Graduation rates, success and high school quality.

The efficacy of high school grades in retention modeling has been criticized as ineffective, owing to the diversity in both grading standards in high schools and competitiveness of high school curriculums. Informed by theories from status attainment, credentialism, and cultural capital, the present study aims to create a standard with which to test both the efficacy of G.P.A. as a predictive variable, and the relationship that individual student characteristics have to the institution from which they graduate. Using student retention data from Baylor University and high school level institutional data from the Texas Education Agency, I assess the individual and institutional level factors influencing student success in college. Student success is operationalized as college G.P.A., four year graduation, five year graduation and six year graduation. Ordinary least squares regression models, and hierarchical linear models are run to assess the relative effect of individual level variables and institutional level variables on students’ undergraduate G.P.A. Binary logistic models and multinomial models are run to predict the relative effect of individual and institutional level variables on students’ four year, five year, and six year graduation rates. The results of this research support the importance of individual level variables in predicting student success. In addition to individual and institutional level variables, interaction variables are included to assess the relationship between students’ high school G.P.A. and the quality of their high school. The institutional level variables were found to have random effects on predicting student success.Ph.D