Advanced methods for the analysis of complex event history data: Survival analysis

Professor Stephen Jenkins, University of Essex

This course is an introduction to survival analysis, also known as ‘event history analysis’. These are the statistical methods used to analyse data on spell lengths of time in a state, e.g. how long a marriage lasts, or the duration of a spell of unemployment. The course focuses on the case in which the data on spell lengths are grouped into bands: survival times are measured as discrete variables rather than continuous ones. This case often arises with social science data, but most textbooks do not discuss it in detail. For simplicity, the course focuses on models for a single spell per observation unit.

Course materials

The course is based on the material in Jenkins (2005a) with hands-on sessions based on the accompanying Lessons (Jenkins 2005b).

Jenkins, S.P. (2005a). Survival Analysis, unpublished Lecture Notes manuscript, Institute for Social and Economic Research, University of Essex.

Downloadable via

Jenkins, S.P. (2005b). Lessons (to accompany Survival Analysis op. cit.), 9 pdf files, Institute for Social and Economic Research, University of Essex.

Downloadable via www.iser.essex.ac.uk/survival-analysis 

Further reading

Allison, a sociologist, provides a gentle introduction to all the important issues. Box-Steffensmeier and Jones (2004) is a graduate text, written by political scientists. Blossfeld et al. (2007) is the latest edition of a graduate text, written by sociologists. Economists might prefer the nice introduction to survival analysis in Chapter 20 of Wooldridge (2002). All the materials cited contain examples, most derived using Stata.

Note: some of the documents on this page are in PDF format. In order to view a PDF you will need Adobe Acrobat Reader