a computer interface easy to use, and ergonomically comfortable to use continuously (Norman, 1986; Schneiderman, 1987.)  By concentrating experimental studies on this individual “user,” however, cognitively oriented IS researchers may inadvertently limit the kinds of “user” phenomena they can study.  Often “the users” they study are situated at the interface to the computer system, where research de-emphasizes complex contextual elements and focuses intently on the exchange of information between the system and the individual, and on choices the individual makes about how and when to use ICTs. Within the complex social settings that commonly constitute organizations, however, individuals don’t always have the opportunity to choose the systems they would prefer to use.  For organizational individuals, “the user” concept is a mischaracterization that engenders faulty expectations that ICTs are predominantly “single-user,” “empowering,” and “widely applicable.”  Moreover, when aggregated to predict organization-wide activity, the one-size-fits-all concept of  “the user” leads to frequent overestimates of ICT use (Libmann, 1990; Dutton et al., 1993; Wilson, 1995; Baldwin and Rice, 1997; AUTHORS6, forthcoming.)

The theories that shape this understanding and influence the design and use of ICTs rely primarily on cognitive social psychology and cybernetic models that are contextually under-developed, leaving nearly all of the organizational and environmental context outside the model. "The user" concept these theories construct is not wrong, but by focusing on individualism, it provides relatively little detail about the contexts that shape ICT use, and so diminishes the importance of organizational structures and complex social environments.  This focus tends to amplify technology specifics and to attenuate the social context, particularly people's relationships with those who have requested information or who they are trying to persuade with information gathered and packaged through the use of ICTs.  When we criticize "the user" concept in this paper, we're referring to this inflation of individualistic aspects, and to the diminution of contextual aspects in cognitively focused IS research.

Within several research disciplines related to IS studies, there is a growing realization that ICTs, like online information services, have achieved only limited success as “useful” information systems, in part, because they are based on models that reflect this “user” concept (Grudin, 1990; Beath and Orlikowski, 1994; Westrup, 1997; Salzman; 1998.)  These thoughtful observations and impassioned critiques have motivated more social awareness and more human-centered design practices, but they do not provide a theoretical basis for an alternative to the “user” concept that can be coherently integrated into ICT design principles. We believe that a theoretically based reconceptualization of the “user” as a social actor is required to move IS research beyond this impasse.  Therefore, we have begun to model a view of the organizational individual as a social actor by building on the concepts of new institutionalists (Scott, 1987; DiMaggio and Powell, 1991) and social constructionists (Goffman, 1974; Giddens, 1984; Touraine, 2000), as well as the rich descriptions of empirical studies that show how organizational contexts constrain and enable individuals using ICTs. This approach gains strength from Scott's (1995) synthesis of institutionalist theory concepts, which outlines a framework for bringing cognitive insights into a contextually well-structured conceptualization of ICT use.

In this paper, we will present our social actor view through a series of critiques and constructions that are based on a multi-disciplinary literature review, and are grounded in our own empirical findings.  In the following sections, we will examine the theoretical constructs that shape the ICT “user” concept, and then present some empirical challenges and theoretical analyses that help to reconceptualize the “user” as a social actor.  We will then present findings from a study of online information service use that characterize four ICT-related dimensions of a social actor.  Building on these findings and the theoretical insights they support, we will present a model of the social actor that is fundamentally integrated with ICT use.  We will conclude our discussion with an invitation to the IS community to try this reconceptualization – to move beyond “user” studies by examining the ICT use of social actors.

2.0 The “User” Concept

The “user” concept is grounded in the cybernetic models of Herbert Simon, particularly his ideas of bounded rationality and learning through information feedback and adaptation (Simon, 1955.) These models describe an atomic individual with well-articulated preferences and the ability to exercise discretion in ICT choice and use, within certain cognitive limits.  They also describe how “information” from objects, the environment, and interactions with other atomic individuals is cognitively processed as feedback to fine-tune the preferences that influence discretion. Within these models, however, “information” is highly decontextualized.

Over the past two decades, cybernetic models and concepts from cognitive social psychology (Fiske and Taylor, 1991) have formed the basis for widely-held understandings of individual mental models and communication behaviors involving the use of ICTs. “User” studies typically seek to inform ICT design by examining how task models, ergonomic factors and cognitive psychodynamics define the limitations of human interaction with the computer system (Norman, 1986; Shneiderman, 1987.)  Theoretically, all humans have the same set of capabilities and limitations, albeit to differing degrees, ranging from novice “users” to expert “users.”  Studies that seek to understand “user” satisfaction with information systems (IS) and IS services, for example, frequently draw on Simon’s decision theory models of consumer choice that explain the satisficing behaviors of consumers and their coordinations through free-market interactions (cf. Malone & Crowston, 1990.)  Studies that seek to understand the ways in which ICTs may affect "user" behaviors often rely on attribution theory to construct cognitive models that explain people's perceptions and rationales for their subsequent actions (Mishra et al., 1996.)  By adopting research models that reflect the “user” concept, researchers implicitly agree to model an artificially constrained set of contextual factors in controlled experimental settings, or to leave context outside the study entirely. Their studies most often take the form of laboratory experiments or surveys that evaluate the task/technology fit of computer systems at the individual use level.  Theoretical insights drawn from “user” studies have been applied extensively in the design of ICTs.  The design of online services, for example, has relied heavily on individualistic models to explain the use (and, frequently the non-use) of online products by librarians, research intermediaries and “end-users”—i.e. people who gather information from online databases for use in their own decision-making and work-related tasks (Bellardo, 1985; Borgman, 1989; Newby et al, 1991; Nicholas et al., 1988.)

These studies have improved our understanding of how a "good information source" fits the people who use it.  Therefore, we wish to state clearly that the discussion which follows is not meant to invalidate this line of research or the results of these studies, but to call into question the “user” concept, which these studies have constructed over time (and which many researchers have assimilated into their studies of organizational IS).  In this way our criticism is similar to that made by Nye and Brower (1996) of cognitive social psychology, more generally.  However, in IS research, pervasive use of the term “user” presents a further complication.  The term has become institutionalized, so that many IS researchers, including ourselves (e.g. AUTHORS1, 1977), have at times unreflexively employed the term to talk about ICT use -- even when describing social interactions that extend beyond individualistic “user” conceptualizations.  Therefore, in addition to an alternative to the "user" concept, IS researchers need an alternative to the term.

3.0 Challenging the “User” Concept

A key criticism of the “user” concept is that it excludes context.  Field studies frequently show that ICT use projections based on “user” studies do not accurately predict use outside laboratory contexts--that “user” study findings simply don’t scale up to the organizational or industry level (Baldwin and Rice, 1997; AUTHORS5, 1997.)  Researchers acknowledge that, within the firm, individuals rarely have the opportunity to choose the ICTs they use (Karahanna and Straub, 1999.)  Instead, they select from a set of resources chosen at the organizational level. Such studies indicate that, from an instrumental perspective, the “user” concept is too narrowly defined.

Another group of studies has critically challenged the “user” concept from a human relations perspective.  Building on the early work of socio-technical design researchers, these studies examine the ways in which ICT designs that are based on the “user” concept may be inadequate, dehumanizing or disruptive to cohesive, productive working contexts (Nygaard, 1986; Ehn, 1988; Greenbaum and Kyng, 1991.)  Mumford’s socio-technical ETHICS approach was particularly influential among Scandanavian ICT designers in the 1970s, and continues to shape their research methodologies and perspectives; although the popularity of socio-technical studies (STS) approaches has waxed and waned over the decades, as economic conditions and labor relationships have evolved throughout the world (Mumford, 1995; Mumford, 2000.)

Today, these two perspectives form the basis for the most cogent criticisms of the “user” concept that one currently finds in an inter-related set of literatures derived from Human-Computer Interaction (HCI) studies, Computer Supported Cooperative Work (CSCW) studies, and the European tradition of Information Systems (IS) research.

HCI and CSCW Critiques
Many critiques of the “user” concept begin with objections to the largely artificial separation between ICT “developers” and ICT “users,” then advocate a more participatory design process.  Participatory design researchers believe that when the workers who are expected to use the ICT under development also take part in its design and implementation, the use context will be more fully reflected in the final system, and workers’ tacit knowledge about the task can be brought to bear (Greenbaum and Kyng, 1991; Nardi and Miller, 1991; Gutwin and Greenberg, 1998; Guzdial et al., 2000.) Their studies enrich conceptualizations of “users” by focusing on the cognitive complexity of their tasks and the adaptive nature of situated work.  Some have even raised objections to the rational systems view of software development, which supports the “user” concept; preferring an informatics view that regards “information systems as networks of people” in which each participant interprets the process in his or her own way (Nygaard, 1986.) Their studies critically examine the fuzzy definition of a "user," recognizing that “user” categories could be based on functional roles or interest groups; and that “users” can assume more than one role simultaneously.  Furthermore, as “users” extend their knowledge of informatics, performing some of the day-to-day system modification themselves, their roles merge with those of developers.

Among those who directly experience the everyday negotiations of information systems development and maintenance, there is a strong sense that “user” concepts and characterizations are inadequate (Grudin, 1990; Bannon, 1991; Woolgar, 1991; Westrup, 1997.) Along with other researchers, these authors have noted that “users” don’t hold the same view of themselves that IS analysts do, and they don’t like to be referred to as “users”  (Markus and Benjamin, 1996; Beath and Orlikowski, 1994; Grudin, 1990.)  In fact, “users” don’t think of themselves as primarily having anything to do with the computer at all.   They see themselves as professionals, working with others, and using computers in support of those interactions.  They are rewarded for the work they do, not for using computers.  The realtor who goes online to search for data on recent property sales is rewarded for selling real estate, not for doing online research.  The attorney who compiles a list of relevant citations from a LEXIS database can charge for time spent online, but is hired to build a case for the client.

One might have expected that as “end-users” and developers began to share the same tasks, the term “user” would have disappeared.  The very term “end-user” should indicate that the value of the “user” concept has broken down.  But, as Westrup (1997) has noted, the ICT “designer” is also continually constructed in the process of ICT design and use, in large part because system development methodologies are firmly rooted in atomic “user” concepts, and they strongly reinforce the dichotomy between “users” and “designers.”

Particularly among CSCW researchers who study ICT use within groups and among organizational collaborators, there is agreement that the term “user” paints an inappropriate and somewhat pejorative picture of the people for whom information systems are created, but they are divided on how to remedy the situation. Scholars who recognize the “end-user’s” capacity for innovative uses of ICTs have suggested that one way to tap that wellspring is to provide them with highly configurable systems (von Hippel, 1998.)  However, this approach has been criticized for adhering to the “ICT as a tool” perspective which also supports the “user” concept (Westrup, 1997.)  Most CSCW researchers have cast their lot with some kind of participatory design solution. When taken into organizations, however, the systems that these approaches produce have met with mixed reviews (Gasson, 1999; AUTHORS4, 1994.)  As developers and “users” work together on system design, power imbalances frequently prevent “users” from making a real contribution (Blomberg et al., 1994.)

