Do odds ratios really control for the availability of occupational positions in status contingency tables?
RODERICK J. HARRISON Department of Sociology, Harvard University 484 William James Hall, 33 Kirkland Street, Cambridge, Mass., 02138, USA Techniques based upon odds ratios or log-linear modeling of status contingency tables are now almost universally accepted as providing adequate controls for structural changes in the size and job composition of the occupational structure, and thus permitting comparisons of the process of mobility or of exchange or circulation mobility unconfounded by structural effects. This paper questions whether the techniques are fully adequate. Formulae for odds ratios are derived for matrices generated by vacancy models, demographic manpower models and first-order Markov mobility models the major approaches to explicitly modeling the processes of mobility that might underlie or generate status contingency tables of the sort analyzed in the literature. In each application odds ratios can differ, implying changes in the underlying process of mobility, when only t
Related Questions
- Will ESR workstructures need to be revised as a result of mapping and linking to access control positions in UIM?
- What happens when a position is ended in ESR in relation to access rights via linked access control positions?
- Do odds ratios really control for the availability of occupational positions in status contingency tables?
