| Criminologists conducting longitudinal research are often faced with the problem of developing estimates from a sample with two groups of people: those who are retained over the course of the entire study and others who drop out before the study is completed. This attrition of subjects over time in longitudinal research can lead to difficulties in estimating basic quantities. One such quantity is the prevalence of continuity and change in criminal behavior over extended periods of time. In this paper, we examine simulation evidence and an empirical application which formalize the key issues that arise in investigations with subject attrition. |
Updated 05/20/2006