IS Research Critiques
In IS literature, researchers have criticized the treatment of “users” during information systems development (Markus and Bjorn-Andersen, 1987; Beath and Orlikowski, 1994), but few have challenged the “user” concept itself (Westrup, 1997.)  Several organizational IS studies have stressed the need for a larger environmental scope when dealing with the influences of ICT use, noting that individual ICT use is influenced not only by organizational contexts, but also by interorganizational, cultural and global contexts (Czarniawska-Joerges, 1992; AUTHORS5, 1997; Walsham and Sahay, 1999.)  Some socio-technical IS studies echo the concerns of CSCW studies, indicating that a focus on interactions can better describe the ways in which people come to use ICTs to support their organizational and interorganizational activities, and emphasizing that organizational individuals often fail to use ICTs in expected ways, and frequently reshape technologies to suit their needs (Kraemer et al., 1987; AUTHORS2, 1987; AUTHORS3, 1992; Hirschheim et al, 1996; Kumar et al., 1998.)  Walsham's (2001) recent overview of many of these works has made important connections between the studies.  His analyses summarize their insights, highlight the global nature of ICT-related changes, and draw attention to the dramatic identity shifts that often accompany these changes.

However, a much larger proportion of IS research reflects the “user” concept, and we believe this approach will continue to guide IS studies until researchers have a well-constructed alternative.  To illustrate this predominance, we present a quick characterization of MIS Quarterly articles published over the past few years. In this related work (AUTHORS7, forthcoming) , we found it helpful to classify IS research perspectives, and to contrast context-rich, interaction-focused studies with individualistic, “user” studies, and with other contextually under-developed approaches.  In our classification, IS researchers who adopt an Individualistic “user” view focus attention on the person at the computer and her attitudes or preferences about information gathering and ICT use (see Table 1.)  A Socially Thin Interaction (STI) view is more expansive than the Individualistic view, often referring to people as “users”, but sometimes mentioning their interactions with others when describing ICT use.  Generally, studies that reflect this view retain an inside-the-firm focus and do not develop a detailed perspective of interfirm interactions, nor do they examine how these drive ICT use.  A Socially Rich Interaction (SRI) view focuses on the ways in which people use ICTs and ICT information products when interacting with others, often with an awareness of interorganizational contexts where roles and power relationships differ as do perceptions about the appropriate use of technology.  Studies that reflect this view do not embrace the “user” concept, although they may retain the term “user,” and often advocate a participatory approach to ICT research and design.

Table 1: Conceptual Focus in IS Research Studies (MISQ Article Classification)
Technology Focus
Computers, Networks 
Technologies of Interaction
Study Participants
People performing formal task systems
Human activity systems
Contextual Scope
Task structures and their execution and coordination
Complex and multivalent social relationships in organizational settings that can extend outside a focal group
MISQ Articles (1996-1999) 

Of the 70 articles we classified, nineteen expressed the Individualistic “user” perspective,  thirteen adopted the Socially Rich Interaction perspective, and twenty-two articles fell somewhere in the middle. The final sixteen articles made no mention of  IS use or “users,” focusing instead on the organizational economics of IT, or issues of IS pedagogy and research methodology.  This quick count indicates that, although CSCW and European tradition IS studies have made important breakthroughs, the IS literature is still strongly influenced by the “user”
concept.  Many of the Individualistic and STI articles (e.g. See Appendix B: Choudhury and Sampler, 1997; Igbaria et al., 1997; Gefen and Straub, 1997; Webster, 1998; Karahanna et al., 1999) work from context-free or thinly contextualized models that breakdown when they are applied to complex “users” in professional settings (Hu et al., 1999.) A large portion of the remaining STI and many SRI articles reflect a management perspective that often focuses on the actions of managers and their effects on overall system success, rather than the actions of “users” themselves (e.g. See Appendix B:Nidumolu et al., 1996; Kettinger and Lee, 1997; Pinsonneault and Rivard; 1998); or make instrumental assumptions about how and when ICTs should be used, viewing other uses (particularly non-use) as aberrant (e.g. See Appendix B:Harkness et al., 1996; Francalanci and Galal, 1998.)  Such research efforts have helped managers avoid many of the pitfalls that plagued early ICT implementations, but management-focused ICT design approaches have only achieved limited success in creating “useful” information systems (cf. AUTHORS11, 1996. )  Among the remaining SRI-classified articles, those that seriously challenge the “user” concept also reflect the theoretical foundations that guide IS, HCI and CSCW critiques (e.g. See Appendix B: Markus and Benjamin, 1996; George, 1996; Kumar et al., 1998.)

4.0 Theoretical Foundations of “User”- Critical Literature

The perspectives that inform “user” critiques range from theories about worklife quality, workplace power and the influence of social institutions; to theories about the construction of interaction practices and technological artifacts. Although they vary in focus, each contributes to a more robust conceptualization of how, why, when and where people encounter and use ICTs, and to what effect.

Socio-technical critiques have focused on the ways in which ICTs may affect the quality of worklife, and shift the balance of power in the workplace.  This view is grounded in human relations theory (cf. AUTHORS10, 1980.)  STS advocates generally believe that better working conditions will result from some sharing of power and an appreciation of the tacit knowledge and adaptive capabilities that workers contribute to organizational processes.  They are keenly aware that structural constraints may prevent this, but they believe that social actors are capable of mobilizing change.  STS-related initiatives, like the early participatory design movement, seek to change the structures and technologies of production in ways that will benefit workers as well as managers. They believe that technologies can be constructed to support better quality-of-life work practices, and that these will, as a natural by-product, create more productive environments, based on the human relations concept that satisfied workers are productive workers.  Socio-technical studies guided by these theories adopt an activist viewpoint.  These portray managers, professionals and workers as creative agents, constrained by structure, but capable of initiating change.  They view ICTs and related technologies as potentially limiting, and possibly even harmful, without intervention in the processes of design and implementation by those who will use them.

More recent CSCW critiques have adopted an interpretive, constructionist perspective.  These rely on theories that take a balanced view of agency and structure, regarding each as one side of the same coin, each constituting the other in critical ways (Giddens, 1984.)  Advocates of this view see social actors as participants in the shaping of social structures through their iterative every-day practices, and as capable of initiating change through these very processes (Berger and Luckman, 1967.)  Social actors interact with variously constituted “others” to form the basis of social institutions and identities (Goffman, 1959; 1974.)  Technologies, particularly ICTs, are integral to these interactions and so shape identity and institutions.  In use, ICTs are an extension of practice and also a part of structure – having dual effects and creating unintended outcomes. CSCW studies guided by these theories adopt a cautiously optimistic view of ICT-related change.  These portray managers, professionals and workers as active participants in ICT design and use, regardless of their official role; but they are just as likely to be changed by a technology as to influence its shape or to alter institutional environments through its use.

Some equally effective IS critiques place more emphasis on the intransigence of institutions and the processes of institutionally constrained action, than on the potential for agent-directed change via ICT design and use (AUTHORS12, 1988; King et al., 1994.)  These are based on new institutionalist explanations of organizational change that encompass a more systems-oriented, historically-focused understanding of the possibilities for action within social institutions  (Scott, 1987; Friedland and Alford,1991; Tolbert, 1995.)  Institutionalization is the process by which an organization develops a distinctive character structure -- a set of norms and routines, a way of doing things.  Practices can become routinized within one particular firm, or they may become standardized throughout an industry. Advocates of this view reject the rational-actor models that shape the “user” concept.  They focus, instead, on the characteristics of groups, organizations, industries, and societies that cannot be reduced to aggregations or direct consequences of individuals’ attributes, preferences or motives (DiMaggio and Powell, 1991.)  They contend that, with respect to the adoption, development and use of technologies, the actions of organizations are shaped by the institutional environment (Meyer and Rowan, 1977.) In this view, social actors are pressured to perform legitimate actions and interactions within institutionalized arrangements. ICT-related change happens, but it is not agent-directed.  Rather agency is channeled through a complex, multi-level system of networks and organizational affiliations that constitute local and global environments.

Although they differ in the degree of importance they give to agency and structure, each perspective demands a more concerted examination of these social constructs in context when seeking explanations and predictions of ICT use outcomes, and each provides some guidance for research design.

5.0 Research Methods

In order to develop a better understanding of ICT use, and to develop an alternative to the “user” concept, we designed a study to examine online information services from the perspective of the people who were actually using or not using them.  Mindful of the main criticism of the “user” concept, and guided by the theoretical concepts that have informed its critiques, we focused our qualitative research on the organizational contexts of situated use.

Online information services.  If context matters, one way to characterize that influence would be to examine the differential use of a single ICT type.  We chose to study online information services because, although prior studies concurred that firms varied widely in their online use, this variation was not well understood (Bellardo, 1985; Nicholas et al., 1988; Borgman, 1989; Newby et al., 1991).

Site selection. Between October, 1995 and March, 1997, we examined the use of online services by 26 California firms in three industries: biotechnology/pharmaceuticals, law, and real estate. (See Table 2.) The firms were all located within two areas of California (Orange County and the San Francisco Bay Area) that each support an active legal practice, a strong real estate market and a viable biotech/pharmaceutical industry.  With data from online databases, such as Dun and Bradstreet, we analyzed the firms within each area and industry using revenue and employee statistics to form a rough impression of how large and how successful each firm might be.  We then correlated this information with online usage data provided by a cooperating online vendor  to identify sample sites that did not use their online services, as well as those that did. From these augmented selection lists, we chose sites for our cross-sectional study to include firms that were non-users as well as regular users of online resources, late as well as early adopters, large firms, small firms, and poorly financed as well as richly capitalized organizations.

Industry environments.  We had originally intended to select sites from only two industries that each reported high use of online resources: the biotech/pharmaceutical industry and the legal industry.  According to Scott's environmental framework, however, this design would restrict our study to two highly institutionalized industries (Scott, 1987.) We had expected the legal services industry to be highly institutionalized.  In some respects, The Law can be considered an institution in itself.  The biomedical field, although highly technical, is also highly institutionalized and heavily regulated--from physicians and hospitals to drug manufacturers and biotechnology groups.  By including the real estate industry--an industry that is somewhat less institutional, also non-technical, and not noted for high use of online services--we ensured that our study organizations would be sampled from a wider variety of industry environments.

Table 2: Online Services Study Site Sample
Orange County
San Francisco Bay Area
Real Estate 

Interviews.  We gathered data primarily through semi-structured, on-site interviews. As reflected above, our preliminary literature analysis indicated that at least three kinds of theorizing could help to shape a reconceptualization of the “user”: socio-technical approaches, social-constructionist approaches and institutionalist approaches.  These theories describe institutional relationships, social interactions and organizational technologies in ways that suggest new possibilities for thinking about the “user” as a social actor; and we used these concepts to guide collection of data about industry environments and to design our interview instrument.  Our questions to informants focused on their firm's use of online resources and print-based media, as well as their own personal contacts.  We discussed the information resources they have and use, and we talked with them about how and why they gather information and when they go online. (See Appendix A for a sample interview instrument.) Often, they would demonstrate their use of a particular online service.  During these interviews, people consistently mentioned their firms' interactions with outside organizations when they talked about using information, and they often linked changes in their data gathering practices with changes in key interorganizational relationships. Thus, while collecting data on information gathering practices at the individual level, we were able to direct attention toward the interfirm associations that influence the use of online services.

In law firms we interviewed librarians, paralegals, associate attorneys, and firm partners.  In biotech companies, we included information center directors, information specialists, scientists, marketing managers, and directors of strategic partnering units.  At real estate brokerages we included brokers and realtors.  We used a nested interviewing approach, relying on inside referrals to collect responses from between one and four informants at each site. Wherever possible, we also observed the use of online resources and services in the day-to-day activities of the people we interviewed. (See Table 3.)  A low number of interviews was typical in firms that did not use online resources at all.  Also, in a few small firms, the nested interviewing approach did not lead to any other inside referrals because online resources were used by only one person at the firm.  While this may seem to be a limitation, in fact it allowed us to better analyze organizational use, because our set of individual informants fully represented online service use within the firm.

