Research That Matters (January 17 - 20, 2008)

Background and Purpose: The academic achievement of low-income and minority youths has become a significant public policy concern (Orfield et al., 2004). Although there are many empirical studies showing that parental SES makes a significant difference for achievement and school completion, it is unlikely that parental SES influences youth's academic achievement directly, instead its effects are mediated by parenting, school, and other factors in the domains influencing youth's development. This paper presents a combined model describing the mediating effects of parenting, school and student factors in relation to the effects of family background characteristics on academic achievement of Mexican American youths, who are among the most academically vulnerable populations of youths in the U.S. The research questions this paper addresses are: do parenting, school, and student factors mediate the effects of parental SES on academic achievement? What are the direct and indirect effects of parenting and school factors on achievement? And, what parenting, school, and student factors contribute most to academic achievement for this population of youths directing us to promising points for intervention?

Method: A model of relationships between family background, parenting, school, student factors, and academic achievement was developed based on prior research with Latino and Mexican American populations. Structural equation modeling (SEM) with MPlus 3.1 (Muthen and Muthen, 2005) was used to assess the theoretical model using a subset of data from the National Educational Longitudinal Study (NELS) representing Mexican American youth in the U.S. (n=2206). The combined model evaluates the effects of eight family background factors (income, mother's and father's occupation and education, family size, immigrant generation, and language), three parenting factors (parental involvement, allocation of resources to education, and help with homework), three school factors (cultural competence, presence of Latino/a teachers, and student composition) and six student factors (positive school behaviors, academic and occupational expectations, language proficiency in English and Spanish, prior grades and being held back) on 8th grade test scores.

Results: In the resulting structural model shows no significant direct effects of background family factors on academic achievement, indicating that they are fully mediated by parenting, school, and student factors. The most significant direct predictors of test scores are student expectations (ß=0.432, p<0.001), student composition of the percentage of poor and minority students in school (ß=-0.224, p<0.001), prior grades (ß=0.176, p<0.05), parental help with homework (ß=-0.138, p<0.001), and English proficiency (ß=0.132, p<0.05). In addition, parental allocation of resources (ß=0.540, p<0.001), percentage of poor and minority students in school (ß=-0.252, p<0.001), parental involvement (ß=0.245, p<0.05), and presence of bilingual education (ß=0.150, p<0.05) had significant indirect effects on test scores.

Conclusions and Implications: This empirically validated model can provide a basis for policy and program interventions to promote academic achievement among Mexican American youths. Findings suggest that increasing youths' academic and career expectations, parental allocation of resources to education, and decreasing segregation of the most disadvantaged students in schools may be some of the most promising targets for interventions with this population.