|
Article Excerpt
Abstract
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 the selection of information and communication technologies (ICTs) that people make. In many ways, these studies have improved our understanding of how a good in formation 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. Most people who use ICT applications utilize multiple applications, in various roles, and as part of their efforts to produce goods and services while interacting with a variety of other people, and often in multiple social contexts. Moreover, the socially thin user construct limits 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 shap e 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 in formation systems.
Keywords: Social actor model, IS research methodology, socio-technical theory, social interaction, online use, IS users
ISRL Categories: A107, IB02, AH0604, AD0102, GA, GB, AE, AJ
Introduction
Advancing the research about information and communication technology (ICT) use within organizations requires some theoretical advances in understanding organizations (Orlikowski and Barley 2001) and technology (Orlikowski and lacono 2001). This kind of research advance requires a complementary advance in our conceptualization of the user--the active agent in information system use. The most common conception of the user in IS research is of an atomic individual with well-articulated preferences and the ability to exercise discretion in ICT choice and use, within certain cognitive limits.
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 underdeveloped, 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 whom they are trying to persuade with information gathered and packaged through the use of ICTs. Moreover, within the complex social settings that commonly constitute organizations, individuals don't always have the opportunity to choose the systems they would prefer to use. Therefore, when aggregated to predict organi zation-wide activity, a thinly socialized concept of the user leads to frequent overestimates of ICT use (Baldwin and Rice 1997; Dutton et al. 1993; Lamb et al. 2003; Libmann 1990; Wilson 1995).
Within several research disciplines related to information systems (IS) studies, there is a growing realization that ICTs, such as 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 (Beath and Orlikowski 1994; Grudin 1990; Salzman 1998; Westrup 1997). These thoughtful observations and impassioned critiques have motivated more social awareness and more human-centered design practices, but they do not provide an alternative to the user concept that can be coherently integrated into information systems design. We believe that a theoretically based reconceptualization of the user as a social actor is required to move IS research beyond this impasse.
Recent suggestions for advancing IS research have looked to structuration theory and institutional theory for concepts that can accommodate the complexity of designing ICTs and understanding their use in social settings. However, even researchers who have a richly contextualized conception of users as social actors often write ambiguously about agency and workers, so their conceptions can be readily interpreted as either individualistic or social (cf., Orlikowski and Barley 2001; Walsham 2001). To help our peers curb against this tendency, and also to respond to their call for better integration of institutional analysis in IS research, we have begun to model an unambiguously social view of the organization member as a social actor by building on the concepts of new institutionalists (DiMaggio and Powell 1991; Scott 1987, 1995 ) and social constructionists (Giddens 1984; Goffman 1974), as well as the rich descriptions of empirical studies that show how organizational contexts constrain and enable individuals using ICTs. In this paper, we will present our social actor view through a series of critiques and constructions that are based on a multidisciplinary 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 present key 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.
Challenging the User Concept
The user concept reflected in IS research 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 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 coordination through free-market interactions (cf., Malone and 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 nonuse) 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).
Field studies frequently show, however, 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; Lamb 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. Although user studies have improved our understanding of how a good information source fits the people who use it, IS field studies indicate that, from an instrumental perspective, the user concept is too narrowly defined. (2) In this respect, our key criticism of the user concept--that it excludes context--reflects the more general criticism of cognitive social psychology made by Nye and Brower (1996).
STS and CSCW Critiques
A series of studies has strongly challenged the user concept and its exclusion of context from a theoretical perspective, even though it has not provided a concise alternative to the simpler user concept. Building on the early work of human relations theorists and 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 (Ehn 1988; Nygaard 1986). Mumford's (1995, 2000) socio-technical ETHICS approach was particularly influential among Scandinavian 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. 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, have sought 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. Participatory design researchers purport 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; Gutwin and Gre enberg 1998; Guzdial et al. 2000; Nardi and Miller 1991). 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.
