Diagonal Mobility Models: A Substantively Motivated Class of Designs for the Analysis of Mobility Effects

Designs for the analysis of mobility effects are flawed by the manner in which the effects of origins and destinations are parametrized. Consequently, suggestions that mobility experiences do not affect attitudes and behaviors are inconclusive. A new class of models (diagonal mobility models) for the analysis of mobility effects which are demonstrably grounded in sociological theory is proposed, and estimation of these models is discussed. To illustrate the use of these models, a reanalysis of the relationship between fertility and intergenerational mobility is presented, using data from the 1962 Occupational Changes in a Generation (OCG-1) survey. The effects of origin and destination on fertility do not depend upon particular origin or destination classes; the effect of destination on fertility appears to be stronger than the effect of origin; and there is only weak evidence of mobility effects.