Content of current issue

After two consecutive special issues in March and May, the first on IS in mergers and acquisitions and the second on cross-cultural IS research, the current issue seems to be a special issue (but actually, it is not). The five papers in this issue are all about using, continuing to use or discontinuing the use of IT, particularly IT that has a strong social element. Each paper brings something new to the party, but by putting all five papers together, this issue is certainly special in offering a richer, more relevant picture of IT use than the models we have used in the past.

The first paper by Anol Bhattacherjee and Chieh-Peng Lin (A unified model of IT continuance: three complementary perspectives and crossover effects) provides an excellent start to the volume by sorting out what may be seen as the three main reasons for continuing to use IT: reasoned action governed by perceived usefulness and subjective norms, experiential response reflected in satisfaction and habitual response. Their evidence comes from a longitudinal survey conducted within a large insurance company in Taiwan.

Next, Tim Barnett, Allison Pearson, Rodney Pearson and Franz Kellermanns (Five-factor model personality traits as predictors of perceived and actual usage of technology) see personality as a significant factor in shaping the intention to use IT. Indeed, the personality factor enhances the predictive power of continuance models. Evidence for this comes from a longitudinal study of a web-based classroom system, and this is the only study in this issue that measures actual use, as well as questionnaires and self-reports.

The next two studies extend general models of using IT to capitalize on the specific way we use IT for social interaction. Tao Hu, William Kettinger and Robin Poston (The effect of online social value on satisfaction and continued use of social media) define the user’s online social value of interacting with social media to explain the user’s intentions to continue using the IT and gaining satisfaction from gaining social value. Online social value results from utilitarian and hedonic benefits vs the costs associated with risk and effort. Meanwhile, Chao-Min Chiu and Hsin-Yi Huang (Examining the antecedents of user gratification and its effects on individuals’ social network services usage: the moderating role of habit) define gratification, rather than satisfaction, to explain the user’s intentions to continue using social media. They go beyond simply using the reasoned action and habitual use model to show the reasons for using social media. In this way, the authors are able to point to understanding, acting and playing as the three goals users have to satisfy by continuing to use social media.

We round off the party with Ofir Turel (Quitting the use of a habituated hedonic information system: a theoretical model and empirical examination of Facebook users). He argues that users have different reasons for continuing and discontinuing their use of IT, and that these considerations will affect differently what the users wish to do. It is a study of popular hedonic applications for personal use such as Facebook. The evidence in this paper is based on a study of Facebook users enrolled in a business class at a large North American University.

Of course, not all studies published in EJIS have to make a practical contribution. However, it is always refreshing to see papers with messages that may change behavior and practice. The unified model sends a strong message to managers and developers to invest in technology and training that will have a positive influence on habits and experience. The social media research on what comes before the sense of gratification shows how social interaction determines the components of comprehensive support in social network services. Studying the subject of discontinuance will warn providers that making habitual behavior stronger will not always bring advantages. At the same time, helping the user to decide when, if at all, to quit might be the right way to go. These implications are a sign of relevance that EJIS cherishes.

Context for future submissions

Most papers in this issue use only questionnaires to measure attitudes, perceptions and self-reported behavior. Although questionnaires no doubt have many advantages for collecting data in comparison to, say, direct observation, they run the risk of making us less able to place our studies in a particular context. Questionnaires that assess constructs necessarily rely on abstraction, approximation and translation. And although these studies will need to be de-contextualized at some stage of the study, there is usually room in a questionnaire-based paper to consider the context, and even the idiosyncrasies of the particular study, beyond the essential practice of specifying and controlling possible co-factors. I believe that examining the use of IT in its context should be encouraged. The remainder of this editorial is a call to get a greater sense of context in questionnaire-based papers, as well as to use other research genres that lend themselves more easily to contextualization.

