Is There a Mobility Effect? On Methodological Issues in the Mobility Contrast Model

Abstract

Social mobility scholars have long been interested in estimating the effect of intergenerational mobility, typically measured by differences in the socioeconomic status between parents and offspring, on later-life outcomes of offspring. In a 2022 article “Heterogeneous Effects of Intergenerational Social Mobility: An Improved Method and New Evidence,” Luo proposes a new approach called the mobility contrast model (MCM) to define and estimate mobility effects. We argue that the MCM is inherently flawed due to its reliance on the coding scheme used for the categorical variables of social origin and destination. Specifically, when different coding schemes are applied, the estimands defined in the MCM bear distinct meanings, involve different but equally arbitrary constraints, and sometimes yield contradictory results. Moreover, regardless of the coding scheme, these estimands do not adequately capture the sociological concept of a mobility effect. To illustrate this, we reanalyze the Occupational Changes in a Generation Study data used in Luo’s study, highlighting the inconsistency of results when dummy coding versus effect coding schemes are used.