ABSTRACT
Survival analysis or event history analysis is a set of methods for modeling the length of time until the occurrence of some event. The term ‘survival analysis’ is used most often in biomedical applications where the event of interest is not repeatable, such as studies where the length of time until relapse or loss of life is modeled as a function of some independent variables, often including a treatment indicator. Failure analysis and hazard modeling are related terms. In social research, more neutral terms like event history analysis or duration analysis are commonly used. This chapter uses the term survival analysis to refer to all these methods. An important feature of survival data is that for some cases the final event is not yet observed, and such observations are said to be censored. This chapter provides a brief introduction to survival analysis, and shows how standard survival analysis can be related to multilevel analysis. Next, it discusses the multilevel analysis of survival data.