So what does ‘contextualizing’; mean? In essence, it is about considering and examining the context when planning and doing research in order to arrive at a better understanding of the phenomena under question. We do a pretty good job of telling the reader about the sample and the setting in the methods section and later in the limitations section. And we do it so that others know to which contexts they can or cannot apply parts of the study. Not only this, but readers expect discussions of generalization problems in the limitations section that go beyond the mere facts about the setting and sample described in the method section. In fact, discussions are there to give the implications of contextual data, even if these are obvious to some authors. For instance, blurring the boundary between work and pleasure does not necessarily mean that we can assume that students using Facebook at home will be using it in the same way at work. Authors need to make the argument why they think so.

It is more difficult to understand the indirect effect of context on what is not addressed by the researcher, particularly those aspects that the readers may assume mistakenly allow them to project conclusions from the study to their own studies in other contexts. For instance, different organizations place different restrictions on how to use IT. So it is vital to mention the implications of these restrictions on the variability of constructs related to use, because this will help to generalize or compare across organizations. At the same time, restrictions change with time. Organizational norms and regulations that were prevalent when social media was first introduced may well change dramatically once the full implications of privacy come to be reflected in new protective rules and practices. What is more, if these changes become part of a theory of how social media affects communication in organizations, it may help to observe directly how the new European regulations on privacy affect the practices of using Facebook in European organizations, rather than to assess perceived benefits after the fact. I would hazard a guess that a more contextualized approach to studying use may be more suitable to see what interactive forces shape IT use in organizations, which are often hidden in the mainly uni-directional models tested by the less contextualized research methods.

The two papers in this issue given over to the continued use of social media show just how this particular type of IT might help untapped theories to be applied, and how dedicated constructs might be used that are not part of models for using general IT. My hunch is that the more we study the emotions that are tied to social interaction, such as gratification, the greater the worry will be about the interactive effects of the context. Before recommending design and action, we should make sure that the limitation section discusses the implications of a particular context. Every new IT changes the context and the user’s interaction with the new IT in a particular context. It is not only about what and how we measure in the new context (as shown in these papers) but how the context might affect the measures. For instance, it is useful to specify moments in history that affect how habit affects gratification when social media is used.

Research that only uses questionnaires about attitudes and self-reports obviously makes the physical and organizational setting in which attitudes are formed more abstract. It also stops short at observing action, which always happens in context. More generally, we have argued at various times for IS research to represent subjective and social worlds, as well as the technical and observable world with appropriate tools for the different worlds. An excellent example is the case for mixed methods argued by Agerfalk in his 2013 editorial. The need to describe the context is all the greater in the internationally and culturally diverse research we cherish in EJIS. Ideally a mix of methods will bring to the fore several important aspects to discuss contextual aspects beyond the gender and age of subjects, whom are students or mechanical Turk subjects.

Direct observation certainly seems to make contextualization simpler, and almost ‘natural’. Indeed, the EJIS has encouraged research that represents and examines in detail the context of developing, using and benefiting from IT. Actually, EJIS boasts a diverse set of methodologies (see previous editorial by Rowe, 2012). Two great examples of this are case studies that report on actual behaviors and ethnographical studies based on detailed behaviors in specific situations. Just as mentioning the context unnecessarily disturbs everyday communication, papers that are bogged down with practical details also harm the way that research is communicated. As editors, we should only ask for details about the context when they are needed and are useful to the reader. A simple example is when reporting how an organization uses social media. This may well help readers to compare results of similar studies at organizations with different norms. Prime, complex examples are studies that explore the effect of national culture on the decision to stop using IT, and its effect on the social hierarchy.

I certainly do not want to prescribe how contextualization should be practiced in each of the wide set of genres that the EJIS promotes. But I hope and expect authors and editors who are knowledgeable about specific genres to look for ways to set out the context more effectively in their own area of expertise. In the future we will learn more from our published work. And I don’t want to suggest that every publication in EJIS sets out a context, but I do encourage authors of future submissions to give a little more thought to contextualization, and to make the most of the context of their study.