Methods: Analyses were conducted using the Longitudinal Immigrant Student Adaptation Study (LISA), which consists of a youth and parent sample surveyed annually from 1997 to 2001. Complete data for the sample (N=308) provides self-report data from youth and their parents (ages 8-14 at Year 1) regarding the effect of family involvement (Year 3) on depressive symptoms (Year 5), controlling for parental education, income, and employment, and child's age, gender, grade level, country of origin, and time since migration. Exploratory factor analysis was utilized to assess the factor structure of the outcome variable, depressive-symptoms, and the primary independent variable, family involvement. Due to significant levels of missing data, multiple imputation analysis was conducted.
Findings: An ANOVA demonstrated no significant differences in family involvement among Dominican (M = 1.33, SD = 1.31), Mexican (M = 1.66, SD = 1.24), and Central American (M = 1.91, SD = 1.39) youth F(2, 184) = 2.86, MSE = 1.73, p = .06. A regression analysis supported a significant negative effect of family involvement (Year 3) on depressive symptoms (Year 5) for Dominican immigrant youth (b = -1.19, p < .05), but not for Mexican or Central American adolescents, while controlling for sociodemographic variables. The model produced a good fit. That among Dominican adolescents, lower family involvement was found to be more severely associated with depression than for the Mexican or Central American groups is not surprising given the high rates of transmigration and its maintenance of contacts with family in the home country among Dominicans (Levitt, 2003; Schiller et al., 1995).
Discussion: It is possible that youth who maintain familial ties in the home country may have lower levels of close local family relationships and robust family involvement in the host country. Therefore, lacking these family resources in the United States, Dominican youth may be less able to cope with the stressors they are exposed to as immigrant youth in the United States (Kasinitz, 2002). Results support the possibility of constancy of depressive-symptom levels over several years as well as the existing literature regarding the relationship between gender and depressive symptoms. Findings support family involvement as a protective factor on depressive symptoms and highlight the importance of understanding its differential role across Latino(a) populations. Researchers and policy makers are recommended to deepen their understanding of unique migration and settlement patterns of Latino(a) groups with respect to protective factors and mental health outcomes and to address these concerns in policy and advocacy efforts.