Event History Analysis of Behavior Modification

Abstract

This paper examines the antecedents of behavior modification in the context of

smoking cessation behavior. Event history analysis is used as a useful research method for

investigating a dynamic process. This method is especially useful for understanding the

process in which individuals move from one qualitative state to another and identifying the

variables which affect this process. A partially parametric model, called proportional hazard

model, is used to quantify the relationship between transition rate and a set of explanatory

variables. Theoretical and methodological implications are then drawn and discussed.