A journey through numeracy: Correlates of success in initial college mathematics

Date of Award


Degree Type


Degree Name

Doctor of Philosophy (PhD)




Louis J. Heifetz


performance prediction, Mathematics education, Statistics

Subject Categories



This research was undertaken to identify and analyze the correlates of mathematical success at the collegiate level. Multiple regression prediction equations were developed for each of the five initial mathematical sequences that undergraduate students at Syracuse University take. Ultimately, the goal of this study was to predict the grade in students' initial two-course college mathematics sequence solely from variables derived from their high school transcripts.

Subjects in the study consisted of 213 undergraduates (103 men and 110 women) who had completed their high school education in New York State under the New York State Regents curriculum and had completed their sophomore year at Syracuse University. The variables used as predictors were derived from high school mathematics and foreign language course performances, overall high school grade point average, Scholastic Aptitude Test (verbal and quantitative) scores, and gender.

Given the fact that each initial mathematical sequence is typically made up of different "types" of students (in terms of academic interests, goals, and pursuits; mathematical track records, etc.), it was found that different high school variables contributed significantly to prediction of mathematical performance for the different sequence types. The most promising multiple regression prediction equations (in terms of the percent of variance accounted for in the dependent variable), turned out to be the no math sequence group and the financial-management calculus group (accounting for roughly 89% and 77% of the variance in sequence grade respectively). The predictor variable comprised of SAT-verbal scores was the only statistically significant variable that was common to both of these equations. Outside of SAT-verbal, math related variables turned out to significantly predict the no-sequence students' initial grades; whereas language related variables together with overall high school performance and gender, significantly predict grades in the business calculus sequence.

The high Calculus sequence group and the probability and statistics sequence group (accounting for roughly 41% and 30% of the variance in sequence grades respectively), each had standard score of tenth grade mathematics performance as their most important predictor. The low Calculus sequence group (accounting for nearly 43% of the variance in the dependent variable) had a standard score based on the sum of all grades in high school mathematics courses and depth of concentration in a single foreign language course as significant contributing predictor variables.


Surface provides description only. Full text is available to ProQuest subscribers. Ask your Librarian for assistance.