Computer With

As Orlikowski (2000, p. 425) notes, "Technology per se can't increase or decrease the productivity of workers' performance, only use of it can." This simple, yet powerful idea, underpins one of the largest bodies of research in our field: the study of information technology (IT) acceptance and use (e.g., Venkatesh et al. 2003). That work has been enormously beneficial in helping us learn how to get users to use systems, but system use alone is not sufficient to obtain benefits (Seddon 1997); the use must be effective. At first glance, this notion of "effective use" might seem obvious. It is, after all, part of this journal's own mission:

In this article, we show that the concept of effective use is, in fact, quite complex. Moreover, it is extremely under researched. Marcolin et al. (2000, p. 53) identified this problem a decade ago and predicted a shift "from the study of use ... to the study of effective use," but their prediction has not yet been fulfilled. We believe it is important to take that step. (1) The purpose of this paper is to present a theory of effective use that we hope can support and motivate such research.

The way we developed our theory was driven by a second motivation: to consider the unique nature of information systems (IS). That is, we do not wish to propose a theory to explain the effective use of any artifact, such as a theory of human motivations or tool affordances. Rather, we ask what is an information system and what does this imply for what effective use involves, and what drives it? Answering this question requires us to take a specific view on what constitutes information systems and to derive specific implications from this view. As we will explain, we chose a theory known as representation theory that offers a particular view of information systems, and we drew a number of implications from it. Other views of information systems could have, and hopefully will be taken, and other implications could have, and hopefully will be drawn. The work we present is simply an initial attempt to develop a platform for research on effective use, and to develop a theory of a specific IS phenomena (here, effective use) from a general theory of IS (here, representation theory (2)).

We begin by defining key terms and reviewing the relevant literature on effective use and representation theory. We provide an especially detailed overview of representation theory in that section to provide a clear foundation for how we use it in the paper. After providing this review, we present our theory of effective use. This involves presenting a general framework, applying the framework, and deriving two models from it. We then discuss the implications of our theory and conclude the paper.

2. Background

In this section we define effective use and summarize the two bodies of research that inform our theory: research on effective use, and representation theory.

2.1. Defining Effective Use

We define effective use as using a system in a way that helps attain the goals for using the system. Our definition is adapted from Burton-Jones and Straub (2006), who defined system use in terms of a user, system, and task, and defined a task as a "goal-directed activity" (p. 231). To move from use to effective use, we simply shifted the emphasis from using the system to perform a goal-directed activity to using it in a way that helps attain the relevant goal. Four assumptions underlie our definition. First, we recognize that "use" can occur at any level of analysis, but we focus our initial theory on the individual level for reasons of scope. As we show, even at this level it is quite complex. Second, we assume that systems are never used just to use them; rather, they are used to achieve other goals (Gasser 1986). As a goal is a "cognitive representation of a desired end point" (Fishbach and Ferguson 2007, p. 491), we assume that the relevant goal for effective use is simply whatever end point the system is used to attain. Third, we assume that goal attainment has objective qualities; it may be hard to evaluate in some cases, but it is not completely subjective. Operationally, it is assessed in terms of performance (Sonnentag 2002). Thus, we view effective use and performance to be relatively objective notions. Finally, we recognize that different stakeholders (e.g., designers, users, managers) may have different views on the goals for using a system. Rather than taking just one of them, we will explain how researchers can study effective use from different stakeholder perspectives.

Our definition of effective use may initially appear similar to the concept of perceived usefulness (or performance expectancy), which is the degree to which an individual believes that using the system will help him or her better attain significant rewards (Venkatesh et al. 2003, p. 23). However, the constructs differ in scope because effective use focuses on rewards that stem from the way a system is used, whereas perceived usefulness focuses more broadly on rewards that stem from use, not just the way it is used (e.g., it could include rewards that stem from the context in which an IS is used). The constructs also differ in terms of raters, as perceived usefulness refers to a user's expectation or perception (i.e., it resides in the user's mind), whereas we view effective use objectively (i.e., it is an observable behavior).

2.2. Past Research on Effective Use

After a detailed review of the literature, we found no studies on the nature or drivers of effective use for systems in general. However, three small streams of research inform this area. First, there are studies that have not studied effective use explicitly, but have done so implicitly by studying outcomes that stem from how systems are used, e.g., if users exploit their knowledge of an IS when using it (Burton-Jones and Straub 2006) or apply and adapt their knowledge of systems (Barki et al. 2007). Such studies often draw on task technology fit theory (Goodhue and Thompson 1995), arguing that benefits stem from using the right features for the task at hand (e.g., Ahearne et al. 2008, Devaraj and Kohli 2003).

Second, there are studies that examine concepts that are similar to effective use, but that deviate sufficiently in spirit or content to be considered differently. For instance, the concept of "faithful appropriation" (Chin et al. 1997, DeSanctis and Poole 1994) is closely related to effective use in that it relates to outcomes that stem from use, but it differs in that it traditionally focuses on whether or not the system is used in a manner consistent with the designer's intentions. Likewise, the concept of "infusion" emanated from organizational-level research examining how fully systems are integrated into organizations (Saga and Zmud 1994). Although defined in terms of performance outcomes, it differs from effective use in that it refers to how extensively the system is integrated into the work and how fully it is used. Studies operationalizing infusion at the individual level have similarly focused on this notion of extensive (Meister and Compeau 2002) or full and sophisticated use (Jain and Kanungo 2005), notions that are absent from our definition of effective use. There are also studies, both in our field and others, that have examined factors that may support effective use, but are not effective use per se, such as the competence of users or the usability of systems (Desjardins et al. 2005, Xie and Wolfram 2002).

Finally, there are studies that explicitly studied effective use, or synonyms thereof. It turns out that there are very few such studies. To be as sure of this as possible, we conducted two literature reviews. The first was a detailed but informal review based on our own knowledge of the IS literature. This yielded three articles published in a range of outlets (shown in the bottom rows of Table 1). Concurrently, we conducted a formal literature review of four fields that study IT use: human computer interaction and computer supported cooperative work (HCI/CSCW), health informatics, information science, and information systems. For each field, we carried out a structured review of all articles published in three highly regarded and diverse outlets during the years 1990-201