The Reliability and Validity of a Group-Based Measure of Adolescents' Friendship Closeness

Abstract

Objective: Six items were administered to a multiethnic sample of 435 middle school—age participants in a group-delivered safer sex intervention to determine their reliability and validity. Method: Exploratory analyses were followed by confirmatory factor analyses, and then correlations of scale scores with theoretically related variables were computed. Results: Findings favored a unidimensional scale with high internal consistency (with as few as two and as many as five items). Scale scores were generally related to other constructs as expected. Conclusions: The items show evidence of reliability and validity. In addition to the items' use in peer group research, the findings have implications for practice in terms of group leader training and facilitation, and the items' use as assessment tools.