Forty-eight 9- to 11-year-old children were recruited from two public elementary schools in the Salt Lake City (Utah) school district. Twelve same-sex pairs of third and fourth graders (the maximum available) were formed from the collaborative school and twelve pairs were formed from the traditional school. At each school, an equal number of boys and girls participated.
Because the collaborative school regularly blended age groups in the same classroom and the traditional school did not, we avoided differences in familiarity by not pairing children from the same classroom. In addition, we did not pair children who reported having close bonds outside the classroom. The researcher was not familiar to the children of these grade levels in either school.