Authors : O.M. Odeniya, A.A. Abiodun and B.A. Oyejola
Abstract: Discrete time survival analysis approach is often used when only the interval in which the event of failure occurs is known or the event itself occurs in discrete time scale. In this study, the approach was used to analyze data on age at first pregnancy among Nigerian women. Literature also reveals that, in some situations, in addition to observed covariates collected on each individual, there often exists unobserved heterogeneity in the data at individual or cluster level which if not accounted for during analysis may lead to biased estimates and unreliable conclusion. In this study, Discrete Time Logit Model was used to investigate the effect of some covariates (risk factors) on the hazard of age at first pregnancy. To account for unobserved heterogeneity (frailty) the random effects of ethnicity was added to the model. The data used for the study were extracted from the 2005 National HIV/AIDS and Reproductive Health Survey (NARHS). The results of the analysis showed that age at first pregnancy depended on geopolitical zone, location of residence, level of educational attainment, marital status, religion and the age of first sexual initiation. It was also observed, using Akaike Information Criterion (AIC) that the model that accounted for unobserved heterogeneity due to ethnicity was preferred to the one that did not.
O.M. Odeniya, A.A. Abiodun and B.A. Oyejola, 2013. Discrete Time Survival Analysis of Age at First Pregnancy Among Nigerian Women. Research Journal of Applied Sciences, 8: 215-220.