Retention of teachers on students’ academic achievement in public secondary schools in Kakamega county, Kenya

Abstract

Teachers’ retention in schools has been necessary because it enhances academic attainment. The gist of this study was to evaluate the effect of teachers' retention on students’ academic achievement in public secondary schools in Kakamega County. The study was guided by Education Production Function Theory (EPFT), Theory of Constraint (TOC) and Social Exchange Theory (SET). The target population was 268 public secondary Schools, 1500 teachers, 268 principals and one (1) County staffing officer. The sample size of 30% of the population was deemed to be adequate. The sampling techniques were simple random sampling, stratified sampling and systematic sampling. Data collection procedures were done with the help of administering questionnaires to respondents, document analysis and interview schedules were the major principal instruments used. The validity of the research instruments was assessed with the help of two research supervisors from Kisii University. The reliability of the index of 0.7 for teachers and 0.89 for principals was obtained using a test-retest approach. Data presentation was done with the help of tables, graphs and pie charts. A mixed research method was used. Statistical Package for Social Sciences (SPSS) version 23 was used for data analysis. Findings revealed that 38 (9.3%) teachers had left on promotion, 240 (59%) voluntarily and 129 involuntarily (31.7%). The study revealed, therefore, that most teachers (over 66%) either left voluntarily or on promotion. The study recommended that since most teachers either left voluntarily or on promotion, this was their right, while others were retained to ensure stability without interference through uncalled-for transfers, which enhanced students’ academic achievement. The study was undertaken by ensuring that ethical considerations for research were put into perspective, such as informed consent from participants and non-usage of unethical language as a means to obtain data from respondents.