Robert E. Kraut

Videos

Is Social Media Good or Bad for Well-being?

Moira Burke and Robert Kraut

A video produced by Facebook explaining research linking social media use and psychological well-being. It is linked to a Dec, 2017 Facebook blog post Hard Questions: Is Spending Time on Social Media Bad for Us?

Online Communication and Psychological Well-being: A Research Journey

Robert Kraut

Presentation for the 2016 SigCHI Lifetime Achievement in Research Award. ACM Conference on Human Factors in Computing Systems, May, 2016.

Since the dawn of the Internet era, researchers, policy-makers and the general public have questioned how new technology is influencing our psychological well-being. This talk will review research I’ve done with students and collaborators over the past 18 years to cast light on this question. Our earliest work in the 1990s indicated that more Internet use, independent of type, predicted increases in depression and declines in other measures of well-being. However, later research that differentiated communication with stronger ties from communication with weaker ties and differentiated types of communication shows that whom one communicates and the nature of the communication are the important factors, not Internet use per se. In Facebook, for example, receiving targeted, composed communication from strong ties predicts improvements in well-being, while viewing wide-audience broadcasts, receiving one-click feedback, and receiving composed communication from weak ties does not. In online health support groups, it is messages exchanged in public that predicts improvements in well-being, while private message exchanges predict declines.

Online Social Support: Advances in Measuring Support and Understanding Its Effects

Robert Kraut

Talk presented at Georgia Tech GVU Center Distingushed Lecture. Atlanta, GA. February 4, 2016.

Many people with serious diseases use online health groups to exchange social support. For these groups to be effective, members must both seek support and provide it, and the support they exchange must have some benefits. For the groups to be sustained, some members must continue to participate. This talk describes analysis of behavior in two large, online cancer communities examining how people get support and the impact of support on members’ satisfaction, commitment and psychological well-being. We use machine learning techniques to automate content analysis of millions of messages, measuring the extent to which messages contain such support-related actions as seeking and providing informational and emotional support, asking questions, expressing empathy and self-disclosing personal thoughts and feeling. These variables are used in longitudinal regression analyses to predict the type of support people receive, their satisfaction with the exchanges and their commitment to the group. The research illustrates differences in the how informational and emotional support are produced and their effects on satisfaction and commitment.

Social Design for Collective Intelligence

Robert Kraut

Invited Plenary talk presented at the Conference on Collective Intelligence, Cambridge, MA April 19, 2012

Collective intelligence is more likely to emerge online if the social environment is designed to overcome challenges that are common in many groups, communities and organizations. Many online groups that exhibit collective intelligence, like the Wikipedia community, depend upon volunteers with low commitment and high turnover who work in an environment without the trappings of conventional employment organizations, such as hierarchical management, employment contracts or pay for performance. Under these conditions, Wikipedia and similar communities face special difficulties recruiting and socializing newcomers and developing policies and processes that encourage their commitment and contribution.

This talk reviews a program of research to understand what leads to commitment and contribution in online communities and how to better design them for these purposes. It reports on both archival research examining natural variations in group structure and interaction that predict greater commitment and contribution and experimental research deploying interventions to improve these outcomes.

The talk illustrates an evidence-based perspective on social design, in which theory and empirical evidence from the social sciences are used to structure online interaction to solve problems endemic to online groups.

Designing Online Communities from Theory

Robert Kraut

Invited talk presented at Stanford University, Human-Computer Interaction Seminar (Seminar on People, Computers, and Design), Stanford, CA

Online communities are the fastest-growing portion of the Internet and provide members with information, social support, and entertainment. While a minority, such as Wikipedia, MySpace, Facebook and the Apache Server project are highly successful, many others fail. To be successful, online communities must overcome challenges common in almost all groups, organizations and voluntary associations. They must handle the start-up paradox, when early in their lifecycle they have few members to generate content and little content to attract members. Throughout their lifecycle, they must recruit and socialize newcomers, encourage commitment and contribution from members, solve problems of coordination and encourage appropriate behavior among members and interlopers alike.

The social sciences can tell us a lot about how to create thriving online communities. Social science theories can inform choices about how to get a community started, integrate newcomers, encourage commitment, regulate behavior when there are conflicts, motivate contributions, and coordinate those contributions to maximize benefits for the community.

This talk focuses on ways to build members’ commitment to online communities, based on theories of social identity and interpersonal bonds. It provides an overview of the relevant theory, describes results of a 6-month field experiment in which an existing site was redesigned based on principles derived from social identity and interpersonal-bond theories, and describes the results of an agent-based model that examines how different approaches to moderating the content in a group influence social identity and interpersonal bonds. Finally, it illustrates a number of design claims that translate the abstract principles from the social sciences into concrete design choices.