Table 3: Role Distribution of Informants Interviewed
Informant Role  Number Interviewed
Librarian/Law Librarian
Library/Information Center Director
Information Specialist
Information Systems/Research Director
Paralegal/Legal Secretary
Basic Research Director/Vice President 
Marketing Analyst
Marketing Director/Vice President
Regulatory Liaison/Director
Information Systems/Software Developer/Manager

Data Analysis.  Throughout the study, we used a theoretical data sampling approach to guide our selection of study sites and to refine our inquiry instruments (Strauss and Corbin, 1990.)  This method combines concurrent qualitative analysis with ongoing data collection.  We followed that analysis with several rounds of coding and thematic analysis at four different levels.  Where we had conducted interviews with more than one informant at a firm, we compared and contrasted transcribed interview data at the individual level. Where we had conducted more than one industry firm study in an area, we analyzed interview transcriptions and developed themes at the firm level. We also supplemented our code analysis with data reduction through site summaries (Miles and Huberman, 1994.)  We then compared and contrasted the data categories that had emerged within each industry, and refined our thematic analysis by performing a cross-industry comparison of the qualitative categories that characterize these sites, these informants, their activities and their use of online information resources.  Both open coding and thematic coding methods were initially used to analyze the interview data.  However, we found that, with iterative refinement, thematic coding techniques developed data categories that were more appropriate for comparative analyses.

6.0 Findings
Our thematic analysis of the interview data provided a rich set of findings to inform an alternative view of the ICT “user”.  These findings characterize four dimensions of a social actor: affiliations, environments, interactions and identities.  In prior work, we have theorized about these findings at the industry and organizational levels (AUTHORS6, forthcoming.) We now examine the data from the view of an organizational individual, with the intention of bringing these levels of analysis together in later sections of the paper to construct a multi-dimensional view of social actors.


We found that the use of online databases by organizational individuals is best explained from a perspective that understands these ICTs as interaction technologies.  Organizational individuals, like attorneys, commercial real estate brokers or members of biotechnology companies, use online services to exchange information and interact with affiliated organizations, like clients or regulators, in ways that are considered legitimate within the industry.  A commercial real estate brokerage, for example, may compile a 3-inch thick report to support a pension fund client’s property investment by combining online information with print-based reports and proprietary data analyses.  Its pension and investment fund clients routinely demand more supporting data for real estate investments than individual investors require.  Brokers understand that the data gathering and packaging processes they follow for fund and trust investments are intended to meet the fiduciary requirements of the investor and to minimize financial risk, as this broker explains:

[Pension fund clients] need to have demographics.  They need to have lots of [comparative data.]  They need to have projected earnings.  A lot more homework is done, and rightfully so, especially if it's a pension fund or a life insurance company, or a publicly traded REIT, in that those types of owners are using other people’s money.  Either as a public shareholder, or if you own life insurance, and you pay your monthly fee to the life insurance company, that’s your money.  If you’re in a pension fund, obviously that’s your money.  It's the little man’s money that they’re investing.  So they have to be very, very careful. [Affiliations: client demands]
Real estate brokerages that service fund and trust clients reported using much more online information than other brokerages.  They had also developed sophisticated data gathering and management practices, to support the information intensive activities of brokers and researchers throughout the firm.  In the biotech industry, firms that interacted directly with regulators, like the U.S. Food and Drug Administration (FDA), also reported using much more online data than firms that dealt with such agencies only indirectly.  Much of the associated online activity involves submitting application packages for new drug approvals, but even after approval, the affiliation with the FDA continues to drive the use of online services, as this researcher describes:
There is a lot of information generated because of the regulatory agencies who need to be kept up to date that now that the product is marketed, it remains safe and effective.  [We search online] regularly.  With a new product, every quarter.  And then every half year.  And then after that every year.  And every year thereafter.  You have to do what we call periodic filings with the FDA.  And we search the literature very thoroughly on set dates and forward that on through [our Regulatory Department] who sends it on to the FDA with reports.  [Affiliations: load shifting arrangements]
Officially, the responsibility for oversight belongs to the regulator and the underwriter, but, as the examples show,  these organizations effectively shift the oversight-related data gathering workload onto biotech companies and investors (who then shift it to brokers), by the nature of their affiliations with these firms.   The nature of the affiliation with a regulator or a client is important.  When a large pension fund client decides that it wants to work with only one broker at a firm, instead of three or four -- no matter how many properties the client holds or in which geographic territories those properties are located -- that can spur changes in other relationships that are fundamental to brokers.  Providing “full service” requires making an exception to traditional territorial arrangements:
Commercial brokerage firms are very geographic and have territory.  What we have is Orange County.  That's our territory...Orange County primarily.  I say that with a footnote because a lot of our brokers, especially investment brokers, have clients that have holdings all over.  And of course they're going to use the same person, so that's a big exception. [Affiliations: full service]
Law firms have also had to provide “full service” in order to retain their biggest clients.  For some, this has meant finding better ways to share information within the firm.  For others, it has meant attracting attorneys who specialize in new practice areas, and then supporting their ICT needs. When a Bay Area law firm took in a group of lawyers who had been downsized out of their biotechnology firm jobs, the firm’s data gathering practices shifted accordingly, as this librarian reflects:
We [now] have a very healthy biotech segment of our clientele.  I mean I can tell just from the kind of work that we have to do in the library that takes us over to the medical school library a lot more than it used to. Biomedical devices is a very large part of our practice, and we have an FDA practice now. [Affiliations: changing jobs]
This new FDA practice provides more services for biotech clients who choose not to interact directly with the FDA.  The attorneys in this new specialization can fulfill the data gathering mandates of the FDA for law firm clients that do not have the infrastructure to manage this themselves or who do not have the capacity to manage another clinical trial in-house.  Realtors and brokers change their brokerage affiliation, perhaps even more frequently than other professionals; and our informants mentioned that such changes are often the source of new client affiliations and new information practices within brokerages.

Data from this study emphasize that, when seeking to understand ICT use and ICT “users”, it is critical to examine the network of relationships that call for the exchange of information and the use of ICTs.  Those relationships take shape within particular industry environments that define standard and legitimate practices of communication and exchange for its member organizations.  As we have shown, national and international financial and fiduciary standards strongly influence the information gathering and related ICT use practices of real estate brokers.  Regulatory and fiduciary demands for extensive documentation and the increasing availability of information in electronic form seem to be pushing brokers to compile ever more sophisticated information packages for key clients, as this broker suggests:

I would say [our information packages are] somewhat more sophisticated.  There’ll be just a little bit more thoroughness to a package that’s going to be presented to a client.  But there again, a lot of it just really depends on who that client is.  If it is an [pension fund] client, they are very much looking to have that big, thick package to go back to their boss and say “Gee, I did a great job.  Look at all this work that I’ve reviewed.”  And so, in that sense they’re more thorough and they’re better quality.  And they’re a little bit more timely.  The information is more timely.  I would say that’s probably the biggest thing that the data revolution has brought, is that the information is that much more timely. [Environments: global financial/fiduciary practices]
The informational demands of banking and regulatory agencies shape the use of ICTs within an organizational field, as do industry competitors and the efficiency dynamics of globalization.  Several commercial brokers reported feeling that they must have an impressive web presence to display their overall competence, mainly because their competitors are on the web.   However, ICTs like the web, can be a double-edged sword for this group. The web can help brokers and realtors advertise their properties and reach more potential clients, but at the same time it threatens their livelihood by “giving away” property information that has belonged exclusively to the broker community (see the MLS discussion, below).  One informant believes this is part of an overall trend in the commercial real estate industry toward greater efficiencies and fewer survivors:
I think that there are fewer commercial real estate brokers now than there were, I'm sure, 10 years ago.  And there will probably be fewer in 10 years going forward.  That's that efficiency thing, I guess.  I mean there are fewer bankers, there are fewer of all of us, and we're working harder.  Isn't it great? [Environments: industry or organizational field]
The need to manage bigger information packages and multiple property transactions across broker territories, combined with the need to compete more aggressively in local markets, has influenced one commercial brokerage in the study to develop a sophisticated set of proprietary software tools for its brokers. By investing in such ICTs, brokerage firms expect to expand their services in ways that will help their brokers attract and maintain large pension fund and investment trust clients.

Law firms that want to handle high-stakes class-action lawsuits must also make a concerted investment in informational infrastructure to coordinate expert testimony, corporate regulatory filings, and plaintiff profiles.  The data gathered for a multi-plaintiff tort case, for example, produces a mountain of paperwork that could rival a bitoech company’s NDA filing in volume (see the NDA discussion, below).  All this data must be logged and classified.  Data entry clerks must be hired and provided with adequately resourced facilities to do their work. A records manager must also coordinate with the trial attorneys to make sure they know about the deposition data and expert reports and have an adequate knowledge of case-related issues, as this IS director explains:

You can see with large litigation cases, you really, in order to manage the documents, you need to have some systems in place…One of the things that we have for these large cases, is we have another office in [central California] which is lower rent.  And we handle the complaints there.  So we have about 50 paralegals that are out in [central California].  And they just handle all of those files, and all of the complaints, and track them and everything else… some of them then go to trial, and we have to get our trial team together.  And so you have to manage a lot of people in these trial teams.  And you have to have these attorneys educated, if its a medical issue, on the medical issues, so that they can go to court and handle that type of situation. [Environments: organizational ICT investment]
Some organizations have made ICT investments at the industry level, as well as the firm level. In our study, we found that the level of infrastructural richness within an industry greatly affected the use of ICTs by industry organizations.  Generally speaking, if the industry information infrastructure was online, firms reported more online use.  For example, in the U.S., both commercial and residential real estate brokerages have pooled their property information in Multiple Listings Services (MLS), maintained by third-party providers for exclusive use by brokers and realtors within a geographic region.  As one broker indicated, this kind of information cartel became an enormous asset to participants:
Its been my experience 5 years ago even 10 years ago or 20 years ago, this thing called the MLS, the Multiple Listing Service, was one of the most highly and closely held secrets [of the industry.]  Real estate agents had access.  Nobody else.  There was no such thing as you being able to find out anything about property. [Environments: infrastructural richness ]
Originally the MLS was a printed booklet, but eventually it went online with broker listings and county recorded data.  Brokers report that they use online MLS’s at least daily, and often hourly.
Lawyers can also rely upon a rich industry infrastructure.  Legal data has been compiled and offered online through third-party vendors like WestLaw and LEXIS since the 1970s; and public taxes have built a network of law libraries that provide free access to primarily print-based materials, but also offer fee-based access to online information.

