Structural Equation Models Assessing Relationships Among Student Activities, Ethnicity, Poverty, Parents' Education, and Academic Achievement

Foreword

Previous research and technical reports produced by the Washington School Research Center have explored the relationships among important school and student variables and academic achievement. Family level income, student ethnicity, the school environment, and specific classroom practices have all been shown to be important determinants of student success in school. At the school level, family income as reported in the percentage of students qualifying for free/reduced lunch is generally the strongest predictor of student achievement. However, the effects of poverty are being at least partially overcome in some schools by other factors. In this sixth technical report produced by the Center, Professors Joireman and Abbott continue the analysis of state-wide data bases to add to our understanding of the factors that help explain the variability we see in student achievement levels.

Their results based on 9th grade data show that, indeed, family income, parents' education levels, and student ethnicity are important in predicting student achievement. Using a sophisticated structural equation modeling approach they are able to sort out the unique contributions each of these factors makes to student achievement, while considering three student factors at the same time. The three "student activities" they include in the analyses are the amount of homework a student does, the number of school or community activities in which the student participates, and the amount of television the student watches. As it turns out, all three of these variables are significant predictors of student achievement apart from family income, parents' education levels, and student ethnicity.

While the statistical analyses are complex, the results can be stated relatively simply. Homework is positively related to achievement. Too much time spent watching TV is related to lower achievement and, unfortunately, the amount of TV watched tends to be greater among non-white students. Participation in school and community activities appears to encourage higher achievement beyond the effects of ethnicity or low-income status. While participation is less in non-white populations, it is greater among low income students. This suggests that encouraging participation may be an important way to affect achievement especially since schools cannot change ethnicity or income.

Interestingly, participation in school and community activities appears to be almost as strong a predictor of school success as does the amount of time a student spends doing homework. Perhaps this finding indicates just how important it is that we create schools where students feel "connected" and where they are important members of a community engaging in an important endeavor.

Jeffrey T. Fouts, Ed.D. Executive Director
Washington School Research Center
Lynnwood, WA

full report (PDF Download)

Back to top