School of Statistics and Planning (SSP) Collection

Permanent URI for this collection

https://dissertations.mak.ac.ug/handle/20.500.12281/5609

Browse

Recent Submissions

Now showing 1 - 5 of 2678
  • Item
    Determinants of littering behaviour among students at Makerere University
    (Makerere University, 2026) Nankya, Patricia
    Littering remains a persistent challenge on university campuses despite the availability of waste management infrastructure, raising questions about the factors that drive student disposal behavior. This study assessed the major determinants of disposal behavior among students at Makerere University, Uganda, focusing on students' knowledge and awareness, attitudes, environmental and institutional conditions, and demographic characteristics. A quantitative cross-sectional design was employed, with data collected from 280 students using a structured questionnaire administered online via KoboCollect. Data were analyzed using R version 4.5.3 and IBM SPSS Statistics version 27, employing descriptive statistics and binary logistic regression. Results show that although most students demonstrated high environmental knowledge (63%) and positive attitudes toward cleanliness (76%), neither knowledge nor attitudes significantly predicted disposal behavior. Inadequate and inaccessible waste bins, alongside weak enforcement of anti-littering regulations, were widely perceived as key contributors to littering. Place of residence was significant at the 1% level, with on-campus residents showing higher odds of good disposal behavior than off-campus residents (OR = 3.778, p < .001). Bin accessibility on campus and college (specifically COCIS) were both significant at the 10% level (OR = 1.990, p = .052; OR = 4.115, p = .090, respectively). The findings suggest that disposal behavior among university students is driven less by knowledge or attitude. The study recommends improving waste disposal infrastructure particularly within high-density colleges, strengthening enforcement, and directing targeted interventions toward off-campus students.
  • Item
    What influences the quantity of beer consumed by Kampala residents? Some insights from Buziga-Salaama village
    (Makerere University, 2026) Namuli, Philippah. Tamale
    This study examined the factors influencing the quantity of beer consumed in Buziga-Salaama village, Kampala. Primary data was collected from 96 respondents using a structured Kobo Collect questionnaire administered through face-to-face interviews. Respondents were selected using simple random sampling. The data was analyzed using univariate, bivariate and multivariate analysis. Ordinary Least Squares regression was used to estimate the effect of gender, age, income, price of beer, education level and bottle size on the quantity of beer consumed, holding all other variables constant. The results revealed that gender, age and price of beer were statistically significant determinants of beer consumption at the 5% level of significance. Females consumed approximately 64% fewer bottles of beer than males, holding all other variables constant, implying that males are the dominant beer consumers and that excessive consumption among males may result into reduced productivity, domestic violence and neglect of household responsibilities. A one year increase in age was associated with a 9% increase in beer consumption before declining at older ages, implying that younger residents consume more beer which may be driven by curiosity and the desire to fit in with peers who drink, which may result in long term health complications. A one Uganda shilling increase in the price of beer was associated with a 0.1% decrease in the number of bottles consumed, implying that beer in Buziga-Salaama village is relatively affordable and accessible, contributing to higher consumption levels. Income, education level and bottle size were not statistically significant. Based on these findings, it is recommended that public health interventions targeting excessive beer consumption focus primarily on male consumers and younger residents. Community awareness programs should educate residents on the recommended safe limits of beer consumption and the long term health consequences of excessive drinking regardless of age. Relevant authorities should also consider fiscal policy measures such as alcohol taxation to reduce excessive beer consumption in communities like Buziga-Salaama village.
  • Item
    What are the factors influencing the cost of rent in Kampala suburb? Some responses from Mackay village, Namirembe parish, Rubaga county, Kampala district
    (Makerere University, 2026) Sserwadda. Cyrus. Mathew
    The study examined the factors influencing the cost of rent in Mackay Village, Kampala, Uganda. Specifically, the study investigated the effects of household income, household size, distance from Mengo Market, number of rooms and age of the occupant on monthly rent paid. Primary data were collected from 96 households selected using simple random sampling through structured questionnaires. Data were analysed using descriptive statistics, Pearson correlation analysis and multiple regression analysis. Diagnostic tests were conducted to assess the adequacy of the estimated regression models. The logged regression model was found to be more appropriate for analysing the data having a Breusch–Pagan test produce a probability value of 0.43, indicating no statistically significant evidence of heteroskedasticity. The model also demonstrated high explanatory power, with an R-squared value of 0.89, implying that approximately 89% of the variation in monthly rent was explained by the explanatory variables included in the model. Furthermore, the model was jointly statistically significant (F = 123.96, p < 0.001). The findings showed that household income was positively and significantly associated with monthly rent. Holding other factors constant, a 1% increase in household income was associated with an approximately 0.25% increase in monthly rent (β = 0.25, p < 