Non-traditional approaches to traditional demographic questions: Three essays on fertility and mortality

Ina Katherine Warriner, University of Pennsylvania


This dissertation is composed of three essays examining different facets of demography, ranging from an issue affecting infant mortality and two issues affecting fertility. The first paper is an assessment of the role of cousin marriage in infant mortality. Previous studies indicate that children of related parents experience higher levels of infant mortality than would be the case if the mother were unrelated to the father. This study assesses the effects of consanguineous family structures on infant mortality in Egypt, Pakistan, Turkey, and Yemen. Disentangling the genetic component from other causes of death is a necessary step toward identifying the impact of inbreeding on mortality. In this paper, community effects, such as access to health care and the proportion of inbreeding in a population are considered using a fixed effects model. Results indicate that controlling for community level variables and sociodemographic, variables greatly reduces the effect of cousin marriage on infant mortality. The second paper examines theoretical and methodological issues associated with data on social networks, focusing on the homophily of network partners chosen for conversations about family planning in four rural areas of Nyanza Province, Kenya. Identifying principles behind the selection of network partners is necessary to establish an association between the influence of social networks and behavior outcomes. Findings indicate that homophily based on sociodemographic attributes and personal preferences guides the composition of family planning networks and selection of network partners based on use or non-use of contraception, although expected to some extent, is secondary and less relevant. The third chapter reviews the literature and data on secondary amenorrhea and its impact on fertility. Fertility models in demography account for postpartum and lactational amenorrhea but do not consider the impact of amenorrhea from other causes. Factors that have been associated with amenorrhea include age, nutrition, psychological factors, strenuous physical exertion, disease, and environmental factors. A closer look at the prevalence of acyclicity in populations before the discovery of oral contraceptives in 1960 and in developing countries today suggests that amenorrhea may be more widespread than is generally assumed and should not be ignored when assessing variations in fertility.

Subject Area

Demographics|Social research|Public health|Families & family life|Personal relationships|Sociology

Recommended Citation

Warriner, Ina Katherine, "Non-traditional approaches to traditional demographic questions: Three essays on fertility and mortality" (1999). Dissertations available from ProQuest. AAI9953615.