In addition to high school academic curricula, many other variables can influence postsecondary education outcomes. Therefore, it is necessary to use multivariate analysis techniques to disentangle the net influence of related variables on the outcome of interest.
In this study, covariance adjustment techniques based on simple linear regression models were used to analyze two persistence outcomes:
Independent variables reflected students’ academic experience in high school (academic curricula and college entrance exam scores), demographic characteristics (gender, race/ethnicity, age), socioeconomic characteristics (income and parents’ education), and the economic status of their high schools (the proportion of students eligible for free or reduced-price lunches).
Other variables reflected students’ experiences in their first year in college, including the type of postsecondary institution, full- or part-time attendance, and work status. In addition, because previous research has shown first-year grade-point average (GPA) to be a strong predictor of success in college (e.g., Pascarella and Terenzini 1991), both analyses were run twice, once without GPA, and once including GPA as an independent variable.
High School Academic Curricula
The results indicated a strong association between high school academic curricula and both measures of persistence. Students who participated in rigorous high school curricula were at a distinct advantage over those who completed no higher than core curricula (the comparison group). In addition, there was some evidence that completing mid-level curricula also was associated with higher rates of staying on track to a bachelor’s degree when compared to those completing programs no higher than core curricula. However, the difference did not reach statistical significance after adjusting for the design effect of the dependent variable.5
SAT Scores and Other Variables
In both persistence analyses, prior to including first-year college GPA as an independent variable, SAT composite test score levels and high school academic curriculum levels were significantly associated with the outcome. However, once GPA was taken into account, high school academic curriculum remained a significant predictor of persistence (specifically, completing a rigorous versus core or lower curriculum), but the association between levels of SAT scores and persistence disappeared. In other words, once all related variables were taken into consideration including college GPA, entrance exam scores were no longer associated with the likelihood of persisting, either with respect to institutional retention or staying on track to a bachelor’s degree.
Other variables were also significantly associated with both measures of persistence after holding related variables consistent. For example, students whose parents did not attend college were less likely to persist than those whose parents were college-educated.6 In addition, students who started college attending part time and/or working full time were less likely to persist than their counterparts, as were those who first enrolled in less-selective 4-year institutions compared with those in selective institutions.