Abstract
Data mining and statistical analyses at a less selective institution reveal that the relationships between parents' educational level and some first year success indicators are not linear. Specifically, students who report that either parent or guardian(s) have an educational level beyond a baccalaureate degree or do not report parent education level fair worse in most first year success indicators as well as some pre-college enrollment characteristics than those students whose either parent or guardian(s) have some college or a baccalaureate degree. Mathematics SAT, Critical Reading SAT, placement tests, first semester credits attempted, first semester credits earned, first semester GPA, and one-year retention means among students grouped by five parent education levels are assessed. In addition to the findings, variable construction, definition of first generation students, exploratory analyses, inferential statistics, and predictive models are discussed. Ordinary least squares (OLS) and logistic and regression techniques, limits, and appropriateness of these techniques are investigated.
Get full access to this article
View all access options for this article.