0.001). Distance from Mengo Market had a negative and statistically significant relationship with monthly rent, with a 1% increase in distance associated with an approximately 0.36% decrease in monthly rent (β = −0.36, p < 0.001). The number of rooms was also positively and significantly associated with monthly rent, with a 1% increase in the number of rooms associated with an approximately 1.20% increase in monthly rent (β = 1.20, p < 0.001). Household size, age of the occupant and the squared age term were not statistically significant at the 5% level. The study concludes that household income, proximity to Mengo Market and the number of rooms are significant factors influencing monthly rent in Mackay Village. The findings suggest that higher-income households tend to pay higher rents, properties located closer to the commercial centre command higher rental prices, and housing units with more rooms attract substantially higher rents. The study recommends the promotion of affordable housing, improved infrastructure and accessibility in areas farther from commercial centres, and the development of housing units that cater to different household income levels and accommodation needs.
  • Item
    Design and development of an automated insurance underwriting and premium pricing system using R Shiny with machine learning integration.
    (Makerere University, 2026) Obwangor, David
    The insurance industry increasingly faces challenges related to manual underwriting processes, inconsistent risk assessment, lengthy policy approval times, and difficulties in determining accurate risk-based premiums. These challenges can lead to operational inefficiencies, pricing inaccuracies, and reduced customer satisfaction. This study presents the “Design and Development of an Automated Insurance Underwriting and Premium Pricing System Using R Shiny with Machine Learning Integration”, aimed at improving underwriting efficiency, consistency, and decision-making within insurance operations. The system was developed using the R programming language and the R Shiny framework following the Agile Software Development (ASD) methodology and Design Science Research (DSR) approach. It integrates actuarial pricing techniques based on Poisson-Gamma Generalized Linear Models (GLMs) with Machine Learning (ML) algorithms including Random Forest, Decision Tree, Linear Regression, and Ensemble Models. The system supports the complete underwriting workflow, including customer application capture, risk assessment, premium computation, underwriting evaluation, policy approval, policy database management, user administration, and audit trail management. The developed system incorporates role-based access control, two-factor authentication, policy management, document handling, compliance monitoring, and an embedded Artificial Intelligence (AI) assistant known as ARIA (Automated Risk and Insurance Assistant). The pricing engine generates premium recommendations using both actuarial and machine learning approaches and combines the results to produce a final premium estimate. Interactive dashboards and real-time analytics further support informed underwriting decisions. The study demonstrates that integrating actuarial models, machine learning techniques, and interactive decision-support tools within a single platform can significantly enhance underwriting operations and premium pricing processes. The system provides a cost-effective and scalable solution for insurers seeking to improve efficiency, transparency, consistency, and risk management. Future improvements may include integration with claims management systems, reinsurance modules, external databases, and real-world actuarial datasets to further enhance predictive accuracy and operational capability.
  • Item
    The impact of ATM card ownership on the saving culture of students at Makerere University,Kampala.
    (Makerere University, 2026) Akurut, Sandra
    This dissertation examines the impact of Automated Teller Machine (ATM) card ownership on the saving culture of undergraduate students at Makerere University, Kampala. Conducted between March and April 2026, the study employed a descriptive cross sectional survey design to capture the financial behaviors of 147 randomly selected respondents. Data were collected through questionnaires and interviews, providing both quantitative and qualitative perspectives on students’ saving practices. The research was guided by three objectives: to assess the relationship between ATM card ownership and saving behavior; to explore how accessibility and convenience of ATM cards influence saving discipline; and to identify challenges students encounter in managing their savings while using ATM cards. Findings revealed that ATM card ownership significantly affects saving behavior. While ATM cards enhance financial inclusion and provide secure, convenient access to funds, they also encourage frequent withdrawals and impulsive spending, which undermine saving discipline. Students reported challenges such as limited financial literacy, peer influence, and difficulty in maintaining consistent savings due to the ease of access to money. The study concludes that ATM cards exert both enabling and constraining effects on students’ saving culture. To address these challenges, the research recommends that financial institutions design student friendly savings products, such as restricted withdrawal accounts and tailored savings packages, to encourage disciplined saving. Universities should integrate financial literacy programs into student support services, equipping learners with practical skills in budgeting and responsible card usage. Policymakers, particularly the Bank of Uganda, are urged to strengthen financial inclusion strategies by promoting innovations that balance convenience with saving discipline. Collectively, these measures will not only foster sustainable saving habits among students but also contribute to Uganda’s broader goals of financial inclusion, economic empowerment, and long term financial stability.