The interactions of attorneys, real estate brokers and biotech firm members are shaped by industry environments, and by the nature of their firms’ affiliations.  Client relationships, interorganizational networks, and information infrastructures both constrain and enable organizational individuals.  Clearly, client demands exert a major influence on data gathering practices and the ways that people use ICTs.  Pension fund clients require more documentary forms of communication and investment decision justification.   And regulators rely on documentation from the firms they regulate to provide evidence of compliance.  Interactions between organizational individuals, and their counterparts in regulatory agencies or financial underwriting firms, often involve the exchange of large documentary data sets.  The shear volume of data can make these interactions fundamentally different from other types of communication, as this biotech scientist pointed out when describing interactions with the FDA while developing a new drug:

Every single protocol is filed with the FDA...At the end you'd file what's called an NDA, which is a New Drug Application.  That NDA contains everything that you've ever done on that particular drug.  So it can be 300 volumes of stuff.  It goes by truck sometimes to Washington.
 [Interactions: documentation]
In real estate, the information package is substantially smaller (perhaps only 3-inches thick), but no less important.  Package contents vary from property to property, and from investor to investor, but most brokers present potential buyers of commercial properties with a package that contains a standard set of data that can be used to profile a property in a way that allows for a comparison against competing properties and other competing investment opportunities.
If there’s an individual buyer, he may have a different motive than an [pension fund] buyer, but from the standpoint of packaging the information, we’re going to package it pretty much the same because you need to compare one real estate investment with another real estate investment and make sure that you’re comparing apples with apples.  And there’s a lot of unwritten rules that are predominant in the industry to ensure that everybody is viewing an investment in somewhat the same manner. [Interactions: making information actionable]
These standardized packages make information actionable, and their construction often involves the use of an array of analysis and presentation software by brokers and their support staff.  Getting all these software tools to work together and to produce an impressive information package is not an easy task.  Firm members must frequently exercise a great deal of ingenuity to transport data from one format to another, so that it can be graphically displayed or econometrically modeled.  So, although an organizational individual may not often have an opportunity to exercise much discretion in selecting the information resource, she plays an essential role in making the chosen resource "work" for her firm.  This regulatory director explains that, contrary to sanitized stories about how ICTs are used, in biotech firms, “users” are continually challenged to redesign the organizationally chosen toolset "in-use:"
Well, it would sound from listening to me like it was consistent and we did it in the same way every time.  Don’t believe it.  There are always little wrinkles and of course, as technology advances, there’s different ways to gather data. There’s different ways to format it, to assemble it, to disassemble it, to review it.  That is an additional part of the challenge. [Interactions: design in use]
Organizational individuals, like those in this biotech company, continually develop, modify and communicate effective data gathering and data management practices.  In these cameo depictions, we see that they act primarily as firm members, performing specific and sometimes multiple roles while interacting with affiliates on behalf of the firm. Their actions are socially embedded, and highly constrained by decisions made at the organizational level, including which clients the organization will serve.  Their interactions take on only the limited number of forms that are deemed legitimate within the industry.  As a firm member, there are prescribed ways in which one can interact with others.  For example, when asked about the ways in which physicians might be recruited to perform clinical trials for her biotech firm, this researcher explained that those interactions must be face-to-face:
So you do [online] literature searches first sometimes.  And then you'll identify some physicians that you think might be able to answer a lot of the questions... You meet with those people.

Interviewer:  “Now, will you go to see them?  Will they come to see you?”

Usually we go to see them because if you're thinking about doing future clinical trials, what you also want to do is start evaluating these people as potential investigators.

Interviewer:  “Would you ever ask someone to be in a clinical trial that you didn't go and meet?”

No, you cannot do that.  It’s a regulation that you have to go see them and they have to meet this way... [Interactions: firm members]

While using ICTs to obtain, package and exchange information, firm members are simultaneously constructing identities – for their firm, for themselves, for their competitors, and for their clients.  These identities constitute much of the content that ICT use entails.  When presenting property material to a client, for example, brokers frequently utilize sophisticated multimedia technologies.   Both the package and its presentation signal brokerage firm competence.  Brokers believe that these communicate a sense technological mastery that is becoming more and more important to their clients:
The packages have to be fancier.  They have to have more color graphics, aerial photographs, more bells and whistles.  It takes a lot more to do the same project.  In fact, we've really had to gear up and retrain so that we can do really dazzling packages.  The guys can't compete with a package that would have been suitable last year or two years or three years ago.  [Identities: presentation]
As a firm member, the broker will share in this organizationally competent identity.  Additionally, a broker will profile herself to the client, to show what she has personally done with other properties and for other clients.  Before she can do that impressively, however, she needs to build a track record:
You have to have an awesome resume to go into an AllState and say "I can move this one billion dollar portfolio for you."  [Otherwise, t]hey just won't even listen.  I mean the competition becomes – it’s a select few people that can do that. [Identities: multi-level identities]
Profiling—the construction of informational vignettes that characterize particular aspects of a firm or an individual--was reported to be one of the most common uses of online services in all industries of the study. Members of biotech firms report that they frequently profile their potential partners or competitors, by scouring online databases for information about patents, acquisitions, product sales, and corporate officer biographies.   Real estate informants report that they profile properties, markets, themselves, their firms, and also their clients.  Some profiles are constructed to get a quick snapshot of a firm or individual.  When encountering a client for the first time, a broker might have his research group perform cursory credit checks and database searches to see if the potential client is a “real player” or just “wasting his time.”  Attorneys also profile clients, judges and expert witnesses.  A lawyer will carefully research the experts she plans to call, but, before the trial, she can only make an educated guess about who the other side might call. When, during a trial, the names of the adversary's expert witnesses are eventually revealed, a law firm with a well-resourced information infrastructure can quickly profile those witnesses and provide that information to its attorneys on the case, just in time, as this librarian relates:
I actually did a really interesting case with one of those, where the case was ongoing in a different part of the state.  And I went every end of the day - luckily court tends to wind up about 4 or 4:30 so they could still get a hold of me - and they'd tell me who the defense was going to call the next day.  And I would work that evening to pull up information about them, and fax it to our attorney so that when they went into court the next morning, they were prepared to talk to this person... [This] was a securities case, and we were looking that these people were still actively involved in the securities industry. [Identities: profiling]
 Profiling is also a means of controlling identity perceptions about oneself.  A firm may be concerned with its own organizational identity, especially if it is a national or international organization.  It may profile itself, as one national real estate brokerage does, to evaluate its overall strengths and weaknesses:
[CAInvest ] has done a lot of research on this just by looking at our own [databases.]...[CAInvest] is constantly looking.  At our corporate office, they know every single deal that's been done by every office in the country.  They're constantly combing through that data to see what can they find, and to learn about who is our target.  Know what markets you really excel in and know what markets you're wasting your time and your effort and your money in going after.  Because there are a lot of different segments to the real estate markets.  And some of them [CAInvest] is set up to perform really well in, and some we aren't.  We keep stubbing our toe. [Identities: self-monitoring]
When perceived technological competency enhances the identities of brokers and their brokerage, data gathering practices and presentation skills can migrate, and roles can change within the firm.  At one commercial brokerage in our study, new brokers are purposefully exposed to information resources as part of their first year of professional training.  During that time, they actually act as research assistants. This hybrid state is only temporary, but it helps new brokers in the firm acquire the data gathering and information presentation skills they will need to bring clients into the firm.  In the legal industry, hybrid roles are more prevalent and more permanent.  Intellectual Property firms, for example, hire lawyers with technical training and Bachelor of Science (BS) degrees in addition to their JD degrees, because this is a prerequisite for credentials that those attorneys must hold to practice patent law.  Law librarians may also acquire hybrid skills. At least one in our study has practiced law, and her experience can be very helpful to firm attorneys who know how to utilize it:
It think there are a number of law librarians who have law degrees.  I think that my firm utilizes my law degree, and a lot of law firms don't.  Or they don't recognize the value of it. And I did practice law for a couple of years before coming into here.  So particularly my litigation experience, working with clients, I think allows me to anticipate and understand and be able to participate with an attorney in that particular mode better than people that have not.  [Identities: hybrid]
Hybrid training helps people transition up the career ladder, and it also aids those on the upper rungs.  Interestingly, the same training system that upskills new commercial real estate brokers also helps older brokers remain productive, who have neither the time nor the inclination to acquire new data gathering or high-tech presentation skills, because the “trainees” provide research services to established brokers, as this research director explains:
[The training system is] a big asset to these guys also, because if I'm a 55 year old broker, I probably don't know anything.  I've heard of Windows, but that's all.  And it gives these [trainees] a much greater asset in marketing themselves to a broker.  If you don't ever want to touch a computer, then you'd better have a [trainee].  And that's a young guy who knows all the systems.  So it works out pretty well that way. [Identities: expert/novice]
The older broker benefits from this relationship, and the novice broker acquires valuable professional knowledge by being apprenticed to the expert broker.  Broker teams like these mix the novice/expert – expert/novice identities of younger and older brokers in a way that benefits the brokers and their firm.

The foregoing examples provide a socially rich view of the “user” of online services.  Much of the data we have presented here characterizes the activities of commercial real estate brokers. We selected these examples because when individual brokers are juxtaposed against atomic “users”, their complex, highly context-dependent use of ICTs presents a striking contrast to acontextual “user” concepts.  By supplementing these interview excerpts with confirming and supporting examples about organizational individuals in law firms and biotechnology companies, our analysis transitions beyond the individual level to the organizational and industry levels.

Throughout this set of examples, we can see how ICTs are used, when they’re used, who uses them, and why.  It becomes less clear, however, where to draw a boundary around the “user”. The “user” is enmeshed by a network of relationships that mobilize the exchange of information and the use of ICTs.  These interorganizational networks take shape within and extend beyond particular industry environments.  The interactions that firm members engage in are shaped by these environments and by the nature of interfirm affiliations.  When ICTs are used as part of those interactions – to package, present and exchange information -- they also construct identities for firms and their members.

7.0  Discussion

Our analysis suggests that organizational ICT “users” are better understood as firm members whose ICT-related actions can be characterized along four interdependent dimensions that emerge from our findings: affiliations, environments, interactions, and identities.  Although no existing models neatly depict firm members in terms of these broad categories, a synthesis of institutional theory concepts provides a good basis for understanding firm members who use ICTs as social actors.

Institutional Elements
In section 4, we gave a brief overview of institutionalist perspectives. Our data support these general ideas, but to build an alternative model that ties together the everyday activities of ICT “users” portrayed in our data examples, we need to develop a better sense of the interplay between conceptual constructs.  For this, we rely on Scott’s (1995) theoretical synthesis that examines how structured social arrangements vary in complex ways depending on whether institutions are conceived as regulative, normative or cognitive systems; and how they are carried and at what level they operate.  (See Figure 1.)

Figure 1: Institutional Pillars, Carriers and Levels (a 3-dimensional representation derived from Scott, 1995 pp. 34-59)

According to Scott, theorists have usually stressed one of the three “pillars” of institutional thought: regulative, normative or cognitive. (Refer to the x-axis of Figure 1.) The regulative view gives primary emphasis to the role of oversight, mandates, coercion and sanctions to establish and maintain formal and informal systems of behavior.  The normative view draws on the concepts of values and norms. These define what people should do and prescribe how things should be done, and they legitimize the role-based actions and cultural routines that particular individuals perform as they continually construct social institutions.  The cognitive view adapts symbolic interactionist  concepts to the institutional framework to explain how individuals’ everyday actions are constrained by the overarching infrastructures that provide the resources for and shape the interactions of negotiations in their social worlds. Each pillar provides a different basis for legitimacy--conformity to rules, a moral value basis, or a common identity – which may work together or be in conflict.  Thus, each of the three pillars reflects a different viewpoint, which can result in very different evaluative assessments of the phenomena under study.

