Using Spanish Surname Ratios to Estimate Proportion Hispanic in California Cities via Bayes Theorem*

In: Social science quarterly, Band 96, Heft 5, S. 1511-1527

Abstract

ObjectivesTo generate, via application of Bayes Theorem, accurate estimates about the size of Hispanic populations in California cities from very limited data on the surnames of those living in the cities.MethodsWe make use here of the ratio of those with the name "GARCIA" to those with the name "ANDERSON" in those cities, one of which is far more likely to be Hispanic and one of which is far more likely to be non‐Hispanic.ResultsFor four cities that vary dramatically in their Hispanic populations, using only two common names we are able to estimate the Hispanic population in the cities.ConclusionsWe lay the background for our surprising results by underscoring common fallacies in using surnames for purposes of ethnic identification, such as the belief that the proportion of bearers of a given name who are Hispanic can be specified as a unique percentage. We show that how "Hispanic" any given name will turn out to be is a function of the overall demography of the subpopulation being analyzed, which will also affect the distribution of names within that subpopulation.