Kabira, Judith Kamathi2026-03-052026-03-052025-11https://ir-library.ku.ac.ke/handle/123456789/32654A Research Thesis Submitted in Partial Fulfillment of the Requirements for the Award of the Degree of Doctor of Philosophy of Education (Educational Psychology) in the School Of Education and Life Long Learning, Kenyatta University. November, 2025 Supervisor Doyne K. Mugambi Chrispus K. WawireAcademic self-sabotaging behaviour is a prevalent problem among students in secondary schools in Kenya, with one in five engaging in academic procrastination, and more than half exhibiting disengagement. Evidence links this behaviour to lower academic achievement, academic anxiety, academic stress and poor mental health. Although existing research consistently associates learning environment and academic expectations with academic self-sabotaging behaviour, limited studies have explored these associations within the Kenyan context. Therefore, this study examined how learning environment and academic expectations relate to academic self-sabotaging behaviour as well as their joint predictive weights. Gender differences in academic self-sabotaging and how school type moderates the relationship between learning environment, academic expectations and academic self-sabotaging behaviour were also tested. The study was guided by self-determination theory and situated expectancy-value theory. A convergent parallel mixed-methods research design was used, targeting 2,734 form two students (boys = 1473, girls = 1261) in all public secondary schools in Meru County. Schools were randomly sampled from four strata: boys' boarding, girls' boarding, co-education day and co-education boarding. The quantitative phase involved 400 students (215 boys, 185 girls) selected through proportionate stratified and simple random sampling, while the qualitative phase involved 20 students purposively drawn from the quantitative phase. Data collection tools included a self-administered questionnaire and a face-to-face semi-structured interview schedule. Both instruments were piloted using 46 and 5 Form Two students respectively, from a school not included in the actual study to establish validity and reliability. Quantitative data were analyzed in SPSS version 20 using descriptive statistics and inferential statistics like Pearson Product Moment Correlation Coefficient, independent samples t-test and multiple regression. Qualitative data were analyzed thematically. The results revealed a weak, negative and statistically significant correlation between needs-supportive learning environment, and academic self-sabotaging behaviour (r (396) = -.14, p < .01). Conversely, needs-frustrating learning environment had a weak, positive and significant correlation with academic self-sabotaging behaviour (r (396) = .19, p < .01). The academic expectations of significant others (r (396) = -.23, p < .01) and students (r (396) = -.23, p < .01) had weak, negative and significant correlation with academic self-sabotaging behaviour. Both learning environment and academic expectations jointly and significantly predicted academic self-sabotaging behaviour (F (2, 395) = 23.73, p < .001). Significant gender differences were observed in academic self-sabotaging behaviour in favour of girls (t (395.78) = -3.19, p = .002). School type significantly moderated these relationships. Qualitative findings reinforced these results. The study recommends that schools should promote an autonomy-supportive school environment that meets students’ basic psychological needs while minimizing practices that frustrate these needs. Significant others and students should maintain adaptive academic expectations. Teachers should identify and address gender-specific challenges to come up with solutions targeted at reducing academic self-sabotaging behaviour.enThesis