VOSS: building social attachment
in virtual groups

Project Overview

Science and engineering work is increasingly accomplished by teams of interdependent people performing at different times and places within and across organizations. Prior research shows that it is particularly difficult for members of these “virtual” teams to become attached or committed to their group. Lack of commitment in turn can lead to conflict, lack of cooperation, decreased information sharing and contribution, higher rates of turnover and poor team performance The goal of the proposed work is to better understand how to build attachment to a group in virtual settings.

Group members can be attached to a group in different ways:

  1. through bond-based attachment, in which the connection to the group is through interpersonal relationships with other group members, or
  2. through identity-based attachment, in which group members incorporate the group mission, theme or defining characteristics into their sense of who they are.

These types of attachment have distinct antecedents. We propose that one can design virtual groups to enhance members’ bond-based or identity-based attachment or both. Dyadic relationships between members of a group can be enhanced by recruiting for similarity and by providing shared experiences and opportunities for repeated interaction or mutual self-disclosure. A shared social identity can be developed by recruiting people from a single social category, by defining group boundaries with clear out-groups and by providing common goals and task interdependence. Although previous research suggests that these different ways of building attachment to a group are distinct, it is not clear to what extent they are complementary or competing. Nor is it clear how to use these ideas for purposes of social engineering – to increase commitment to group – especially in virtual environments.

We will examine these and similar ideas by conducting a series of online experiments, creating numerous game groups in Facebook. These experiments will vary features of the groups that theory predicts will promote bond-based and identity-based attachment. For example, we can create groups with and without opportunities for pair wise communication, collective goals, demographic similarity among members or competition from other groups. We will track each group for over a month from its creation. Outcomes include turnover, contributions, member-to-member support, task performance within the group, friendship ties, communication outside the group, and self-reported attachment to the group and its members.

These parametric, experimental studies will be complemented with field experiments conducted in citizen-scientist communities like “eBird.org”. eBird is an online community, hosted by the Cornell Laboratory of Ornithology, in which birder watchers report on the geographic and temporal distribution of bird species. In a series of field experiments in citizen-scientist communities such as eBird, we will introduce features that the Facebook experiments showed successfully promoted attachment. The focus of these field experiments will be to determine how to combine and hone a set of community-design features to increase attachment and contribution.

If you have any questions, feel free to send an email to: tetris-for-research@mailman.srv.cs.cmu.edu.

national science foundation logo this work was made possible through nsf award #0943159.