Capturing the friendship context with a collective property: Friendship group engagement vs. disaffection Carrie J. Furrer * NPC Research, 4830 SW Macadam Ave., Ste. 530, Portland, OR 97239, United States Keywords: Friendship groups Collective properties Group motivation Peer context Group-level phenomenon abstract The purpose of this study was to introduce a motivational property of the group, friendship group engagement vs. disaffection, and to examine different composition models for combining individual reports into an indicator of a group-level property. Data were collected from 312 ninth grade students and their teachers. Self-reported friendship group (FG) engagement and disaffection were related to self-reported classroom engagement, aggression, and peer problem behavior, but not to teacher-reported classroom engage- ment. Group-reported FG engagement and disaffection were related to self-reported FG engagement and disaffection, classroom engagement, aggression, and peer problem behavior, but these relationships were moderated by within-group agreement on the group property. Findings underscore the importance of a priori conceptualization of group-level properties that are theoretically linked to adolescent adjustment, and the need to consider composition models when aggregating individual reports to create profiles of the group. Implications for future work on capturing properties of friendship groups are discussed. Ó 2010 The Association for Professionals in Services for Adolescents. Published by Elsevier Ltd. All rights reserved. Friends and peer groups are a fundamental developmental context for adolescents. As children move into adolescence, their friends are central in their daily lives – their relationships become increasingly intimate and supportive, they spend more time with non-family members, and they develop romantic relationships (Collins & Repinski, 1996; Furman & Winkles, 2010; Hartup,1996; Ladd, 2005; Parker & Asher,1987; Savin-Williams & Berndt, 1990; Sullivan,1953; Youniss & Haynie,1992). Not only are adolescents influenced by their multiple friendships (e.g., Berndt & Keefe, 1996; Farmer, 2000; Hartup, 2005; Kindermann, 2003; Newcomb & Bagwell, 1995), but they also participate in shaping these relationships and co-create the climate or context that emerges from ongoing relationships with friends (e.g., Bronfenbrenner & Morris, 1998; Bukowski, Velasquez, & Brendgen, 2008). The investigation of qualities of the friendship group (or peer group or social network) has taken two general directions. One line of inquiry centers on asking individuals to report on experiences with friends. Constructs such as friendship quality, social support, belonging, relatedness, and social competence are examples of self-referent indicators of one’s friendship context (i.e., typically self-reported questions that refer to the individual, “I feel.” or “I have.”). Research has generally shown that a more positive social context, as inferred from these and other similar constructs, contributes to various indi- cators of adolescent adjustment (see Gifford-Smith & Brownell, 2003 for a review). Adolescents with high quality, supportive friendships (e.g., experienced as loyal, affectionate, and caring) tend to have higher self-esteem, exhibit more prosocial * Research, Evaluation and Assessment, Portland Public Schools, 501 N. Dixon, Portland, OR 97227-1804, United States. Tel.: þ1 503 916 6342; fax: þ1 503 916 3106. E-mail address: cfurrer@pps.k12.or.us. Contents lists available at ScienceDirect Journal of Adolescence journal homepage: www.elsevier.com/locate/jado 0140-1971/$ – see front matter Ó 2010 The Association for Professionals in Services for Adolescents. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.adolescence.2010.07.003 Journal of Adolescence 33 (2010) 853–867