Leroux, Ximena (2020). Cigarette Smoking Among Hispanic and Non-Hispanic, White Adolescents: Segmented Assimilation and the Social Causes of Delinquency.
Research on smoking among Hispanic persons in the United States presents a puzzle. In general, lower socioeconomic status correlates with higher smoking risk. However, Hispanic persons have lower status than non-Hispanic (NH) white ones and are less likely to smoke. Lower smoking rates among Hispanics are consistent across age groups and have persisted over time despite substantial declines in smoking. I set out to explore this puzzle by conceptualizing teen smoking as both unhealthy and delinquent behavior. What community and individual aspects of social life in the United States may account for lower smoking rates among Hispanics adolescents relative to NH white ones? To what extent does the protective effect of Hispanic ethnicity extend to the different immigrant generations? I address these research questions with a new, comprehensive integration of segmented assimilation theory with the four main criminological theories. I test the resulting causal model using data from the first wave of the National Longitudinal Study for Adolescent to Adult Health. I find that, at the community level, Hispanic teens are less likely to live in co-ethnic, supportive neighborhoods than NH white ones. The causal model predicts Hispanic teens, then, are at higher risk for smoking, but multilevel regression analysis indicates no community-level effect on adolescents’ individual-level smoking risk. I also find that, at the individual level, Hispanic and 1.5-generation adolescents, those brought to the United States as children, have lower exposure to pro-smoking social learning and general strain factors and higher exposure to protective social control and self-control ones. In single-level regression models, perceptual neighborhood characteristics and social factors mediate the association of Hispanic ethnicity with current and daily smoking. Both Hispanic ethnicity and 1.5-generation status retain their protective effect against the outcome of daily smoking, but Hispanic ethnicity loses its association with current smoking. The complete mediation of Hispanic ethnicity is due to the measure friends who smoke, so it needs careful interpretation because we know that Hispanic teens, who are less likely to smoke, primarily associate with other Hispanic teens, who are less likely to smoke. Further research is needed to resolve this social tautology.