Drawing on the works of Jeppersen (1991) and Giddens (1984), Scott describes how these institutional elements are “carried” or reproduced—through “cultures, social structures, or routines (and, perhaps also by technologies)” (Scott, 1995; p. 60.) (Refer to the z-axis of Figure 1.)  Cultural carriers are codified patterns of meanings and rule systems that may operate at many levels, from belief systems to organizational cultures.  Social structural carriers are characterized by networks of social positions, or role systems. These structures may be erected to exercise governance, or to differentiate codified systems, like academic departments in universities.  Routines as carriers reflect the tacit knowledge of actors and the habits and procedures that may be based on their unarticulated understandings.  The institutions and their carriers operate at several different levels, from organizational subunits, like marketing departments, to global systems, like international fiduciary regulations. (Refer to the y-axis of Figure 1.) Institutional theorists do not examine phenomena exclusively at the individual level.  Their research focuses on situated action by firm members or other actors in social settings where the influences that shape behaviors and identities can be examined as integral.  Whereas the “user” concept limits understanding of constraints to the person and the ICT resource, institutionalist perspectives broaden that understanding into the wider context of ICT use.

Scott’s synthesis does not contain a social actor model to contrast against the ICT “user,” but it does relate actors situated at multiple levels to other actors, locations, social processes and technologies. From this basis, we can reexamine each of our four analytical categories, affiliations, environments, interactions and identities, to construct a theoretically supported model of a social actor (see Table 4.)

Our data exemplify how social actor relationships are shaped by networks of organizational affiliations (Affiliations: client demands ).  When a commercial real estate firm takes on a pension fund client, that new affiliation brings with it particular expectations for service.  Normative institutionalist concepts that highlight prescriptive practices and obligatory behaviors help to show how such networks of clients, professionals and industry firms structure an organizational field at national and even global levels through interfirm relationships.  Practices that become routinized within one particular firm may become standardized throughout an industry.  The concept of isomorphism characterizes how organizations come to adopt similar market approaches, how they develop industry standards, and how they define legitimate forms of interaction. DiMaggio and Powell (1991) describe two types of isomorphism: competitive and institutional.  Competitive isomorphic pressures are market-related, and tend to push organizations toward improving efficiency or toward better marshalling of scarce resources.  Institutional isomorphic pressures come from other organizations in the industry environment.  These may be regulative (e.g. coercive, legally imposed restrictions that regulate the sale of drugs), cognitive (e.g. mimetic imitation of practices developed by leading industry organizations) or normative (e.g. codes of conduct associated with industry professionals)  -- depending on the affiliation or relationship that is established. These relationships are dynamic, and their related informational exchanges change with “flows” of capital, labor, and other resources (Affiliations: load shifting arrangements).  In many cases, obligatory practices like information gathering for government agencies such as the FDA may also be carried through networked social structures along with capital exchanges from one organization to another, exerting regulative isomorphic pressures.

As relationships change, interaction practices migrate within and across organizations (Affiliations: changing jobs).  This can happen via load shifting arrangements, through the proliferation of professional cultures like CPA’s and MBA’s, or as people within an industry change their employment affiliations from one firm to another. Mimetic and normative isomorphic pressures can precipitate the adoption of practices, particularly if new organization members already have such skills.  When the competition gets tough, real estate brokers feel they must imitate the sophisticated, ICT-enable presentation formats of the leading brokerage firms to signal a high level of competence.  To support the intuition-based decisions of a savvy pension fund investor, brokers must compile a set of quantitative analyses that accounting professionals will consider adequate to justify a multi-million dollar loan package – and they need a skilled research staff on hand who can do this.

Affiliations are multi-level, multi-valent, multi-network (Affiliations: full service). A firm’s relationship with a transnational client to whom it sells investment properties may be simultaneously local, regional, national and global. The interfirm relationships that support this complex relationship may cross departmental and organizational boundaries, and require legitimization within the norms and values of very different professional and regional cultures.
Practices become institutionalized across organizations through isomorphic pressures.  When industry practices become standardized, social actors may be severely constrained in the kinds of interactions they may initiate and the kinds of ICTs they might use. However, firms may be pressured through a number of different affiliations to use or not use ICTs in their interactions with other organizations. Organizational individuals are simultaneously influenced in different ways to use ICTs, because they belong to multiple, somewhat overlapping, networks, where professionals, regulators and others may differ in their shared views of what is legitimate.

These affiliation-related findings emphasize that when we speak about an organizational ICT “user”, we are really talking about a social actor who is an organizational individual, representing the interests of the organization or subunit (and her own interests in the exchange of various forms of capital) and who uses ICTs to facilitate these exchanges and to service these affiliations. Her focal relationships are work-related affiliations, between realtor and client, for example, where the primary objective is to sell property to the pension fund. These affiliations are more important than how the social actor relates to the ICTs she is using – the use of an online resource like DIALOG or Nexis is instrumental to that primary relationship. As firm members, social actors seek to communicate with others in socially legitimated ways, and often through networked ICTs.  Thus, we would expect that their selection of ICTs in this setting would be more strongly influenced by institutional norms, than by personal preferences.

Theoretical outline of the Affiliations dimension of a Social Actor:
Affiliations: This dimension of understanding ICT use by social actors gives us a way to examine relationships as primary and ICT use as supportive ( i.e. a well-socialized actor does not have primary relationships with ICTs).
Organizational individual representing the interests of the organization or subunit (and her own interests) Normative professional norms, cultural values Social structures, Cultures networks, professional hierarchies  Various

Multi-level systems of networks and organizational affiliations form the backbone of the
organizational environments that exert technical and institutional pressures on firms and their members (Environments: global financial/fiduciary practices). Where financial regulations can be enforced, more information will be gathered to assess the risk of a property investment, for example.  Regulative concepts explain how institutions constrain and standardize behaviors through industry standards, government oversight, internationally sanctioned monetary practices, technical standards and market regulations. All organizations face varying degrees of technical and institutional demands (i.e. isomorphic pressures) from their environments.  Scott (1987) has categorized industries as being more strongly or weakly influenced by these demands.  In prior analyses of our study data, we used Scott's classification to help explain why some firms find online technologies essential, while others use them very little, or not at all.  We found that in industries with high institutional pressures, firms produced more documentation, and so used online services more intensively.  In industries with high technical pressures, firms performed more profiling activities, and used more online services for that purpose.

Despite such general influences, however, environmental dynamics vary greatly among industries (Environments: industry or organizational field). For example, although real estate brokers are relatively unregulated, their pension fund clients can exert additional pressures on them to increase internal efficiencies, and to use ICTs in investor-community-approved ways. We have found that there are as many differences among firms within an industry as there are between firms in different industries (AUTHORS6, forthcoming).  Much of that variation has to do with the set of influences that come from the interorganizational relationships of the firm. Industry organizations are more or less incented to gather data and use information resources depending on the clients they serve or wish to attract.  Thus client relationships have a very strong impact on data gathering practices and the use of information resources.  Firms that work very closely with institutions, such as federal regulators, report gathering more data overall than firms that do not interact with regulators as intensively. And, when firms partner with one another, they may shift the responsibilities for gathering data across organizational boundaries.

As the examples from our study show, the use of online information is embedded in communications among organizations.  Within these informational environments, communications are shaped by industry institutions, they follow particular conventions, and they have a certain legitimacy -- ICTs are part of the organizational environment (Environments: organizational ICT investment).  Firms must invest in new ICTs, like a sophisticated legal document management system, to maintain a viable environment, unless such resources are provided in another manner.  In some cases, ICTs are part of the industry, national or global environment (Environments: infrastructural richness), mitigating the investment costs of individual firms.  In the US, for example, real estate firms have pooled their resources to support the development of critical information infrastructures – the regional multiple listings services (MLS).

While discussing our environmentally-related findings, we have shifted our attention to the organization as a legal entity or the industry as an assemblage of collective actors.  Scott’s  multi-level synthesis of institutional theory enables us to shift our view within this overarching framework to examine different aspects of social actor behaviors and characteristics – to transcend levels, and to look at collective action as well as situated individual action --without departing from the basic theoretical constructs. Affiliations aren’t all of the same type.  They are influenced by contexts we call environments, and vice versa.  Although these two social actor dimensions deeply interpenetrate one another, we are able to tease out different aspects by examining them from both normative and regulative institutional perspectives. Our use of this analytical affordance shows where the views complement each other, providing a richer understanding of the ways in which ICT use is constrained and enabled within informational environments.

Theoretical outline of the Environments dimension of a Social Actor:
Environments: This dimension of understanding ICT use by social actors gives us a way to understand ICT use as constrained and enabled within informational environments.
Organization as a legal entity or the industry as collective actors Regulative industry standards, monetary practices, government oversight  Social structures, Cultures regulatory mandates, industry associations, infrastructures  Various

When  people engage in interactions with clients, regulators, partners and others, they are enacting their firm’s affiliations within constrained environments, and their ICTs as interaction technologies are a fundamental part of these communications and exchanges. Whether as a connected and situated individual or a larger group, the social actor may be simultaneously representing that self, additional local groups, the larger organization, or even the entire industry, depending on which affiliations pertain.  As firm members, people routinely perform socially embedded (role-based), highly specified actions on behalf of the firm (Interactions: firm members), and they seek to communicate in legitimate ways (Interactions: documentation). Regulative concepts can improve our understanding of ICT use in certain interactions. Consider the ways in which legal regulations shape the use of ICTs in the compilation of FDA submissions, and the ways in which they restrict the interactions that biotech researchers can engage in, such as the face-to-face recruiting of physicians for clinical trials.

However, normative and cognitive concepts explain many other interesting phenomena associated with ICT-enabled interaction, particularly with respect to organizational and industry interactions. “At the intermediate levels, individuals operate within particular social arenas, such as educational, work and family settings which carry with them many codified cultural rules and social routines. And in everyday interaction, at the microlevel, individuals appropriate and employ these broader cultural frameworks but also improvise and invent new understandings and interpretations that guide their daily activities.  Individuals are not simply constrained but informed and empowered by these preexisting knowledge and rule systems” (Scott, 1995; p. 51).  In other words, although they seek to communicate in institutionally proscribed ways, organizational individuals often find the need to reshape available resources to accomplish that goal --  to build, design, develop and modify interactions to facilitate “flow” changes (Interactions: making information actionable). When brokers create an information package that profiles a property so that it can be compared to other investments, like a stock purchase or a commodity trade, they are reshaping an informational resource in ways that facilitate the flow of investment capital.

Organizational individuals are also commonly called on to improvise – to draw upon their tacit and explicit knowledge to effectively deploy new ICTs on behalf of their firm, or to cobble together effective assemblages of existing technologies that will support new interactions.  They rely upon organizational norms, professional codes of conduct, and ways of working in particular communities of practice to construct  ICT-enabled interactions (Goffman, 1974; Giddens, 1984.) ICTs are shaped by and also shape interactions within organizational contexts, as well as the kinds of practices that may proliferate through interactions with affiliates. In this way, ICTs become part of the interaction process (i.e. “interaction technologies”) as people transform and embed available informational resources into connections and interactions (Interactions: design in use).

