Social engagement drivers

This paper studies how online communities can foster social engagement. The study is carried out in a site for the neighbours of a suburban town, built in a Domino/Lotus environment. Three different aspects are examined: effects of several design element that were intended to increase interaction, users' profiles and subjects of the conversations.
Design
- while the site is structured in "rooms" for different topics, the "frontpage" (main room) concentrates activity and makes new material more visible
- others systems allow to track users' activities
- notification services
Members selection
Method: a survey has been sent to users in order to know their background and experience, general attitudes about the Internet and the site. The survey included 50 questions and covered different topics: attitudes about technology (comfort and skills, security, usage of synchronous and asynchronous skills), time use (hours spent working, unpaid work, leisure...). Several regression analysis have been carried out using the survey responses as independent variables and the logs of the site as dependent variables.

The independent variables used in the regression models were based on estimated factors scores, which where computed for each respondent using factors solutions.
The measure for participation (dependent variable) is the number of sessions.

While the method of research can be interesting, the outcomes are not surprising at all (we have to note that this text has been written in the turn of the century): skilled users are more likely to participate, as would do individual who spend a great deal of their non-work time socialising and in non-work Internet activities.

Content
The paper also study the subject of the longer threads of messages

Stimulating social engagement in a community network
by: David R Millen, John F Patterson