Name-Based Estimators of Intergenerational Mobility.

This article focuses on evaluating and comparing name-based estimators of intergenerational socioeconomic mobility, which use first and surnames as proxies for family links in data lacking direct parent-child connections. It systematically reviews two main classes of estimators—the R² estimator, which measures the informational content of names in explaining socioeconomic status within a generation, and the grouping estimator, which imputes parental status by averaging outcomes within name groups across generations. Using newly digitized Finnish longitudinal data and linked U.S. census samples, the study finds that the interpretation and reliability of these estimators critically depend on sampling properties, especially the degree of overlap between parent and child samples, which varies widely across studies and affects bias direction and magnitude. The authors propose a bias correction method accounting for sample overlap and name informativeness, highlight conceptual caveats such as sensitivity to name frequency and regional or ethnic confounding, and provide practical guidance for researchers applying these methods.