PEER LEARNING: AN ALTERNATIVE TEACHING MODEL FOR GARISSA COUNTY PUBLIC SECONDARY SCHOOLS

Performance in Mathematics by public secondary school students in Garissa County was studied using peer learning and the conventional teaching methodologies. Schools were grouped based on the gender of their students into boys only, girls only and mixed-sex. Using simple random sampling technique, Form Two students in all the single-sexed schools were divided into Control and Experimental groups. The Control groups were taught by the conventional methods while the Experimental, and the students in the Mixed-sex schools which were being studied for effect(s) of interaction, were taught by peer learning methodology under single-blinded setup. Post-test and pre-test Mathematics scores were collected from all the groups. Scores were letter-graded A to E and entered in Microsoft Excel Software. Students who obtained C+ and above, the cut off point for university admission in Kenya were summarized. Statistics such pretest and posttest Means, Standard Deviations and Coefficient of Variation as well as Pearson' Correlation of the Experiment and the Control groups were calculated using Microsoft Excel Software. Using Chi-square test, the influence of peer learning on the exposed groups was tested for significance at 5% confidence level and while confounding factors such as gender of students, text book student ratio, and teacher student ratio were also tested at the same confidence level. The findings suggested that peer learning is a better teaching methodology than the conventional ones for Garissa County public secondary schools. The study also found out that there were significant differences between the Means of the Control and the Experiment groups. There were significant differences between the Mean scores of boys' and that of the girls' schools. The mixed-sex schools did show main significant effects of interaction. Girls gained more than the boys. However, there were high Coefficient of variation within same term than between the two terms. The confounding effects were found to be significant on their own and on peer learning.