Mobility ratios and association in mobility tables

Abstract In the late 1940's a similar problem occurred in the work of three sociologists, working in three countries, on three similar sets of data. Natalie Rogoff, David Glass and G6sta Carlsson all faced the problem of making sense of data on intergenerational occupational mobility. A matrix of frequencies of occupations of respondents by occupations of fathers could be converted, in an obvious and straightforward fashion, into matrices of inflow and outflow percentages. Their joint problem arose in comparing inflow percentages across rows or outflow percentages across columns. The problem was that, as sociologists, concerned with the extent that origins in socially meaningful categories influenced destinations in the same socially meaningful categories, they were stuck with occupational categories that differed from one another dramatically in size. A secondary (though hardly trivial) problem was the fact that, in all their data, the two marginal distributions, the respondents' generation and the fathers' 'generation', were notably dissimilar - a consequence of both differential fertility and a general upward shift in the occupational distributions of the three countries. All sought a technique that would 'make the two time periods comparable with respect to occupational structure'.