The Demographics of Labour Turnover: A Comparison of Ordinal Probit and Censored Count Data Models
- 1 January 1992
- journal article
- Published by CAIRN.INFO in Recherches économiques de Louvain
- Vol. 58 (3-4) , 283-306
- https://doi.org/10.1017/s0770451800044080
Abstract
Summary: As has been found in previous studies, the labor market performance of individuals is often affected by demographic determinants like cohort size, age, marriage status and family size. While most of this analysis was studied for earnings, the paper investigates the issue for labor mobility. Mobility is measured here by the number of new employers and the frequency of unemployment of an individual in a particular period. Given the discrete nature of the data, the ordinal probit model and the censored Poisson as the censored negative binomial model was estimated. Since the choice of the statistical model is not clear a priori, various model comparisons are carried out and some new pseudo-R2 measures are proposed and used in the analysis. Results indicate that demographic determinants matter for labor mobility.Keywords
This publication has 13 references indexed in Scilit:
- Pseudo-R2's in the ordinal probit modelThe Journal of Mathematical Sociology, 1992
- A Theory of Quits and Layoffs with Efficient TurnoverJournal of Political Economy, 1991
- Regression-based tests for overdispersion in the Poisson modelJournal of Econometrics, 1990
- Labour Turnover in UK ManufacturingEconomica, 1990
- Job Performance, Turnover, and Wage GrowthJournal of Labor Economics, 1990
- Quit Rates over Time in a Job-Rationed Labour Market: The British Manufacturing Sector, 1971-1983Economica, 1988
- A Tobit-type estimator for the censored Poisson regression modelEconomics Letters, 1985
- THE ACCUMULATION OF HUMAN CAPITAL DURING MARRIAGE BY MALESEconomic Inquiry, 1983
- A statistical model for the analysis of ordinal level dependent variablesThe Journal of Mathematical Sociology, 1975
- Turnover in the Labor ForceBrookings Papers on Economic Activity, 1972