|
|
Carolina
Milesi The purpose of this study is to understand the effect of social background on early student achievement, as mediated by childrearing practices. Four questions are posed: To what extent are childrearing practices explained by social background? What is the direct effect –independent of family background- of childrearing practices on student academic achievement during the kindergarten year? To what extent do childrearing practices mediate the association between family background and kindergartners’ achievement? Finally, do these effects vary by racial and ethnic groups? These questions are addressed in the context of a seasonal pattern of achievement, in which cognitive gain trajectories during the academic year are essentially neutral with respect to social background inequalities. Annette Lareau’s conceptualization of the cultural logic of childrearing is meant to reflect the process by which accumulated family resources are activated in the particular context of schools. Based on nationally representative data from the Early Childhood Longitudinal Study- Kindergarten Class of 1998-99 (ECLS-K), structural equations are used to estimate a multiple group model for White non-Hispanic, Black, and English and Spanish proficient Hispanic children. Results show that the class-based typology developed by Lareau does represent distinctive childrearing approaches. As predicted, middle-class practices are positively associated with student achievement at the beginning of kindergarten year for all groups. By contrast, the effect of working-class practices is negative. An interesting exception to this negative association corresponds to Hispanics assessed in Spanish. After controlling for fall achievement, the impact of both childrearing practices and social background on spring achievement is reduced. Supporting the seasonal learning pattern, the effect of these factors on spring achievement operates almost completely through its previous impact on fall achievement. Given the specification of the model, specific structural effects do not vary significantly across racial and ethnic groups. Possibilities of extending this analysis are finally considered. |