As research on adolescent crowds has increased over the past several decades, researchers appear to be confident in their claims of the consequences of crowd membership, even suggesting targeted interventions. This review of the various methods used to identify adolescents' crowd membership suggests that this confidence may be misplaced. There are diverse methodologies used in this research area that examine different samples of adolescents belonging to each crowd. Social-type rating methods, self-identification methods, grouping by adolescent behaviors or characteristics, and ethnographic or other qualitative methods should be accompanied by greater specificity in terminology to alert researchers to the various phenomena being studied (i.e., "reputational crowd," "interactional crowd," "behavioral crowd," "affiliation crowd"). Additionally, studies comparing the various self-identification approaches and peer ratings are needed, along with reliability studies of peer ratings. More attention to specific methodology to determine crowd membership and its stability will aid the design of theoretical models of adolescent crowds and contribute to developmental outcome research.