Many critiques of the user concept made by a group of researchers, loosely categorized as computer supported cooperative work (CSCW) researchers, draw upon this socio-technical foundation, objecting to the largely artificial separation between ICT developers and ICT users. 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 (Bannon 1991; Grudin 1990; Westrup 1997; Woolgar 1991). Along with other researchers, these authors have noted that users do not hold the same view of themselves that IS analysts do, and they do not like to be referred to as users (Beath and Orlikowski 1994; Grudin 1990; Markus and Benjamin 1996). 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. (3)
The more recent CSCW critiques, referenced above, 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 Luckmann 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 researchers who study ICT use within groups and among organizational collaborators agree 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, therefore, 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; Kling and Elliott 1994). As developers and users work together on system design, power imbalances frequently prevent users from making a real contribution (Blomberg etal. 1994).
Challenges from the European Tradition of IS Research
In the North American IS literature, researchers have criticized the treatment of users during information systems development (e.g., Beath and Orlikowski 1994; Markus and Bjorn-Andersen 1987), but they have not challenged the user concept itself. Researchers in the European tradition of IS research have been more confrontational, basing their arguments on many of the theoretical concepts reviewed above, but also examining critical aspects of institutional power in information systems development and implementation. Several organizational IS studies have stressed the need for a larger environmental scope when dealing with ICT use, noting that individual CT use is influenced not only by organizational contexts, but also by interorganizational, cultural, and global contexts (Czarniawska-Joerges 1992; Lamb 1997; Walsham and Sahay 1999). Some European-tradition IS studies echo the concerns of CSCW studies, indicating that a focus on interactions (between individuals in groups, among groups within an organization, among groups and individuals performing roles in different organizations, etc. ) can better describe the ways in which people come to use ICTs to support their organizational and interorganizational activities. They also emphasize that organization members often fail to use ICTs in expected ways, and frequently reshape technologies to suit their needs (Hirschheim et al. 1996; Kling 1987,1992; Kraemer et al. 1987; Kumar et al. 1998).
European sociologists have also theorized about technology use in ways that provide important new insights about social actors and their technologies (Castells 1996; Latour 1987; Touraine 1988), and these concepts have begun to guide IS research as well (Tuomi 2001; Woolgar 1991). Castells' ideas connect social actors to the global network society, describing a rich environment within which we can view social interactions. Actor-network theorists 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 (Gallon 1991; Latour 1987; Woolgar 1991). Perhaps the most important observation by actor-network theory researchers is that people together with their technologies comprise social networks. Modern affiliations among individuals, groups, and organizations entail the use of ICTs to varying degrees; therefore, all networks can be viewed as heterogeneous socio-technical actor-networks. The techni cal and the social are inseparable, and actor identities necessarily reflect this fusion. (4)
Walshams (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. In our view, the most effective of these IS critiques place an emphasis on the intransigence of institutions and the processes of institutionally constrained action, while acknowledging the potential for agent-directed change via ICT design-in-use (e.g., Bikson 1996; Bowker and Star 1994; Ciborra et al. 2000; Kling and lacono 1984). Their arguments often take root in new institutionalist explanations of organizational change that encompass a more systems-oriented, historically focused understanding of the possibilities for action within social institutions (5) (cf., Friedland and Alford, 1991; Scott 1987; Tolbert 1985). Institutionalist IS researchers 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 (cf., DiMaggio and Powell 1991). Their studies show that, with respect to the adoption, development and use of technologies, the actions of organizations are shaped by the institutional environment. From this perspective, we see that social actors are pressured to perform legitimate actions and interactions within institutionalized arrangements. ICT-related change happens, but it is not entirely agent-directed. Rather, agency is channeled through a complex, multilevel system of networks and organizational affiliations that constitute local and global environments.
Although these user-critical STS, CSCW, and European-tradition IS studies have not explicitly provided an alternative model to the user concept, collectively they indicate that a synthesis of theoretical constructs, supported by empirical evidence, may effectively do so. Their literatures vary in focus, but each contributes to a more robust conceptualization of how, why, when, and where people encounter and use ICTs, and to what effect. They differ in the degree of importance they give to agency and structure, but each demands a more concerted examination of context when seeking explanations and predictions of ICT use outcomes, and each provides some guidance for context-centered research design.
Research Design and Empirical Results
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. Our methodology and findings are critical components of our overall theory development effort. Therefore, in this section and in the referenced appendices, we will present our research design and empirical results in detail to explicate the data-driven construction of the social actor model that is the main focus...
|
