- Individual-Level Model:
- Students who display higher levels of effort in their schooling, as measured by school attachment (.15) and school commitment (.19), have higher GPAs. Furthermore, associations with positive peers (.09) and parent-child discussion (.08) were significant predictors.
- Family SES (.06) was also significantly and positively related to self-reported GPA.
- Contrary to expectations, school involvement was not significantly related to GPA. Also, parental school involvement was not significantly related to GPA.
- Individual-level characteristics accounted for about 47.6% of the within-school variance in GPA.
- Between-School Model:
- After controlling for individual-level variables, only one of the six school structural variables had a significant effect on GPA: school cohesion (.22).
- Surprisingly, school poverty, proportion non-White, school location, school size, and school social problems were not significantly associated with average GPA when school cohesion and individual level predictors were taken into account.