Advanced level physics performance among Uganda students as a function of ordinary level physics and mathematics performance

Abstract

Performance in U.C.E. Physics and Mathematics is used as a selection criterion for admitting students to A-level physics. The extent to which a student’s performance in A-level physics can be predicted from his performance in O-level physics and mathematics is reported. 460 students (364 males and 96 females) from 33 randomly selected A-level schools in Uganda who sat the U.C.E examinations in November, 1986 and the U.A.C.E examinations in March, 1989 were used in the study. Using grades awarded by UNEB to these students in O-level physics 535 and mathematics 456 (as predictors) and A-level physics P510/1. P510/2. P510/3 and P510 (as the criteria). The following hypotheses were tested: Ho1 “There is no significant correlation between performance in Olevel physics and mathematics and the corresponding performance in A- level physics.” Ho2 “Males and females are equally predictable on performance in A- level physics.” Ho3 (a) “There is no difference in A-level physics performance of students from low-performance and those from high performance O-level schools. (b) “Males and females perform equally well in A-level physics.” (c) “There is no interaction between sex and 0-level school quality in determining performance in A-level physics. To test the first two hypotheses, Pearson’s product-moment correlation coefficients were computed between the predictors and criteria and they were found to range between 0.34 and 0.47 for males: 0.35 and 0.68 for the females: and 0.33 and 0.49 for the entire group. All the correlation coefficients were found to be highly significant (p<0.0005) leading to a rejection of hypothesis Ho1. The correlations with the practical paper (P510/3) were found to be the lowest, possibly due to the low-weight given to the physics practical when grading the 0-level physics examination. Overall performance in A-level physics was regressed on the performance in 0-level physics and mathematics. The combination of physics and mathematics was found to improve the prediction of performance in A-level physics, explaining about 30% of the variation therein. The significance of the differences between the correlation coefficients for each sex was tested using the Fisher Z- transformation. Females were found to be more predictable than males when overall performance in Alevel physics was considered. This difference in prediction, however, broke down when the individual physics papers were considered. Hypothesis Ho2 was thus rejected, only in part. Using the overall performance in A-level physics as the dependent variable. A two-way ANOVA was performed to test the significance of differences in performance due to sex and 0-level school quality. The school quality was established dichotomously judging from the overall performance of candidates in the school in 0-level physics. Schools whose mean-grade was better than the national mean were categorized as high-performance and for the reverse as low-performance. Contrary to views expressed in current literature, the study found the performance of females to be significantly higher than that of males in Alevel physics. It was further found that students form high-performance 0level schools performed better than those from low-performance schools. On the interaction between sex and school quality it was found that the difference in performance between students from high-performance and low-performance schools was more marked among the females. The third hypothesis was thus rejected as well. Plausible explanations for the above findings were given and as a result, recommendations were made to students, teachers, educational planners, UNEB and to future researchers.