Methoden vergleichen

	Surname-Based Mobility Analysis	Historical Nominal Record Linkage
Fachgebiet	Social History	Historical Demography
Familie	Regression model	Process / pipeline
Entstehungsjahr	2014	2016
Urheber	Gregory Clark	Ivan Fellegi and Alan Sunter (probabilistic theory); James Feigenbaum, Ran Abramitzky, Leah Boustan (historical ML methods)
Typ	regression-estimation	measurement-linkage
Wegweisende Quelle	Clark, G. (2014). The Son Also Rises: Surnames and the History of Social Mobility. Princeton University Press. ISBN: 9780691162546	Abramitzky, R., Boustan, L., Eriksson, K., Feigenbaum, J., & Perez, S. (2021). Automated Linking of Historical Data. Journal of Economic Literature, 59(3), 865-918.
Aliasnamen	Clark surname method, Group surname mobility, Surname-group status persistence, Implied intergenerational elasticity	Record linkage, Census linking, Fellegi-Sunter matching, Historical individual linkage
Verwandt	3	3
Zusammenfassung	Surname-based mobility analysis estimates how strongly social status is inherited across generations without linking a single parent to a single child. Developed by Gregory Clark, it exploits the fact that surnames cluster: certain names were borne disproportionately by elites, others by the poor. By tracking how over-represented or under-represented a surname group is among elites, university graduates, physicians, attorneys, the wealthy, across successive generations, one observes how fast that group's relative status regresses toward the population mean. The speed of regression yields an estimate of underlying intergenerational persistence, conventionally denoted b. Clark's striking and contested finding is that this group-level b is far higher, around 0.7 to 0.8, than the 0.3 to 0.5 typically found by conventional parent-child studies, implying that the deep, latent component of social status is far stickier than single-generation correlations suggest. The method extends mobility measurement into eras and places where individual linkage is impossible.	Historical nominal record linkage is the task of recognising when records in different sources, two censuses, a census and a draft register, a baptism and a marriage, refer to the same person, even though no shared identifier exists and names are misspelled, ages misreported, and places renamed. Linkage is the engine behind longitudinal historical micro-data: it builds the life-course panels that underpin studies of migration, mobility, mortality, and the long-run effects of early-life conditions. Three families of methods dominate. Deterministic linkage applies hand-crafted rules; the probabilistic Fellegi-Sunter framework weights field agreements and disagreements by their discriminating power; and supervised machine learning, trained on hand-linked examples, learns to classify candidate pairs. Modern historical practice, led by Abramitzky, Boustan, Feigenbaum, and collaborators, emphasises transparent, replicable algorithms and, crucially, explicit measurement of linkage error, since false matches and missed links can bias every downstream estimate.