THE RELATIONSHIP OF BELIEFS ABOUT MATHEMATICS, MATHEMATICAL SELF-CONCEPT AND ARITHMETIC SKILLS TO MATHEMATICS ANXIETY AND PERFORMANCE IN BASIC STATISTIC

The purpose of the present study was to determine the relative contributions of acceptance of erroneous beliefs about mathematics, self-concept regarding mathematical ability, and arithmetic skills to anxiety in mathematical situations and to performance in the first semester course in basic statistics.;The sample consisted of 92 adults, 16 male and 76 female, who were registered for the first semester course in basic statistics in the School of Education, Health, Nursing and Arts Professions at New York University. Participants were asked to fill out the Beliefs About Mathematics Scale and the Mathematical Self-Concept Scale, developed for this study, the Mathematics Anxiety Rating Scale, and the Arithmetic Skills Test during the last half hour of their first class session of the semester. Approximately eight weeks after the beginning of the semester, participants were given a midterm examination; a portion of this examination was constructed and scored by the reseacher.;Hypotheses 1 stated that there would be a positive linear relationship between mathematics anxiety and beliefs about mathematics. This hypothesis was supported, and a low to moderate relationship was found.;Hypothesis 2 stated that there would be a negative linear relationship between mathematics anxiety and mathematical self-concept. This hypothesis was supported, and a moderate to strong relationship was found.;Hypothesis 3a stated that the combination of beliefs about mathematics and mathematical self-concept would account for a portion of the variance in mathematics anxiety. This hypothesis was supported, and it was found that mathematical self-concept was the only predictor to contribute significantly to the regression.;Hypothesis 3b stated that the combination of beliefs about mathematics, mathematical self-concept, and arithmetic skills would account for a portion of the variance in mathematics anxiety. As stated, this hypothesis was supported; however, it was found that the addition of arithmetic skills to the regression did not improve prediction significantly.;Hypothesis 4 stated that the combination of beliefs about mathematics, mathematical self-concept, and arithmetic skills would account for a portion of the variance in performance in basic statistics. This hypothesis was supported; however, it was found that arithmetic skills was the only significant predictor.;Hypothesis 5 stated that the combination of beliefs about mathematics and mathematical self-concept would discriminate among groups of basic statistics students with respect to level of mathematics anxiety and level of course performance. This hypothesis was supported: one discriminant function, in which mathematical self-concept played the primary role, was found significant, and students could be classified into the four possible anxiety-performance groups with 57% accuracy. It was also found that students' mathematical self-concept corresponded more closely to their level of mathematics anxiety than to their level of performance.;Additional analyses showed that performance in arithmetic skills was substantially related to both beliefs about mathematics and mathematical self-concept, that students high in mathematics anxiety and low in mathematical self-concept tended to give up more easily in response to the Arithmetic Skills Test, that mathematics anxiety tended to be more debilitating for the older student, and that students who withdrew from the corse tended to score relatively high on erroneous beliefs about mathematics.