| Jennifer Glanville
University of Iowa
The measurement, consequences, and antecedents of school engagement: Comparisons among African American, Latino, and white students
FINAL REPORT
Previous research has observed a strong relationship between school engagement and academic success and retention. Some scholars suggest that a lower level of engagement is at least in part responsible for the achievement gap between White and African American students. Consequently, the concept of school engagement is the focus of a good deal of interest from policy makers. However, we need to know more about the measurement of engagement and its dimensions, the nature of the inter-relationship between engagement and achievement over time, and how these relationships compare across race and ethnicity to draw informed policy-oriented conclusions.
The first study of this project employs confirmatory factor analysis to examine previous measurement strategies that employed data from the National Educational Longitudinal Survey of 1988 (NELS:88). Based on these findings, theory, and previous research, a multidimensional, higher-order measurement model of engagement is then proposed and validated. The higher-order factors are behavioral engagement, measured by attendance, preparation, effort, and at-risk behavior (such as fighting) and psychological engagement, measured by positive relationships with teachers, interest in academics, and motivation. Time spent on homework, a common measure of behavioral engagement, does not load on the higher-order behavioral engagement factor. The measurement model demonstrates measurement invariance across White, African American, Latino, and Asian students, which is important because the consequences and causes of school engagement cannot be compared across race and ethnicity unless the measures are invariant.
The second study examines the effects of different dimensions of engagement on academic achievement, using cross-lagged longitudinal models. The results suggest that several dimensions of behavioral engagement enhance achievement, but that their estimated influences are more modest when the models take the influence of achievement on engagement into account. Psychological engagement does not predict achievement gains, net of the other explanatory variables in the model. Additional analyses observe that the effect of at-risk behavior varies by school, such that compliance with behavioral expectations is more beneficial in schools with higher levels of achievement.
A third study examines whether school engagement is equally beneficial across racial and ethnic groups. Institutional and cultural theory suggests that engagement may not be as beneficial for minority students as for Whites. The results suggest that behavioral engagement predicts test score gains for non-Latino White students, but not for African American and Latino students. In addition, prior academic achievement significantly and positively predicts subsequent behavioral engagement among White students, but not among African American and Latino students.
Taken together, the three studies suggest that school engagement can be beneficial for academic achievement, but that policies aimed at increasing engagement among minority students are not necessarily the best strategies for reducing achievement gaps. However, because the results suggest that the effects of some types of behavioral engagement vary across schools, more attention to the sources of this variation could suggest viable policy prescriptions.
Back to Funded Research Grants Page |