Despite the existence of successful online groups, the majority of newly created ones fail. They don’t survive, fail to attract enough digital users, or don’t meet goals their founders set for them. Herbert A. Simon Professor of Human-Computer Interaction Robert Kraut has dedicated much of his recent research to investigating why some groups succeed where others fail. With the help of a new NSF grant, he and collaborators at the University of Minnesota and the University of Michigan hope to create a computational theory of online communities. This computational theory will be both a scientific one to explain the variations in success of existing online communities, and an engineering one that can be applied to improve design choices.
Kraut, who began his career as a traditional social psychologist, has focused much of his research on the design and social impact of information technologies in small groups and larger communities. This new grant will extend this prior research by developing an overarching theory of community success, which can explain what attributes and processes determine success for online communities and help predict the effects of design interventions they are introduced into a community.
Prior research into online communities primarily consists of empirical studies and specialized theories to explain narrow outcomes, such as membership commitment or contribution. These empirical studies typically fail to account for the large number of design choices available to community managers and their potential influence on a variety of community outcomes. Although commercial firms routinely use A/B testing to make specific design choices in their communities, these tests tend to be atheoretical and don’t generalize beyond the setting in which they were conducted. There have been few attempts to build a comprehensive, evidence-based theory to explain how online communities’ attributes and processes interplay to determine their success.
The famous psychologist Kurt Lewin once wrote, "nothing is as practical as a good theory.” According to Kraut, this new project will build a practical theory of online communities, which will be useful both for understanding and improving them.
The project will ground its theory using empirical data from crowd lending teams, health support groups and peer support forums within STEM classes, and will test it by deploying theory-inspired interventions in those sites. Kraut envisions the verified model to become a tool to improve communities, provide guidance in design choices for future communities, and explain variations in success for existing communities.
Find more research about social computing from the HCII.