手法を比較

	Historical Social Mobility Tables	Historical Nominal Record Linkage
分野	Social History	Historical Demography
系統	Process / pipeline	Process / pipeline
提唱年	1992	2016
提唱者	Robert Erikson and John H. Goldthorpe; log-linear methods from Leo Goodman	Ivan Fellegi and Alan Sunter (probabilistic theory); James Feigenbaum, Ran Abramitzky, Leah Boustan (historical ML methods)
種類	descriptive-tabular	measurement-linkage
原典	Erikson, R., & Goldthorpe, J. H. (1992). The Constant Flux: A Study of Class Mobility in Industrial Societies. Clarendon Press. ISBN: 9780198273837	Abramitzky, R., Boustan, L., Eriksson, K., Feigenbaum, J., & Perez, S. (2021). Automated Linking of Historical Data. Journal of Economic Literature, 59(3), 865-918.
別名	Origin-destination mobility tables, Intergenerational mobility analysis, Log-linear mobility models, Erikson-Goldthorpe mobility tables	Record linkage, Census linking, Fellegi-Sunter matching, Historical individual linkage
関連	3	3
概要	Historical social mobility tables measure how much a person's social position depended on the position of their parents in past societies. The core device is the mobility table: a cross-tabulation of origin class (typically the father's) against destination class (the child's), built from linked parent-child pairs drawn from marriage registers, censuses, or genealogies. Following the framework Erikson and Goldthorpe codified for modern sociology and that historians adapted using HISCLASS, the table is analysed not by raw movement, which is dominated by changes in the class structure itself, but by odds ratios and log-linear models that isolate relative mobility, the strength of association between origins and destinations net of structural change. This distinction between absolute and relative mobility lets historians ask whether genuine fluidity, equality of opportunity, rose or fell across industrialization, migration, and demographic transition, independent of how the shape of the class structure shifted.	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.