Probability Models for Duration: The Data Don′t Tell the Whole Story
Abstract
Probability models for duration have been applied to a wide range of individual-level and organizational phenomena. Interestingly, seemingly similar models may produce different results. Using divorce as an illustration, we discuss a hierarchy of duration models of different complexity and show how an analysis of the results across models helps explain why different studies may have come to different conclusions. This analysis also leads to substantive insights that even the most complete model by itself does not provide.