Testing student models of academic achievement.

다국어 초록 (Multilingual Abstract)

The purpose of the present study was to test a model featuring student variables as predictors of academic achievement. Several leading educational researchers have proposed theoretical models to explain direct and indirect influences on students' educational outcomes. The models proposed by these researchers have focused primarily on the impact of home and classroom variables on academic achievement. The initial goal of this project was to test a comprehensive model of academic achievement including student, home, and instructional variables; however, limitations in the size of the sample as well as the quality of the data prevented the realization of this goal. As a result, I focused on models of academic achievement including student variables only (i.e., prior achievement, motivation, interpersonal skills, study skills. participation, and problem behaviors).
Students, teachers, and parents completed measures as part of the project, and data was collected for a total of 102 children across Grades 3 through 5. Structural equation modeling (SEM) was used to test the fit of the hypothesized student model with the data. In addition, exploratory SEM analyses were used to assess the fit of the student model across known groups of students. Exploratory SEM analyses also were used to test the fit of a model of achievement including a home variable, parent involvement in schooling, in addition to student variables.
Results indicated that the hypothesized student model did not fit the data particularly well, however, the best-fitting model was developed with the revision of a few pathways in the hypothesized model. Within this best-fitting model, only motivation and prior achievement demonstrated large total effects with current academic achievement. Of the remaining four variables, study skills and participation demonstrated small total effects, and interpersonal skills and problem behaviors demonstrated negligible total effects. The best-fitting student model demonstrated acceptable fit across males and females, but it demonstrated poorer fit across students distinguished by disability and minority status. Finally, the model of academic achievement including home and student variables demonstrated acceptable fit with the data. These results along with the limitations of the study provide several directions for future research.