Theoretical outline of the Interactions dimension of a Social Actor:
Interactions: This dimension of understanding ICT use by social actors gives us a way to understand how ICTs are shaped by and shape interactions within organizational contexts and the kinds of practices that may proliferate through interactions with affiliates.
Organizational individual or group representing self, local groups, larger organization (even industry) simultaneously based on relevant affiliations  Normative, Cognitive professional norms, cultural values, imitation, symbolic action Cultures, Routines communities of practice, professional codes, industry norms Organization,  organizational subunit

Interactions (and reciprocal interactions) shape the identities of organizational individuals and the collective identities they construct and present to affiliates, regulators and the general public.
When interactions embed the use of ICTs, social actor identities have an ICT use component (Identities: presentation).  For example, when brokers pitch a property using multi-media graphics and “fly-over” video clips, they mean to convey a sense of their own (and also their firm’s) technological competency – believing that to be a quality that the potential investor will value.  Cognitive institutional theorists, like Coleman (1990) have drawn on social constructionist concepts to explain that “[t]he social construction of actors is not limited to persons: Collective actors are similarly constituted and come in a wide variety of forms” (Scott, 1995, p. 43.)  This ability to theorize about both collective actors and individual actors is an important advantage because, as our data exemplify, ICT-enhanced networks heighten professional, ethnic and multiple other identities (Identities: multi-level identities; expert/novice), and ICT-enhanced interactions among firm members often transcend roles (Identities: hybrid).  In work settings, people may be called upon to use their personal identities, or project-based identities to serve the larger organization, when, for example,  a broker presents his “track-record” as evidence of his own and his firm’s capabilities; or when an attorney must demonstrate to a client that her expertise covers biotechnology as well as the law, in order to maintain client confidence and future business for the firm.

Much of what institutional theorists rely upon to understand identity construction, at all levels, is drawn from Goffman’s (1959,1974) descriptions of the ways in which interactions are shaped or framed by social institutions and enacted by the presenter and her audience.  He observed that over time, people have used many resources, including technologies, to present multiple aspects of themselves to different audiences. The things we own, use and display to others make statements about who we are.  Our technological possessions and competencies are very much a part of identity, and so it is not surprising that social actors use ICTs to construct identities and control perceptions (Identities: profiling; self-monitoring.)  Firm members commonly use online information to profile experts, clients and competitors, and they often use proprietary data to monitor themselves – to see themselves as others would see them.  For Goffman, the self and presentation of the self can only be understood through a person’s interaction with others. That is, identity is co-constructed by interactors, and reciprocity is a fundamental ingredient in identity construction.

ICTs are fundamental to the social construction and representation of reality and the “self” within organizational contexts. People’s identities influence how they work with ICTs; and firm members consciously project identities that reflect their ICT competencies. They appropriate their own identity characteristics from collective units, like the projects and communities of practice in which they participate, and they contribute their own embodied competencies, like ICT use expertise, to collective social actor representations, like their department. But, even though they may present themselves as a coherent collective actor, people within an organizational unit aren’t all alike because of the mix of affiliations, environments and reciprocal interactions in which they engage over time.

Theoretical outline of the Identities dimension of a Social Actor:
Identities: This dimension of understanding ICT use by social actors gives us a way to understand how ICTs are fundamental to the social construction and representation of reality and the “self” within organizational contexts.
“Self”multiply connected, situated representation as organizational individual or collective actor Cognitiveimitation, symbolic action, cognitive frames Cultures, Routines communities of practice, projects Organization,  organizational subunit depending on self representation

The social actor concept
From the foregoing discussion, it is clear that the categories that emerge from our data analysis help to harness Scott’s synthesis of institutionalist concepts as support for an explanation of ICT “users” as social actors who communicate and interact with others through multiple social networks of affiliations. (See Table 4.)

Social actors often have conflicting and ambiguous requirements about the activities they perform, and the socially legitimate ways in which to perform their work as attorneys, biotechnology researchers, inspectors, plant managers, real estate brokers, investors, students or teachers.  Many organizational individuals exercise limited discretion in ICT choice and use,

Table 4: Multi-dimensional View of a Social Actor
CHARACTERISTICS and BEHAVIORS of  connected and situated individuals
EMPIRICAL EXAMPLES in the study data presentation 
Affiliations   Social actor relationships are shaped by networks of organizational affiliations Affiliations: client demands
Relationships are dynamic, and related informational exchanges change with “flows” of capital, labor, and other resources Affiliations: load shifting arrangements
Relationships are multi-level, multi-valent, multi-network (i.e. global/local, local/global, group, organization, intergroup, interorganization, culture)  Affiliations: full service
As relationships change, interaction practices migrate within and across organizations  Affiliations: changing jobs
Environments  Organizational environments exert technical and institutional pressures on firms and their members  Environments: global financial/ fiduciary practices
Environmental dynamics vary among industries Environments: industry or organizational field
ICTs are part of the organizational environment  Environments: organizational ICT investment
ICTs are part of the industry/national/global environment  Environments: infrastructural richness
Interactions Organizational individuals seek to communicate in legitimate ways  Interactions: documentation
Organizational individuals build, design, and develop interactions that facilitate “flow” changes  Interactions: making information actionable
ICTs become part of the interaction process, (“interaction technologies”) as people transform and embed available informational resources into connections and interactions Interactions: design in use
As firm members, people perform socially embedded (role-based), highly specified actions on behalf of the firm  Interactions: firm members
Identities  Social actor identities have an ICT use component  Identities: presentation
ICT-enhanced networks heighten ethnic and multiple other identities (global/local tension) Identities: multi-level identities  Identities: expert/novice
ICT-enhanced connections among firm members transcend roles (project-based)  Identities: hybrid
Social actors use ICTs to construct identities and control perceptions Identities: profiling  Identities: self-monitoring

since they are constrained by organizational and institutional contexts.  Social actors interact within organizational environments, where coordinations center around the exchange of resources and information between members of firms and institutions.  Social actors commonly use computers, information products and other ICTs in their interorganizational and interpersonal interactions.  However, social actors are not primarily “users” of ICTs.

Table 4 represents our attempt to synthesize these ideas into a multi-dimensional view of a social actor. We believe that this social actor model of interpenetrating layers can help researchers better examine the interplay between agency, ICTs and structure in organizational settings.  As we have noted, many of these ideas about social actors are recurrent in other research analyses, but, until now, they have not been systematically selected and integrated into a guiding research and design model.

8.0  Conclusions

Other analysts and theorists have contributed to a rich literature about social actors, and, with a bit more elaboration, we could have used their work to further support our model, rather than relying entirely on Scott’s synthesis.  Although IS researchers have called for a better integration of insitutionalist theory into ICT-related studies (Orlikowski and Barley 2001; Orlikowski and Iacono, 2001), and this paper responds to that call, sociologists who have theorized about technology use in recent years have also provided important new insights about social actors and their technologies that deserve consideration (Castells, 1996; Latour, 1987.)

Alternative paths toward a social actor concept
According to Manual Castells (1996), social networking has become the main principle of social organization. His analyses show that, at the beginning of this new millenium, organizations and economies are becoming more informational, more global and above all networked.  Interactive information networks are taking shape among organizations as well as individuals (Castells, 2000.)  They are reinforced through new transportation and communication technologies, so that they often extend nationally and internationally. Castells stresses that globalization is a key phenomenon shaping what he calls “the network society,” and that ICTs are key components and enablers of globalizing practices among organizations and across markets.

Even if we understand them as instruments of mediation which affect social and power relations by their capacity to connect and disconnect, Castells contends that ICTs are limited – and here institutionalists would agree.  Coordination among institutions, organizations and individuals – particularly global coordination -- is only achieved through chains of (often standardized) social interactions (such as, deregulations of international trade and global financial transactions.)  Ultimately, Castells concludes that all networks are networks of individual actors (Castells, 2000.)  However, these individuals are clearly not isolated, atomic ICT “users,” but rather socially embedded actors who take on multiple roles as they use ICTs to act on behalf of their organizations and themselves -- simultaneously reshaping identities through those interactions (Castells, 1997.)

Castells’ ideas connect social actors to the global network society, and help to paint the background against which we can view social interactions, but they don’t help us further characterize social actors.  Actor-network theorists, on the other hand, have also given ICTs a central focus in their theorizing about social systems, and have carefully pointed out the differences in stability within the networks they have studied (Latour, 1987; Woolgar, 1991; Callon, 1991.)  But, perhaps the most important observation by actor-network theory (ANT) researchers is that people together with their technologies comprise social networks.  Affiliations among individuals, groups and organizations entail the use of ICTs to varying degrees –all networks are heterogeneous socio-technical actor-networks--the technical and the social are inseparable.  This perspective helps to explain further how commercial broker identities, for example, have acquired a technology component—social actors are inextricably tied to the technologies they use to present themselves.

Another key tenet of ANT is that actors  define one another through interactions that often involve intermediaries, including ICTs and other people (hybrids.)  Intermediaries translate the actions and interests of one actor into the actions and interests of another, thereby aligning the network to enable collaboration and coordination.   Such alignments are often fragile, and ephemeral, requiring continual effort to maintain a series of network connections.  The networks are also fundamentally dynamic, often requiring novel use of existing or new ICTs and the reconfiguration of organization member roles.  These concepts provide a powerful lens through which to view the social actor and her interactions through ICTs.  They help us see more clearly that brokers and biotech researchers, when acting as firm members, are intermediaries, who perform socially embedded and highly specified actions on behalf of their firms in ways that align the interests of pension fund investors and regulatory agencies through interactions that entail the use of ICTs.

In application, ANT theory  has contributed important explanations of network level phenomena, such as network mobilization (through enrolment and translation) during the development of new technology projects (Akrich, 1993; Latour, 1996; Tuomi, 2001.)  Although ANT theorists tells us that networks are often dynamic and that social actors are pliable, other ANT concepts suggest that both may also become inflexible and irreversible if networks stabilize and intermediaries become taken for granted.  In our view, the key concepts of actor-network theory could also provide an integrated explanation of the social actor dimensions and characteristics that we exemplified earlier with our study data.  However, several variations on ANT concepts have been explored over the years, and although some consensus has developed around particular concepts, ANT researchers themselves disagree about exactly what constitutes ‘ANT theory’, as it has evolved over time (Latour, 1998).

Other analysts have helpfully focused on particular aspects of the social actor.  Geser (1992), for example, has treated the organization itself as a social actor to examine collective action and to explain the dynamics of multiple, simultaneous, often conflicting, interactions among organizational actors and between their internal sub-units.  Munck (1995) has also theorized about collective action by considering social movements to be social actors that interact with existing institutions through the efforts of movement organizers to develop strategies and build movement identity.  Our own approach retains this focus on interconnection and action, but more closely follows Touraine’s (2000) general method for studying social actors by allowing the social actor unit to vary in accordance with self-representation (i.e. as an individual, a group, an organization, or a social movement) and relationships to other actors.

Moving beyond the concept of “users”
We believe that the social actor model provides a better way to conceptualize ICT research and design--it can move IS research beyond the concept of "users." As a framework for data collection and analysis, it provides a theoretically sound mechanism, via institutionalist theory synthesis, for transcending levels, and for examining characteristics and behaviors from regulative, normative and cognitive perspectives as they become relevant.

