Difference between revisions of "Helping hands: design for member-maintained online communities"

From WikiLit
Jump to: navigation, search
(changed Collected data time dimension)
m (Text replace - "|collected_datatype=" to "|data_source=")
Line 64: Line 64:
artifacts that might be produced at each stage. We discuss the stages below.
artifacts that might be produced at each stage. We discuss the stages below.
|collected_datatype=Experiment responses, Wikipedia pages
|data_source=Experiment responses, Wikipedia pages

Latest revision as of 20:28, January 30, 2014

Publication (help)
Helping hands: design for member-maintained online communities
Authors: Daniel Regis Cosley [edit item]
Citation: University of Minnesota  : . 2006. United States, Minnesota.
Publication type: Thesis
Peer-reviewed: Yes
DOI: Define doi.
Google Scholar cites: Citations
Link(s): Paper link
Added by Wikilit team: Added on initial load
Article: Google Scholar BASE PubMed
Other scholarly wikis: AcaWiki Brede Wiki WikiPapers
Web search: Bing Google Yahoo!Google PDF
Format: BibTeX
Helping hands: design for member-maintained online communities is a publication by Daniel Regis Cosley.

[edit] Abstract

Online communities provide millions of people every day with information, companionship, support, and fun. These communities need regular maintenance to function. Tasks such as welcoming new members, reviewing contributions, and building community-specific databases typically fall to a few dedicated members. Concentrating responsibility in the hands of a few valuable leaders makes communities vulnerable to leaders' leaving and limits communities' ability to grow and provide value. We study the design of member-maintained online communities, systems where many members help perform upkeep. A key design challenge is motivating members to contribute toward maintenance. Social science theories help to explain why people contribute to groups. We use these theories to design two general mechanisms for increasing people's motivation to contribute. The collective effort model from social psychology suggests people are more likely to contribute to a group if they believe their contributions matter. Editorial review can foster this belief by promoting good content and suppressing bad content. We build review systems that involve the whole community, where review is performed by peers, experts, or no one. Peer review performs about as well as expert review in both motivating contributions and providing effective review, but no review does very poorly. We also explore whether contributions must be reviewed before being made available to the community. Mathematical models suggest that making contributions available right away increases value more quickly, and does just as well in the long run, as requiring prior review. These models can inform the design of review systems. Public goods theory from economics suggests people will contribute more to group resources if the cost of contributing drops. We use intelligent task routing---matching people with tasks they are likely to do---to reduce contribution costs. We develop a number of generally useful task routing algorithms. Experiments in a movie database and in Wikipedia show these algorithms are very effective at increasing people's motivation to contribute. By using theory to support our designs, testing them in multiple domains, and distilling our results into usable artifacts such as guidelines, models, and algorithms, we hope to help designers build better systems and better communities.

[edit] Research questions

"In this thesis, we explore the potential, problems, and design of member-maintained communities, online communities where many members participate in the group’s upkeep. We investigate how to design communities to increase members’ willingness to participate and lead to better community outcomes."

Research details

Topics: Technical infrastructure, Contributor motivation, Other antecedents of participation, Contributor engagement [edit item]
Domains: Information systems, Sociology [edit item]
Theory type: Design and action [edit item]
Wikipedia coverage: Sample data [edit item]
Theories: "Public goods theory from economics suggests people will contribute more to group resources

if the cost of contributing drops. We use intelligent task routing—matching people with tasks they are likely to do—to reduce contribution costs.

We believe that design for contribution—using social science theory to inform system design, and vice versa—is one way to bridge this gap in online communities. Economists and computational agent builders use mechanism design to induce agents to behave in ways that accomplish social goals of truthfulness, fairness, or efficiency (Varian 1995). Similarly, online community designers might build systems that induce members to contribute by exploiting theories about why people participate in groups.

A first question is whether we should believe that theories about group behavior from social psychology, sociology, economics, and related disciplines will work in online settings. These theories have been developed over years of watching people interact offline, but social presence theory suggests that online communication is less rich than face-to-face and may lead to different outcomes (Short, Williams, and Christie 1976).

A general problem with applying social science theories to system design is that the theories are hard to apply directly, another manifestation of Ackerman’s’ socio-technical gap. Further, it is hard to know the scope of a theory, which contexts it works in and which it does not. This is an especially difficult problem in the study of groups (McGrath, Arrow, and Berdahl 2000). The CommunityLab project, a collaboration between the University of Minnesota, the University of Michigan, and Carnegie Mellon University, aims to bridge the gap by deploying these theories toward multiple designs and in multiple contexts....Third, we draw on the similarities between our experiments and those described above to outline a general methodology for using theory in design.

Based on theory, we did not have strong beliefs about the relative strength of peer and expert review, though we know that people often perceive expert reviewers to be more valuable than peers.

There is not much theory about how to compute motivation, and the collective effort model is short on quantitative predictions. Theories of critical mass (Marwell and Oliver 1993) and the diffusion of innovations (Rogers 1995) might have more quantitative insight to contribute, along with the body of work in experimental economics.

Both the collective effort model and the theory of public goods suggest that reducing costs should motivate people to contribute more.

Our contribution model is a usable theory for designers of editorial review mechanisms. A number of other theories, from activity theory to communities of practice, may be useful in designing systems that motivate contributors. Based on our experiments and other work from the from the CommunityLab group, we present a methodology for using theory to drive designs. Our goal is to help designers find effective solutions to their specific problems while also generating design artifacts that can be used by other designers. Figure 11.1 shows the steps of the methodology and the design artifacts that might be produced at each stage. We discuss the stages below." [edit item]

Research design: Experiment [edit item]
Data source: Experiment responses, Wikipedia pages [edit item]
Collected data time dimension: Longitudinal [edit item]
Unit of analysis: Article [edit item]
Wikipedia data extraction: Live Wikipedia [edit item]
Wikipedia page type: Article [edit item]
Wikipedia language: English [edit item]

[edit] Conclusion

"The Internet connects computers, but more fundamentally, it connects people. Millions of people per day derive social benefits from online communities. Many fewer pay the costs of maintaining these communities. In this thesis, we studied ways to spread these costs across many people in the hope of building more valuable communities. By applying theories of contribution to real problems, we were able to design policy mechanisms for reviewing contributions and algorithms for matching people with tasks that increased the number of people who contribute and the amount of work they do."

[edit] Comments

""Asking people to do work that is tailored to their interests leads to large increases in both the number of people who contribute and the amount of work they do.""

Further notes[edit]