Poster #69 - Longitudinal Associations between Perceived Discrimination, Peer Drinking, and Alcohol Use among Mexican American Adolescents

• In Event: 2-110 - Poster Session 07
  In Poster Session: PS 07 Section - Diversity, Equity & Social Justice

Integrative Statement

While empirical evidence strongly indicates that perceived discrimination (PD) can have negative psychological consequences (e.g., Schmitt et al., 2014), studies focusing on the link between PD and alcohol use (the majority of which are cross-sectional) have provided mixed results (Gilbert & Zemore, 2016). Minority stress models (e.g., Clark et al., 1999) conceptualize PD as a stressor, suggesting a direct link between PD and alcohol use. Others have conceptualized PD as a form of social rejection (e.g., Branscombe et al., 1999), which introduces the theoretically-based possibility that PD may be indirectly associated with alcohol use via increased deviant peer affiliations (e.g., Factor et al., 2013). The present study evaluates these perspectives by examining the longitudinal associations between PD, affiliations with peers who drink (peer drinking), and alcohol use among Mexican American adolescents.

Data were collected annually for 6 years from 646 Mexican Americans as they progressed from early to late adolescences (Wave 1: Mage = 12.39 years, SD = .60; 50.3% girls; 71.8% US-born). Response rates varied by wave but 89% of the adolescents remained in the study until the final wave. The adolescents completed a measure of perceived ethnic discrimination (Brittian et al., 2013) and were asked to report, in the past three-months, how many of their friends drank alcohol and how often they personally drank alcohol.

We estimated two autoregressive cross-lagged path models via Mplus Version 7.4 (Muthén & Muthén, 1998-2015). All of the paths were allowed to estimate freely in the first model (unconstrained model). Like-paths were constrained to be equivalent across the waves in the second model to evaluate whether the associations varied by age (constrained model). Due to infrequent use of higher response categories, the variables were rescored to be on 3-point scales and treated as ordinal in our analyses. Gender and nativity (i.e., US- vs. foreign-born) were included as covariates.

Both the unconstrained and constrained models provided a good fit to the data (see Table 1). Moreover, constraining the like-paths to be equivalent across the waves did not result in a significant drop in model fit, suggesting no age-based moderation. As shown in Table 2, PD was associated with subsequent increases in peer drinking, peer drinking was associated with subsequent increases in alcohol use, and alcohol use was associated with subsequent increases in peer drinking; PD was not significantly associated (directly) with subsequent alcohol use. Finally, alcohol use was associated with subsequent decreases, while peer drinking was associated with subsequent increases, in PD.

PD was associated with alcohol use but only indirectly via increased peer drinking affiliation among Mexican American adolescents, consistent with the conceptualization of discrimination as a form of rejection (Branscombe et al., 1999). PD may have different meanings for adolescents, who likely have a rudimentary understanding of the pervasiveness and impact of discrimination within society, and adults, who likely have a deeper understanding of discrimination and its personal and societal consequences. We discuss the implications of these differences for future research on PD and alcohol use.

Authors

• Brian E. Armenta, University of Missouri
  Presenting Author

• Kenneth J. Sher, University of Missouri
  Non-Presenting Author

• Miriam Martinez, University of Missouri
  Non-Presenting Author

• Olivia Emile Atherton, University of California, Davis
  Non-Presenting Author

• Richard W. Robins, University of California, Davis
  Non-Presenting Author