Differential Use of Regression and Discrimination Analyses in Attitudes Formation towards Computational Physics Education in Nigeria

Computational physics as an aspect of physics and/or physics education has gained tremendous ground at Nigerian educational levels, especially in universities across the nation. There is no doubt that computational physics is the use of computers to solve scientific problems, especially in physics and applied mathematics. Computational physics complements the areas of theory and experimentation in traditional scientific investigation. It has been successfully applied to areas such as molecular modeling, electronic circuit design, protein folding, atmospheric science, aerodynamic design and testing, and material science. The application of computational physics cannot be underestimated. However, physics education as an aspect of scientific knowledge taught in science education in tertiary institutions in Nigeria is the product of the research process and, as such, entails the gathering of theories and laws synthesized from the data through statistical analysis for generalization of ideas and scientific reasoning. In physics and/or physics education, continuous independent variables such as mathematics test scores, attitude to physics, and computer self-efficacy, among others, contribute to the prowess of individuals in computational physics. Therefore, to ascertain the contributions and/or prediction at the relative and the composite, it is measured or known as regression analysis. On the other hand, the attitudes of students toward studying computational physics education in universities across the nation are measured. Two research hypotheses were developed and tested at a 0.05 level of significance. A self-designed validated questionnaire was used to gather information necessary for the conduct of the research. Ten federal universities were randomly sampled for the study. The first hypothesis states that there will be no significant composite contribution of students mathematics test scores, attitude of physics and computer self-efficacy to studying computational physics in Nigeria (R = . 99; R2 = 98; F (3, 6) = 107. 90; p < . 05). On the other hand, testing the relative contribution of knowledge about differential regression and discrimination analyses in computational physics in tertiary institution in Nigeria (β =1. 02, t = 5.07; p < .05) and discrimination analysis (β =158, t = 1.33; p < . 05). Hence, the use of regression analysis is absolute when predicting the relative and the composite joint contribution of two or more continuous independent variables on continuous dependent measure. Prior knowledge about computational physics right from secondary school in Nigeria would go a long way in preparing students and increasing their passion for physics education.