Unlike some commonly used IS models, the social actor model is not tied to individualistic concepts that don’t scale up.  IS researchers have tried to augment individualistic approaches like TAM with contextual elements to increase the predictability of their models (Venkatesh and Davis, 2000; Karahanna and Straub, 1999).  Clearly, they recognize that more richly contextualized models can provide better understandings of ICT use and adoption.  However, their theoretical bases (TRA, TPB) don’t easily allow for augmentation of this sort.  In contrast, institutional theory doesn’t really allow for socially thin analyses. When building a context-rich alternative to an atomic individualistic “user”, we contend that the thinking should come from theory that understands cognition as inextricably contextualized phenomena. Cognitions matter, but, as Scott’s synthesis emphasizes, within organizations these are institutionally shaped, and cognitive changes are negotiated through social interaction.  Cognitive science views rarely go beyond individual, internalist sources for generating constructs to guide IS research.  However, Scott’s synthesis provides an important conceptual gadget that allows institutionalist analyses of the social actor to vary in scope according to self-representation, as Touraine has recommended. We believe that it is important for IS researchers to have a theoretical basis for their studies that allows for that kind of transition, and that facilitates the multiple perspectives needed to see all sides of a multi-dimensional social actor.

As presented here, the social actor  model provides a bridging mechanism for IS researchers to use to pursue the research agenda called for by leaders of the field (Orlikowski and Barley 2001; Orlikowski and Iacono, 2001.)  It is eminently extensible.  We have tried to describe it as an association of deeply interpenetrating layers that generously allow for much modeling work to be done by others.  In our own studies, we have found it to be a helpful guide for data collection and analysis – providing a framework for insightful understandings of intranet integration (AUTHORS8, 2002) and the role of ICTs in identity construction among oceanographic scientists (AUTHORS9, 2002).

Moving beyond “user” terminology
Our multi-dimensional social actor view encourages us to think of social actors as webs of people and technologies working together—continually reconstructing roles (e.g. developer, client, user) and shifting identities to reflect what’s going on in the world around them.  Multiple technologies are intrinsic components of identity – the information system is part of the group.  Instead of labeling people as passive “users”, we begin to view them more accurately as selectors of information for embedded interorganizational communications, and as active facilitators of changing information infrastructures.  They are ICT “designers-in-use”, who take advantage of opportunities to modify work roles and develop new information practices; and they are ICT–use proliferators who spread information practices within and across organizations.

So, an important step the IS community could take in moving beyond the “user” concept (at least symbolically) is to drop the term “user” from information systems development vocabularies, or to at least minimize unreflexive use of the term.  “The power of words is not total, but they may subtly and indirectly inhibit the adoption of new areas of research and approaches to development” (Grudin, 1990 p. 269.)  Based on the works we cited earlier, it seems clear that the term “user” has helped to reinforce the “user” concept, and we hope that it can be replaced by more specific labels.  Therefore, we encourage IS researchers and designers to radically reduce their use of the term “user” --  only resorting to it when no other role or term is workable.  We don’t expect people to simply perform a mental “search-and-replace” operation.   In our MIS Quarterly article review (see section 3), we noticed that several papers effectively avoid the term “user” by referring to people by their role (e.g. managers, department members, clerical staff, nurses and librarians.)  To eliminate the term “user,” one must be more explicit about interactions among people, organizations and their technologies.  We have actually tried to follow this advice ourselves, but we have sometimes slipped back to using the term “user.”  The term and the concept have become institutionalized within the IS community, and as with all resilient institutionalized practices, change will be difficult to initiate. Nevertheless, as Hirschheim et al. (1996) have pointed out, within the domains of IS development change, language change is the least difficult to effect.  In that case, we would venture to suggest another terminology change to help IS researchers and designers move beyond “user” concepts: that is, to replace the commonly used expansion of the acronym “IT” (i.e. information technologies) with one that is better suited to a network society: “interaction technologies.”

Throughout this paper, our goal has been to convince individualistic IS researchers and designers to move their work onto the social canvas. Our social actor approach provides the community with something to work with other than the socially thin concept of “user.” We realize, however,  that in use, the approach may reshape the kinds of questions researchers can investigate and reconfigure the methodologies they find effective.  To construct a socially rich view of ICT use, we have drawn out the more reliable (and researchable) aspects of institutional theory, and set out a direction for new theoretical work in IS.  More can be done, and so we invite IS researchers and designers to augment and refine our view of a multi-dimensional social actor. It will take a concerted effort by IS researchers to shape a comprehensive understanding of interaction technologies in the network society.  Such an effort could make an important contribution, not only to IS-related disciplines, but to social theory, as well.


Akrich, Madelein (1993.) "Inscription et Coordination Socio-Techniques: Anthropologie de Quelques Dispositifs Energetiques," Paris: Ecole Nationale Superieure des Mines de Paris, These pour le doctorat Socio-Economie.

AUTHORS1 (1977.) “The Organizational Context of User-Centered Organizational Design,” MIS Quarterly, Vol. 1 No. 4, pp.41-52.

AUTHORS2 (1987.) “Defining the Boundaries of Computing Across Complex Organizations" in R. Boland and R. Hirschheim (eds.), Critical Issues in Information Systems, New York: John Wiley.

AUTHORS3 (1992.) “Behind the Terminal: The Critical Role of Computing Infrastructure in Effective Information Systems’ Development and Use,” in William Cotterman and James Senn (eds.), Challenges and Strategies for Research in Systems Development, New York: John Wiley, pp. 153-201.

AUTHORS4 (1994.)  “Digital Library Design for Usability,” Digital Libraries ’94 Conference.

AUTHORS5 (1997.) “Interorganizational Relationships and Information Services: How Technical and Institutional Environments Influence Data Gathering Practices,” Unpublished Ph.D. Dissertation, University of California, Irvine.

AUTHORS6 (forthcoming.)  "Informational Environments: Organizational Contexts of Online Information Use," Journal of the American Society for Information Science and Technology.

AUTHORS7 (forthcoming.) “Socially Rich Interaction through Information and Communication Technologies: Moving Beyond the Concept of Users,” in R. Zmud (ed.) MISQ Workshop Proceedings, May, 2000, Norman, Oklahoma.

AUTHORS8 (2002).  “Intranet Boundaries: Social Actors and Systems Integration,” Proceedings of 15th Bled Electronic Commerce Conference, e-Reality: Constructing the e-Economy, Bled, Slovenia, June, 2002.

AUTHORS9 (2002).  "Social Scientists: Managing Identity in Socio-Technical networks," Proceedings of the Thirty-fifth Annual Hawaii International Conference on Systems Sciences. Kona-Kailua, Hawaii, January, 2002.

AUTHORS10 (1980). “Social Analyses of Computing: Theoretical Perspectives in Recent Empirical Research”, Computing Surveys, Vol. 12, No. 1, pp. 61-110.

AUTHORS11 (1996). "How the Marriage of Management and Computing Intensifies the Struggle for Personal Privacy" in David Lyon and Elia Zureik (Eds) New Technology, Surveillance and Social Control, University of Minnesota Press.

AUTHORS12 (1988.) “The Institutional Character of Computerized Information Systems,” Office: Technology and People (5:1), pp. 7-28.

Baldwin, Nancy Sadler and Ronald E. Rice (1997.) “Information-seeking behavior of securities analysts: Individual and institutional influences, information sources and channels, and outcomes,” Journal of the American Society for Information Science, v48, n8, pp. 674-693.

Bannon, Liam J. (1991.) “From Human Factors to Human Actors: The Role of Psychology and Human-Computer Interaction Studies in Systems Design” in J. Greenbaum and M. Kyng (Eds.) Design at work: Cooperative Design of Computer Systems.  Hillsdale: Lawrence Erlbaum Associates, pp. 25-44.

Beath, Cynthia M. and Wanda J. Orlikowski (1994.) “The Contradictory Structure of Systems Development Methodologies: Deconstructing the IS-User Relationship in Information Engineering,” Information Systems Research, Dec. 1994, pp. 350-377.

Bellardo, T. (1985).  An Investigation of Online Searcher Traits and Their Relationship to Search Outcome, Journal of the American Society for Information Science, 36(4):241-250.

Berger, Peter L. and Thomas Luckmann (1967.) The Social Construction of Reality. New York: Doubleday Anchor.

Borgman, C. L. (1989).  All Users of Information Retrieval Systems Are Not Created Equal: An Exploration of Individual Differences, Information Processing and Management, 25(3):237-251.

Blomberg, J., L. Suchman and R. Trigg (1994.) “Reflections on a Work-Oriented Design Project,” in R. Trigg, S. I. Anderson, and E. Dykstra-Erickson (eds.), PDC ’94: Proceedings of the Participatory Design Conference (pp. 99-109.) Computer Professionals for Social Responsibility, Palo Alto, CA.

Callon, Michel (1991.) “Techno-economic networks and irreversibility,” in John Law (ed.) A Sociology of Monsters: Essays on Power, Technology and Domination, London: Routledge.

Castells, Manual (1996.) The Information Age: Economy, Society and Culture. Vol I: The Rise of the Network Society.  Oxford: Blackwell.

Castells, Manual (1997.) The Information Age: Economy, Society and Culture. Vol II: The Power of Identity.  Oxford: Blackwell.

Castells, Manual (2000.) “Toward a Sociology of the Network Society,” Contemporary Sociology, Vol. 29(5):693-699.

Czarniawska-Joerges, Barbara (1992). Exploring Complex Organizations, Newbury Park, CA: Sage Publications.

DiMaggio, Paul J. and Walter W. Powell (1991.) “Introduction” in W.W. Powell and P.J. DiMaggio (eds.) The New Institutionalism in Organizational Analysis, Chicago: University of Chicago Press.

Dutton, W. H., J. Wyer and J. O'Connell (1993). "The Governmental Impacts of Information Technology: A Case Study of Santa Monica's Public Electronic Network." In Rajiv Banker et al. (Eds) Strategic Information Technology Management, Harrisburg, PA: Idea Group.

Ehn, Pelle (1988.) Work-oriented design of computer artifacts. Falköping, Sweden: Arbetslivscentrum /Almqvist & Wiksell International.

Fiske, S. T. and S. E. Taylor (1991). Social Cognition. New York: McGraw-Hill, Inc.

Friedland, Roger and Robert R. Alford (1991.)  “Bringing Society Back In: Symbols, Practices and Institutional Contradictions,” in W.W. Powell and P.J. DiMaggio (eds.) The New Institutionalism in Organizational Analysis, Chicago: University of Chicago Press.

Gasson, Susan (1999.) “The Reality of User-Centered Design,” Journal of End User Computing, Vol. 11, No. 4.

Geser, Hans (1992.) “Towards an Interaction Theory of Organizational Actors”, Organization Studies, Vol 13(3).

Giddens, A. (1984.) The Constitution of Society. Cambridge: Polity Press.

Goffman, Erving (1959). The Presentation of Self in Everyday Life. New York: Doubleday.

Goffman, Erving (1974.) Frame Analysis, Cambridge, MA: Harvard University Press.

Greenbaum, J. and M. Kyng (1991.) Design at Work: Cooperative Design of Computer Systems, Hillsdale, NJ: Lawrence Erlbaum.

Grudin, Jonathan (1990.) “interface,” in Proceedings of the Conference on Computer Supported Cooperative Work, Los Angeles, CA, October 7-10, 1990, pp. 269-278.

Gutwin, Carl and Saul Greenberg (1998.) “Design for individuals, design for groups: tradeoffs between power and workspace awareness,” in Proceedings of the ACM 1998 Conference on Computer Supported Cooperative Work, Seattle, WA.

Guzdial, M., J. Rick and B. Kerimbaev (2000.) “Recognizing Support Roles in CSCW” in Proceedings of the ACM 2000 Conference on Computer Supported Cooperative Work, Philadelphia, PA.

