HIV/AIDS Knowledge: An Age-Period-Cohort Analysis of Trends in Malawi from 1992-2010

Theresa M. Fedor, University of Pennsylvania

This paper explores how individuals update knowledge of HIV/AIDS over time in Malawi, a high HIV prevalence country in sub-Saharan Africa. HIV knowledge uptake could potentially be operating across both periods and birth cohorts. Individuals of all ages may become more informed about HIV at a relatively equal pace from year to year, implying an effect across period of time. Alternatively, individuals born in later years may be more likely to accept new knowledge relative to older generations, implying an effect according to birth cohort. The goal of this paper is to address whether HIV knowledge is changing over time in Malawi more strongly by period of time or by birth cohort. I use a cross-classified random effects age-period-cohort (APC) method developed by Yang and Land, which alleviates the major empirical identification problem posed by the simultaneous estimation of age, period and cohort.