Changes in HIV/AIDS Knowledge, Attitudes and Behaviors in Malawi

The three chapters of this dissertation collectively assess how HIV/AIDS knowledge, attitudes are behaviors are changing in Malawi. The first chapter assesses how married individuals use knowledge of HIV status to make behavioral changes to reduce HIV risk or make decisions about divorce. Instrumental variable models controlling for selection into HIV testing are estimated using data from the Malawi Longitudinal Study of Families and Health (MLSFH). Results indicate that knowledge of HIV status does not affect chances of divorce but does reduce the number of reported sexual partners among HIV-positive respondents, and increases reported condom use with spouses for both HIV-negative and HIV-positive respondents. The goals of the second and third chapters are to dig beneath behavior itself and look at how potential behavioral changes are motivated, as well as how basic HIV knowledge has changed. Chapter 2 examines ways that HIV prevention efforts may have changed beliefs and attitudes towards HIV risk and HIV prevention, in particular attitudes towards a woman's right to protect against HIV risk. Using MLSFH data, I compare participants and non-participants in a program providing extensive HIV counseling and testing. Results suggest that participants are more likely to believe that women have the right to take steps to protect themselves from HIV risk, are less likely to be extremely worried about HIV infection, and are more likely to think condom use is an acceptable means of protection against HIV. Chapter 3 explores how individuals update knowledge of HIV/AIDS transmission and prevention over time in Malawi. HIV knowledge uptake could potentially be different according to an individual's age, the time frame in which an individual was born, or could be changing predominately over time for all individuals (age, period or cohort). Using Demographic and Health Surveys Data for Malawi in cross-classified random effect age-period-cohort models, I find that period effects dominate over cohort or age effects, meaning that knowledge of effective HIV prevention tactics has increased most strongly over time, net of age and birth cohort effects.