ABSTRACT
Survival analysis, or event history analysis as it is often labeled in the social sciences, includes statistical methods for analyzing the time at which "failures" or events of interest occur. Survival analysis is an important statistical topic that when treated in its full depth and breadth fills entire volumes. This chapter focuses on basic theory and application for three survival analysis techniques that are commonly used with complex sample survey data. Survival analysis models are classified into four major types parametric survival models, nonparametric survival models, Cox proportional hazards (CPH) model, and discrete time event history data, based on the assumptions that are made concerning the probability distributions of survival times. The chapter introduces nonparametric Kaplan–Meier (K–M) analysis of the survivorship function. It presents a description and application of logit and complementary-log-log (C-L-L) models for discrete time event history data.
