Using Snowball-Based Methods in Hidden Populations to Generate a Randomized Community Sample of Gang-Affiliated Adolescents

Abstract

Snowball-based methods frequently have been used when working with hidden populations whose members are difficult to identify and locate. By the very nature of the processes of these methods, such samples usually are not random or representative, thereby, resulting in selection bias. This study provides a framework on how to generate a randomized community-based sample of Mexican American adolescent females involved with gangs whose understanding merits further examination given their apparent increase in number and degree. Research findings from a more robust sample can accentuate appropriate policy formation as well as prevention and intervention programs aimed at improving the lives of high-risk youth.