Hirschheim, Rudy, Heinz K. Klein, and Kalle Lyytinen (1996.) “Exploring the Intellectual Structures of Information Systems Development: A Social Action Theoretic Analysis,” Accounting Management and Information Technology, Vol. 6, No. 1 / 2, pp. 1-64.

Hu, P., P. Chau, O. Sheng and K. Tam (1999.) “Examining the Technology Acceptance Model Using Physician Acceptance of Telemedicine Technology”, Journal of Management Information Systems, (16)2, pp. 91-112.

Jary, David and Julia Jary (1991.)  The Harper Collins Dictionary of Sociology. New York: HarperCollins Publishers.

Karahanna, Elena and Detmar W. Straub (1999.) “The Psychological Origins of Perceived Usefulness and Ease of Use,” Information & Management, 35, pp. 237-250.

King, J. L., V. Gurbaxani, K. L. Kraemer, F. W. McFarlan, K. S. Raman and C. S. Yap (1994.) “Institutional Factors in Information Technology Innovation,” Information Systems Research (5:2), pp. 139-169.

Kraemer, K. L., S. Dickhoven, S. F. Tierney and J. L. King (1987). Datawars: The Politics of Modeling in Federal Policymaking.  New York, NY: Columbia University Press.

Kumar, Kuldeep, Han G. van Dissel and Paola Bielli (1998.) “The Merchant of Prato—Revisited: Toward a Third Rationality of Information Systems,” MIS Quarterly, June, 1998, Vol. 22, No. 2, pp. 199-226.

Latour, Bruno (1987). Science in Action: How to follow scientists and engineers through society, Cambridge, MA: Harvard University Press.

Latour, Bruno (1996.) Aramis: Or the Love of Technology, translated by Catherine Porter, Cambridge, MA: Harvard University Press.

Latour, Bruno (1998.) “Bruno Latour Keynote Speech: ‘On Recalling ANT’”, in John Law and John Hassard (eds) Actor Network Theory and After, Blackwell.

Libmann, Francois, 1990. "Study of technology transfer databases", Online Information 90, London, England.

Malone, Thomas W. and Kevin Crowston (1990.) “What is Coordination Theory and How Can It Help Design Cooperative Work Systems?” in Proceedings of the Conference on Computer Supported Cooperative Work, Los Angeles, CA, October 7-10, 1990, pp. 357-370.

Markus, M. Lynne and Robert I. Benjamin (1996.) “Change Agentry – The Next IS Frontier”, MIS Quarterly, Vol. 20, No. 4.

Markus, M. Lynne and Niels Bjorn-Andersen (1987.) “Power Over Users: Its Exercise by Systems Professionals,” Communications of the ACM, June, 1987, Vol. 30, No. 6, pp. 498-504.

Meyer, John W. and Brian Rowan (1977.) “Institutionalized Organizations: Formal Structure as Myth and Ceremony,” American Journal of Sociology, 83:340-63.

Mishra, P., R. J. Spiro, and P. Feltovich (1996.) "Technology, representation & cognition" in H. von Oostendorp (Ed.) Cognitive aspects of electronic text processing. Ablex Publishing Corporation.

Munck, Gerardo L. (1995.) “Actor formation, social co-ordination, and political strategy: Some conceptual problems in the study of social movements”, The Journal of the British Sociological Association, Vol 29(4).

Mumford, Enid (1995.) Effective Systems Design & Requirements Analysis: The ETHICS Approach, MacMillan.

Mumford, Enid (2000.) “Socio-Technical Design: An Unfulfilled Promise or a Future Opportunity?”, in Baskerville et al. (Ed), Organizational and Social Perspectives on Information Technology, Norwell, MA: Kluwer Academic Publishers.

Nardi, B. and J. Miller (1991.) “Twinkling Lights and Nested Loops: Distributed Problem Solving and Spreadsheet Development,” International Journal of Man-Machine Studies, 34:161-184.

Newby, G. B., M. S. Nilan, and L. M. Duvall (1991).  Toward a Reassessment of Individual Differences in Information Systems: The Power of User-Based Situational Predictors.  In J. M. Griffiths (ed), ASIS '91" Proceedings of the American Society for Information Science (ASIS) 54th Annual Meeting: Volume 28; 1991 October 27-31; Washington D.C.; Medford, NJ: Learned Information; 83-81.

Nicholas, David, Erbach, Gertrud, Pang, Yin Wah, and Paalman, Koren, 1988. End-Users of Online Information Systems: An Analysis, Mansell Publishing Limited, New York, NY.

Norman, Don A. (1986.) “Cognitive Engineering” in D. A. Norman and S. W. Draper (Eds.), User-centered system design, Hillsdale, NJ: Lawrence Erlbaum Associates.

Nye, J. L.,  and A. M. Brower, A. M. (Eds.). (1996). What's social about social cognition? Research on socially shared cognition in small groups. Thousand Oaks, CA: Sage.

Nygaard, K. (1986.) “Program Development as Social Activity” in Information Processing 86.

Orlikowski, Wanda J. and Stephen R. Barley (2001.) “Technology and Institutions: What Can Research on Information Technology and Research on Organizations Learn from Each Other?”, MIS Quarterly, Vol. 25, No. 2, pp. 145-165.

Orlikowski, Wanda J. and C. Suzanne Iacono (2001.) “Research Commentary: Desperately Seeking the “IT” in IT Research—A Call to Theorizing the IT Artifact,” Information Systems Research, Vol. 12, No. 2, pp. 121-143.

Salzman, Hal (1998.) “Beyond User Participation: The Politics of Software Development” in W. Karwowski and R. Gonnetilleke, eds. Manufacturing Agility and Hybrid Automation.

Scott, W. Richard (1987).  Organizations: Rational, Natural, and Open Systems, Second Edition, Englewood Cliffs, NJ: Prentice Hall.

Shneiderman, Ben (1987.)  Designing the User Interface: Strategies for Effective Human-Computer Interaction, Reading MA: Addison-Wesley Publishing Co.

Simon, Herbert A. (1955.) “A behavioral model of rational choice”, Quarterly Journal of Economics, 69:99-118.

Tolbert, P. S. (1995).  Institutional Environments and Resource Dependence: Sources of Administrative Structure in Institutions of Higher Education, Administrative Science Quarterly, 30:1-13.

Touraine, Alain (2000.) “A Method for Studying Social Actors”, Journal of World-Systems Research, VI, 3, Fall/Winter, pp. 900-918.                               .

Tuomi, Ilkka (2001.) “Internet, Innovation, and Open Source: Actors in the Network”, First Monday, Vol. 6, No. 1,  http://firstmonday.org/issues6_1/tuomi/index.html

Venkatesh, V. and F. D. Davis (2000.) “A Theoretical Extension of the Technology Acceptance Model: Four Longitudinal Field Studies,” Management Science, Vol. 46, pp. 186-204.

von Hippel, Eric (1998.) "Economics of Product Development by Users: The Impact of 'Sticky' Local Information," Management Science, Vol. 44, No. 5, pp. 629-644.

Walsham, G. (1997.) “Actor-Network Theory and IS Research: Current Status and Future Prospects”,  in Information Systems and Qualitative Research, A. S. Lee, J. Liebenau, and J. I. DeGross (eds.), London: Chapman & Hall.

Walsham, Geoff (2001.) Making a World of Difference: IT in a Global Context. Chichester: John Wiley & Sons.

Walsham, Geoff and Sundeep Sahay (1999.) "GIS for District-Level Administration in India: Problems and Opportunities," MIS Quarterly, Vol. 23, No. 1, pp. 39-66.

Westrup, C. (1997.) Constituting Users in Requirements Techniques” in Information Systems and Qualitative Research, A. S. Lee et al. (eds.), London: Chapman & Hall, pp. 182-204.

Wilson, Patrick, 1995. "Unused Relevant Information in Research and Development", Journal of the American Society for Information Science, 46(1):45-51, 1995.

Woolgar, S. (1991.) "Configuring the User: The Case of Usability Trials" in J. Law (ed), A Sociology of Monsters: Essays on Power, Technology and Domination.  London: Routledge.

We are grateful for the helpful comments of attendees of the MISQ Workshop, the MISQ reviewers, and other colleagues who have helped to sharpen this discussion, including Nancy Pouloudi, Kristin Eschenfelder and Liz Davidson.

Appendix A - Online Services Study Interview Instrument
IRIS: Interorganizational Relationships and Information Services
Semi-Structured Interview 
<company name>

<phone number>

<address> <date>


<name(s) of interviewer(s)> Systematic Study Design  File: <interview guide>

1. Basic Questions
Who am I interviewing?  What "department" are they in? 

What do they do there?

What is the relationship (of this person) to the rest of the firm?

What is this person's educational/experiential background?

What are their professional affiliations?

What's the major focus/product of this firm? What is its relationship to the rest of the industry?
What is the general technological orientation of this firm? Does everyone have/use a computer?

Does everyone have/use email?

Does the firm have and "attitude" about computer/online use?


Who are their competitors?  What is the competive environment like? 

Are there times when competition is intense?

Are there alliances or cooperative associations?

What about mergers, subcontracting, partnering, outsourcing?

How do regulators interact with the firms?

2. InterOrganizational Relationships (IRs)
Clinical Trials, Expert Witnesses, Subcontracting, Investments Mention we have seen OI resources used during these: 
  • Used in forming these relationships? 
  • Used to check out clients/customers?
  • Used to check out competitors/opponents? 
  • Used during these relationships?
Ask about any other IRs 

What is the context of these IRs?

How many of these over the years?

Persistence over time:  Are these new kinds of IRs for this firm?
Frequency:  How often (in a year) do these IRs occur?
Importance:  Do these IRs bring in a lot of revenue to the firm?

Do these IRs bring prestige to the firm?

Which ones involve OI resource use at this firm?  What fraction of OI use is for IR or outside firm research?

What fraction of OI use is not linked to IR at all?

3. Activity Sequences of IRs
What is done? 0
In what order? 0
Where do information resources fit in? Electronic media? 

Print-based media?

Personal contracts?

How is th eIR established? 0
How is the IR maintained? 0
Were things always done this way? 0
Are any new ways doing these being tired? 0

4. Intermediary Roles (re: Activity Sequences)
Who does what?  * Who gets what type of information?
What is their relationship to other intermediaries inside the firm? * Outside the firm?* Is there a shift from personal contacts to databases?* Make this concrete: How many now vs. before?* Is there a shift from org. contacts to other intermediary contacts? (e.g. more lawyers and consultants doing research?)* Make this concrete: How many now vs. before?
How do they share IR information?  * Professional affiliations?
Is intermediation a driver of certain types of activities? * What kind of activity does it drive?* Is OI researching that kind of activity?

5. Conceptualizations of roles of resources
Accuracy or completeness of identity picture that can be constructed from all resoruces * What's perceived to be best? * Why?* Are some resources more legitimate than others?
Who constructs the identities?  * Gathers the information* Compiles the reports
Who interprets the constructions?

6. OI Usage and Changes Over Time
Overall:  Is there an increase/decrease/shift in kind?

Per unit (economies of scale) or across the board?

In Context: When is it used most?

Are these instances increasing/decreasing/shifting?

What is the relation of IR search to other search occasions?

How much do they use? (yearly/monthly) Prior to online what did they do? 

Paper only?

How critical are OI resources? (substitutability)


Appendix B: MIS Quarterly Article Contents Analysis (1996-1999) [pdf]