A social network-based teams intervention: pilot study impact on bonding to school

Bonding to school, the degree to which students feel accepted and have an attachment to the school, has the potential to affect whether they continue their education or drop out. I evaluated a pilot intervention in which students in the treatment group are assigned to in-class work teams. Students completed social network surveys in which they identified with whom they had recent interactions. The student sample consisted of 1,570 students, 787 males and 774 females. Of these, 1,024 attended schools assigned to receive treatment, and 546 attended schools assigned to the control condition. The survey also asked questions about values and interests. Teams were formed using high-centrality students (those who had the most interaction nominations) as seeds. Students with whom they were acquainted and who share values and interests were joined. Students who scored poorly on a measure of school bonding were sequestered until the last step of team formation. Teachers were encouraged to use teams as the basis for seating and for group tasks during instruction. Students in control schools were not assigned to teams. Overall, there was a decline in bonding from pretest to posttest, which is typical as students grow older. Students with the lowest pretest bonding score increased their bonding from pretest to posttest. Multivariate repeated analysis of variance assessed pretest–posttest changes to separately examine students in four pretest bonding subgroups. This analysis revealed that assignment to teams impacted bonding of only those who had positive bonding scores to begin with. Compared to control students in this quartile, those who were assigned to social network-based teams had smaller declines in bonding. Results of this pilot study suggest that there may be some gain in students’ levels of bonding to school that may be facilitated by assigning them to groups that have initially low levels of affiliation in which there are common interests and values. The logistics of using online data collection in team creation is methodologically challenging. Challenges to maintaining team membership are discussed.