Analyzing place of residence as a predictor of undergraduate academic performance at the School of Statistics and Planning (SSP), Makerere University

Abstract

This study analysed place of residence as a predictor of undergraduate academic performance among students of the School of Statistics and Planning (SSP), Makerere University. The specific objectives were to examine the effect of place of residence, distance to campus, accommodation expenditure, study environment, and internet accessibility on students' cumulative grade point average (CGPA). A cross-sectional, quantitative research design was adopted. Primary data were collected from 250 undergraduate students at SSP using a structured, self-administered online questionnaire. Data were analysed at three levels: univariate analysis using frequencies, percentages, means, and standard deviations; bivariate analysis using the Pearson correlation coefficient; and multivariate analysis using ordinal logistic regression (the proportional-odds model), since the dependent variable, academic performance, is measured as four ordered CGPA categories rather than a continuous score. The model was checked for multicollinearity and for the proportionalodds (parallel-lines) assumption. The univariate results showed that most respondents lived in private hostels close to campus and reported a moderately favourable study environment and internet accessibility. At the bivariate level, only internet accessibility was significantly and positively correlated with CGPA (r = 0.17, p = 0.006); place of residence, distance to campus, accommodation expenditure, and study environment were not significantly correlated with academic performance. The ordinal logistic regression confirmed this pattern: internet accessibility was the only variable with a statistically significant effect on the odds of being in a higher CGPA category (odds ratio = 1.40, 95% CI: 1.11–1.76, p = 0.004), while the remaining predictors were not significant at the 5% level. The overall model explained only a modest share of the variation in academic performance (McFadden pseudo R-squared = 0.01) and the overall likelihood-ratio test was not significant (χ²(7) = 9.99, p = 0.189). The study concludes that, among SSP undergraduates, reliable internet access at one's residence is more strongly linked to academic performance than the type of residence, its distance from campus, its cost, or its physical study conditions. It is recommended that the university and private hostel operators around Makerere invest in reliable internet infrastructure, and that students prioritise internet access when choosing accommodation, as a practical step toward supporting better academic outcomes.