Date of Graduation


Document Type



The purpose of the study was to determine the weight of a set of available cognitive, affective, and demographic variables in the prediction of success of students enrolled in the three mathematics courses for elementary pre-service teachers at West Virginia University. The mathematics courses were: Mathematics 33, Introductory Mathematics for Elementary Teachers, Part I; Mathematics 34, Introductory Mathematics for Elementary Teachers, Part II; and Mathematics 131, Informal Geometry and Algebra for Elementary Teachers. Ninety-one students participated in the study. Each completed a questionnaire consisting of two parts: the Mathematics Opinionnaire by Aiken and a questionnaire related to students' backgrounds. From these questionnaires a set of thirty-one predictor variables was gathered for examination. Various combinations of the independent variables were used in a multiple regression procedure of the Statistical Package for the Social Sciences (SPSS). Prior to this a matrix was constructed to examine the correlation of the variables. The thirty-one predictor variables were: number of high school mathematics classes, average grade in high school mathematics classes, college grade point average, ACT mathematics score, ACT composite score, score on an Arithmetic test, times an Arithmetic test was taken, high school grade point average, scores on the Freshman Mathematics Exams (Parts A and B), value and enjoyment of mathematics scores, absences in mathematics classes, residency, attempts at credit by exam in the classes, years between last high school mathematics class and first college mathematics class, major, frequency and sources of outside help with mathematics, semester in college, age, and hours taken during the semester. The criterion variable was final course grade. The hypothesis and research questions were examined by means of an F-test. In each situation, an F-test was used to compare the square of the correlation coefficient (R('2)) to zero. The significance of the weights were also tested in this procedure. Three prediction equations were developed. The five variables which contributed the most weight in the prediction equations were: high school grade point average, college grade point average, enjoyment of mathematics score, score on the Arithmetic Test and residency R('2) = .71416 on these five variables. The variable which contributed the most weight to the prediction equations was college grade point average. The information generated by the use of the predictions equations and by analysis of the available information could be used by institutions in the placement and advising of students.