|dc.description.abstract||The study investigated some factors that influence academic self-concept of primary school pupils, Specifically, absenteeism and class-retention patterns were investigated. Academic self-concept of. the pupils as influenced by absenteeism, class-retention, teacher ratings and academic achievement was studied. Further, it looked at how some home and pupil-related factors were associated with academic self- concept.
It was a survey study that employed an ex-post-facto design. The study was guided by the Symbolic Interaction Theory (Mead 1934)
The study was conducted in Bondo District of Nyanza Province, Kenya. Stratified sampling was used to come up with 5 schools used in the study. All the pupils in classes 2-4 and 6-8 participated in the study. The sample included S headtcachers, 29 teachers and 972 pupils (497 male and 475 female). The pupils' mean age was 11.9 years with a range of 6 to 20.
Research instruments included- questionnaires; school records, academic self-concept ladders teacher rating scales and 2006 KCPE results records. The pupils' questionnaires and academic self-concept ladders were group-administered in the respective classes by the researcher with the assistance of the class-teachers. Teachers' questionnaires were dropped in their respective schools and collected after 2 weeks.
Statistical Package for Social Sciences (SPSS) was used to analyze quantitative data. All the hypotheses were tested at P < .05 level of significance.
According to the findings, pupils in the sample had a high academic self-concept in the subjects tested. Pupils in the lower classes had signifignantly higher academic selfconcept than those in upper classes; class by class analysis indicated that girls tended to have higher self-concept that boys in lower classes but this changed as they moved to upper classes. Nonetheless, in the upper classes, the girls still maintained higher academic self-concept scores in languages than the boys who led in mathematics and science.
Academic self-concept was found to be positively related to teacher ratings and negatively correlated to absenteeism and class-retention.
A significant positive relationship between academic self-concept and academic achievement was observed only at standard 8. In other classes, there was no consistent pattern in the relationship.
In a multiple correlation analysis to find out which of the variables in the study would be fitted into a model to predict academic academic-self concept, academic achievement emerged as the best positive predictor of academic self-concept. With age factor
included, in the model, ago became a major negative predictor of academic self concept at standard seven.
Absenteeism was found to be negatively correlated to academic self-concept (p < .05), teacher ratings and age (across the sample), and was positively correlated to classretention.
There was a significant interaction effect between class-retention and grade. Comparing mean scores of retainees to non-retainees before and after retention revealed that classretention significantly improved the mean performance of the pupils in classes 7 &8 (p = .000) but had negligible effect in lower classes.
Although there were class differences in retention rates, no gender differences were observed.
Teacher ratings were positively related to academic self-concept and the relationship was strongest and significant academic self-concept in mathematics.
On the whole, the findings of the study supported the guiding theory of the study - the Symbolic Interaction Theory.
On the basis of the findings, a number of conclusions and recommendations for education policy makers and other stakeholders were made.||en_US