The 15-year history of collaboration on Wikipedia offers insight into how peer production communities create knowledge. In this research, we combine disparate content and collaboration approaches through a social network analysis approach known as an affiliation network. It captures both how knowledge is transferred in a peer production network and also the underlying skills possessed by its contributors in a single methodological approach. We test this approach on the Wikipedia articles dedicated to medical information developed in a subcommunity known as a WikiProject. Overall, we find that the position of an article in the affiliation network is associated with the quality of the article. We further investigate information quality through additional qualitative and quantitative approaches including expert coders using medical students, crowdsourcing using Amazon Mechanical Turk, and visualization using network graphs. A review by fourth-year medical students indicates that the Wikipedia quality rating is a reliable measure of information quality. Amazon Mechanical Turk ratings, however, are a less reliable measure of information quality, reflecting observable content characteristics such as article length and the number of references.
In recent years, we have witnessed the rapid proliferation and widespread adoption of a new class of information technologies, commonly known as social media. Researchers often rely on social network analysis (SNA) when attempting to understand these technologies, often without considering how the novel capabilities of social media platforms might affect the underlying theories of SNA, which were developed primarily through studies of offline social networks. This article outlines several key differences between traditional offline social networks and online social media networks by juxtaposing an established typology of social network research with a well-regarded definition of social media platforms that articulates four key features. The results show that at four major points of intersection, social media has considerable theoretical implications for SNA. In exploring these points of intersection, this study outlines a series of theoretically distinct research questions for SNA in social media contexts. These points of intersection offer considerable opportunities for researchers to investigate the theoretical implications introduced by social media and lay the groundwork for a robust social media agenda potentially spanning multiple disciplines.
The information systems (IS) literature has focused considerable research on IS resistance, particularly in the health-care industry. Most of this attention has focused on the impact of IS resistance on systems' initial implementation, but little research has investigated whether and how post-adoption resistance affects performance. We focus on a particular type of post-adoption resistance, which we call IS avoidance, to identify situations in which individuals avoid working with adopted IS despite the need and opportunity to do so. We examine the effects of IS avoidance on patient care delivered by health-care groups across three levels of analysis: the individual level, the shared group level, and the configural group level. We find that IS avoidance is significantly and negatively related to patient care only at the configural group level, which suggests that patient care is not degraded by the number of doctors and/or nurses in a group avoiding a system, but rather by their locations in the group's workflow network configuration. We use qualitative data collected over 16 months at the research site to help explain these results. Implications for theory and practice are discussed.
Firms increasingly turn to online communities to create valuable information. These communities are empowered by new information technology-enabled collaborative tools, tools such as blogs, wikis, and social networks. Collaboration on these platforms is characterized by considerable membership turnover, which could have significant effects on collaborative outcomes. We hypothesize that membership retention relates in a curvilinear fashion to effective collaboration: positively up to a threshold and negatively thereafter. The longitudinal history of 2,065 featured articles on Wikipedia offers support for this hypotheses: Contributions from a mixture of new and experienced participants both increases the likelihood that an article will be promoted to featured article status and decreases the risk it will be demoted after having been promoted. These findings imply that, contrary to many of the assumptions in previous research, participant retention does not have a strictly positive effect on emerging collaborative environments. Further analysis of our data provides empirical evidence that knowledge creation and knowledge retention are actually distinct phases of community-based peer production, and that communities may on average experience more turnover than ideal during the knowledge retention phase.
Virtually all of the extensive previous research investigating the effect of information systems proficiency on performance has been conducted at the individual level. Little research has investigated the relationship between IS proficiency and performance at the group level. In this paper, we argue that IS proficiency at the group level may be more than the simple sum or average of the IS proficiency of individual group members.Rather, effective group-level IS proficiency may also be a function of how a group's IS proficiency is distributed across its members. Relying on concepts associated with social network analysis (SNA), we introduce the concept of centrality-IS proficiency alignment. We argue that groups will perform better if their more proficient members are highly central in the group's communication and workflows network. Data from 468 employees in 32 workgroups show that centrality-IS proficiency alignment is significantly and positively related to performance across multiple systems examined individually and with the portfolio of systems examined as a whole. This approach effectively integrates the structural and resource perspectives of SNA,providing a roadmap so that others may follow a similar approach to address broader questions of group-level user-system interactions in the IS literature and more general questions of central resource alignment in the broader organizational literature.
This paper argues that Web 2.0 tools, specifically wikis, have begun to influence business and knowledge sharing practices in many organizations. Information Systems researchers have spent considerable time exploring the impact and implications of these tools in organizations, but those same researchers have not spent sufficient time considering whether and how these new technologies may provide opportunities for us to reform our core practices of research, review, and teaching. To this end, this paper calls for the IS discipline to engage in two actions related to wikis and other Web 2.0 tools. First, the IS discipline ought to engage in critical reflection about how wikis and other Web 2.0 tools could allow us to conduct our core processes differently. Our existing practices were formulated during an era of paper-based exchange; wikis and other Web 2.0 tools may enable processes that could be substantively better. Nevertheless, users can appropriate information technology tools in unexpected ways, and even when tools are appropriated as expected there can be unintended negative consequences. Any potential changes to our core processes should, therefore, be considered critically and carefully, leading to our second recommended action. We advocate and describe a series of controlled experiments that will help assess the impact of these technologies on our core processes and the associated changes that would be necessary to use them. We argue that these experiments can provide needed information regarding Web 2.0 tools and related practice changes that could help the discipline better assess whether or not new practices would be superior to existing ones and under which circumstances.
Information systems (IS) researchers have typically examined the user-system relationship as an isolated dyad between a single, independent user and an individual, freestanding information system. We argue that this conceptualization does not adequately represent most organizations today, in which multiple users interact with multiple information systems within a group. Relying heavily on the theory and methods behind social network analysis, we introduce the concept of multimodal networks to assess both users and information systems as equivalent nodes in a single social network. This perspective allows us to examine the influence of information systems on organizational outcomes as a function of all of the user-system and interpersonal interactions in a group. We explore two different possible mechanisms for this influence: (1) direct user-system interactions by aggregating the strength of all the dyadic user-system interactions in a group, and (2) indirect user-system interactions by assessing the centrality of the information systems within the social network. We survey approximately 600 individuals in 40 healthcare groups to test whether either or both of these mechanisms are associated with two types of organizational performance outcomes-efficiency and quality of care. We find that the centrality of the information systems within the network is significantly and positively associated with both efficiency and quality outcomes, but that the average strength of the user-system interactions is not. Implications are that managers and researchers should examine the wider multimodal network of multiple users and multiple systems when assessing the role of IS in organizations in relation to organizational performance outcomes.