Most undergraduate programs require completion of a mathematics course, but pass rates in these courses are often low. To determine predictors of student success in 100-level mathematics courses at the University of Southern Maine, we used a questionnaire to collect information on possible predictors of grade (student demographics, attitude, educational experience, factors impacting study time) and course-related factors. We then performed univariate analysis using Mann-Whitney and the Kruskall-Wallis tests, and multivariate analysis using ordinal logistic regression modeling, and found that students who were male, older, had missed fewer classes, had taken more 100-level classes, took classes in a once a week format, had a more positive attitude toward mathematics, and had a lower ranked instructor tended to receive higher grades. These results suggest that a supportive learning environment may enhance performance.
Most undergraduate students are required to successfully complete at least one course in mathematics to obtain a baccalaureate degree. However, many students enter college unprepared for the study of mathematics, or other subjects that require quantitative analysis, at the college level, and pass rates in these courses are low (Boughan, 1996). The consequences of these low pass rates can be significant. Undergraduate student rates of entry into and persistence in science, engineering, and mathematics majors are poor nationwide (House, 2000), and failure in an introductory mathematics course may contribute to this trend.
The University of Southern Maine (USM) is a public university in northern New England. To determine if poor performance in entry level mathematics courses is a problem at USM, as it seems to be nationwide, the USM Department of Mathematics and Statistics reviewed pass rates for the fall semester of 2001 in the three entry-level mathematics courses that it offers (MAT 105: Mathematics for Quantitative Decision Making, MAT 108B: College Algebra, and MAT 120D: Introduction to Statistics). It found a 20.7% failure rate for these courses compared to a 9.6% failure rate for all other 100-level courses across the university for the same term. This led to discussions about interventions that might improve student success rates.
A range of factors including student perceptions of/attitudes toward mathematics and science (Olsen and House, 1997), and academic background (House, 2000; Sandler & Tai, 2001) are well established as predictors of student achievement in mathematics and science education. Students' outside workload (Harris, Hannum, and Gupta, 2003) and characteristics of the way a specific class is taught (Lake, 2001) may also play a role at some institutions.
In an effort to recognize the predictors of success and failure in basic mathematics courses at USM, and to suggest interventions that could improve student success, we surveyed students enrolled in the three entry-level math courses at USM during four consecutive semesters in 2003. From each student, we collected general demographic facts, as well as information about academic preparation, attitude toward mathematics, and use of university resources designed to help students in the study of mathematics. We also collected basic information about the teaching of each individual course section. We then used nonparametric univariate statistical methods and ordinal logistic regression to examine the relationships between these predictors and each student's final course grade.
Setting: USM is a comprehensive public university that offers 50 majors and more than 40 academic programs to the increasingly diverse population of southern Maine. USM is comprised of three campuses (in Portland, Gorham, and Lewiston-Auburn, Maine) and three off campus sites (located in Bath/Brunswick, Saco, and Sanford, Maine). Thus, USM serves both relatively urbanized areas (Portland is the largest and Lewiston-Auburn the second largest urban center in Maine) as well as more suburban and rural parts of the state. …