Methods: The Youthnet study of runaway homeless youth ages 13 to 28 in LA, was used for our analysis. The total sample (N=936) was collected in four waves from October 2011 through June 2013. Using an event-based approach, youth were recruited at two drop-in centers in different LA neighborhoods. Data collection comprised a self-administered questionnaire and a social mapping interview conducted by trained research assistants.
Age, gender, sexual orientation, race/ethnicity, sexual abuse before homelessness, and physical abuse before homelessness, were control variables in our model. Family network characteristics of homeless youth were stratified by family relationship type (parent, sibling, cousin) and emotional or instrumental support obtained from each relative type. Emotionally supportive family was defined as relatives youth could turn to when they need help or advice and whom they could confide in. Instrumentally supportive family was measured with items asking about individuals youth could talk to about school and/or work and whom they could borrow $100 from. A multivariate ordinary least squares regression was used to test the supportive family network characteristics of homeless youth based on racial and ethnic backgrounds.
Results: The mean age of youth in our sample was 21.35 years, 72% male, 39% non-Hispanic White, 24.7% Black/African-American, and 14% Latino; 24.6% identified as bisexual, homosexual, or questioning. Compared to the networks of non-Hispanic White homeless youth, being Latino was associated with having emotionally supportive siblings (β=0.035, t=1.99, p=.047) and emotionally supportive cousins (β=0.026, t=2.70, p<.01) in their network. Conversely, being Black/African-American, compared to non-Hispanic Whites, was associated with fewer emotionally (β=-0.061, t=-4.29, p<.001) and instrumentally (β=-0.056, t=-4.04, <.001) supportive parents in their social network.
Discussion: Results indicate Latino youth maintain emotionally supportive connections with siblings and cousins in their social networks, relative to youth from other backgrounds. The Risk Amplification and Abatement Model (RAAM) suggests these supportive home-based ties can promote positive behaviors among homeless youth. Practice settings must support positive emotional connections to extended family networks, particularly with siblings and cousins of Latino homeless youth. Future research should explore more carefully racial and ethnic differences in how supportive family-based network ties may impact depressive symptoms, substance use, and other risk behaviors.