5. Conclusions

Dimitris Pavlopoulos and Jeroen K. Vermunt

Previous

In this paper, we investigated the measurement error in the type of the employment contract in the Dutch LFS by matching its longitudinal component from 2007 and early 2008 with a unique register dataset, the PA. We applied several hidden Markov models, in which the true contract type is treated as a latent state and in which the survey and register information serve as observed indicators of an individual’s true contract. We modeled the measurement error in the two data sources by taking into account that the error in the register is correlated across occasions.

Our results show that the register data contain more error than the survey data, and therefore cannot be used as a golden standard. However, the improvement of the initial registration in the register data can significantly improve their quality as measurement error in the indicator of the contract type that comes from this dataset is serially correlated.

The measurement error results into an underestimation of the percentage of individuals that are working on a temporary contract. In the LFS this percentage is 8.9%, whereas after correction for measurement error this percentage rises to 10.9%. Another effect of measurement error is that it yields severely overestimated transition probabilities. According to the LFS and PA, the transition probability between temporary to permanent employment in a three-month period is 5.7% and 8.5%, respectively, whereas the corresponding latent transition probability is only 3.2%. This finding is particularly important for Dutch policy makers as it clearly indicates that there is much less mobility from temporary to permanent employment than originally thought.

The results of this study remain fairly stable across the model specifications that we tested. This shows that the results are robust to small model misspecifications. However, results remain somehow dependant on model assumptions. Further sensitivity tests and applications can further verify the validity of our results. Future research may focus particularly on sensitivity tests with the use of Monte Carlo simulations.

Acknowledgements

The authors are thankful to Statistics Netherlands for providing access to the data of this article. The authors are also thankful to Frank van der Pol, Wendy Smits, Ruben van Gaalen and to the participants of the ESPE and EALE conferences as well as to the participants of the SILC research group of the VU University Amsterdam for the useful comments and suggestions. The contribution of Jeroen Vermunt was supported by the Netherlands Organization for Scientific Research (NWO) [VICI grant number 453-10-002].

References

Abowd, J.M., and Stinson, M.H. (2005). Estimating measurement error in SIPP annual job earnings: A comparison of census survey and SSA administrative data. Technical Paper, U.S. Census Bureau.

Bartolucci, F., Lupparelli, M. and Montanari, G.E. (2009). Latent markov model for longitudinal binary data: An application to the performance evaluation of nursing homes. Annals of Applied Statistiscs, 3(2), 611-636.

Bassi, F., Hagenaars, J.A., Croon, M.A. and Vermunt, J.K. (2000). Estimating true changes when categorical panel data are affected by uncorrelated and correlated classification errors. Sociological Methods and Research, 29(2), 230-268.

Baum, L.E., Petrie, T., Soules, G. and Weiss, N. (1970). A maximization technique occuring in the statistical analysis of probabilistic functions of markov chains. The Annals of Mathematical Statistics, 41(1), 164-171.

Bentolila, S., and Bertola, G. (1990). Firing costs and labour demand: How bad is eurosclerosis? The Review of Economic Studies, 57(3), 381-402.

Biemer, P. (2004). An analysis of classification error for the revised Current Population Survey employment questions. Survey Methodology, 30, 2, 127-140.

Biemer, P. (2011). Latent Class Analysis of Survey Error. New Jersey: John Wiley & Sons, Inc.

Biemer, P.P., and Bushery, J.M. (2000). On the validity of markov latent class analysis for estimating classification error in labor force data. Survey Methodology, 26, 2, 139-152.

Bollinger, C.R. (1996). Bounding mean regressions when a binary regressor is mismeasured. Journal of Econometrics, 73(2), 387-399.

Booth, A.L. (1997). An analysis of firing costs and their implications for unemployment policy. In Unemployment Policy, (Eds., D.J. Snower and G. de la Dehesa). Cambridge: Cambridge University Press.

Bound, J., Brown, C., Duncan, G.J. and Rodgers, W.L. (1994). Evidence on the validity of cross-sectional and longitudinal labor market data. Journal of Labor Economics, 12(3), 345-368.

Bound, J., Brown, C. and Mathiowetz, N. (2001). Measurement error in survey data. In Handbook of econometrics, (Eds., J.J. Heckman and E. Leamer), Amsterdam: Elsevier, 5, 3705-3843.

Brown, C., and Medoff, J.L. (1996). Employer characteristics and work environment. Annales D’Économie et de Statistique, 41, 275-298.

Cahuc, P., and Postel-Vinay, F. (2002). Temporary jobs, employment protection and labor market performance. Labour Economics, 9(1), 63-91.

Dempster, A.P., Laird, N.M. and Rubin, D.B. (1977). Maximum likelihood from incomplete data via the EM algorithm. Journal of the Royal Statistical Society, Series B, 39(1), 1-38.

Duncan, G.J., and Hill, D.H. (1985). An investigation of the extent and consequences of measurement error in labor-economic survey data. Journal of Labor Economics, 3(3), 508-522.

Gottschalk, P. (2005). Downward nominal-wage flexibility: Real or measurement error. Review of Economics and Statistics, 87(3), 556-568.

Gottschalk, P., and Huynh, M. (2010). Are earlings inequality and mobility overstated? The impact of non-classical measurement error. Review of Economics and Statistics, 92(2), 302-315.

Hagenaars, J.A. (1990). Categorical Longitudinal Data Log-Linear Panel, Trend and Cohort Analysis. Newbury Park, CA: Sage Publications.

Hagenaars, J.A. (1994). Latent variables in log-linear models of repeated observations. In Latent Variable Analysis: Applications for Developmental Research, (Eds., A. von Eye and C.C. Clogg). Thousand Oaks, CA: Sage Publications, 329-352.

Heckman, J.J., and Singer, B.L. (1984). A method for minimising the impact of distributional assumptions in econometric models for duration data. Econometrica, 52 (2), 271-320.

Hilbers, P., Houwing, H. and Kösters, L. (2011). De flexibele schil – overeenkomsten en verschillen tussen CBS- en UWV-cijfers [the flexible periphery – similarities and differences between CBS and UWV-data]. In Sociaaleconomische Trends, 2^e Kwartaal 2011 [Socioeconomic Trends, 2^nd Trimester 2011], (Eds., B. Hermans et al.). Den Haag/Heerlen: Statistics Netherlands, 26-33.

Kapteyn, A., and Ypma, J.Y. (2007). Measurement error and misclassification: A comparison of survey and register data. Journal of Labor Economics, 25(3), 513-551.

Langeheine, R. (1994). Latent variable markov models. In Latent Variables Analysis. Applications for Developmental Research, (Eds., A. von Eye and C. Clogg). Thousand Oaks, California: Sage Publications, 373-395.

Manzoni, A., Vermunt, J.K., Luijkx, R. and Muffels, R. (2010). Memory bias in retrospectively collected employment careers: A model-based approach to correct for measurement error. Sociological Methodology, 40(1), 39-73.

Mars, G. (2011, December). Cijfers over Flexibele Arbeidsrelaties - Confrontatie Van Bronnen en Defiities [Figures on Flexible Labour Relations - Confrontation of Sources and Definitions]. Statistics Netherlands, report nr SAH-2011-H11. The Hague/Heerlen.

Mathiowetz, N.A. (1992). Errors in reports of occupations. Public Opinion Quarterly, 56(3), 352-355.

OECD (2002). Employment Outlook 2002. Paris: Author.

OECD (2012). Country Statistical Profiles. OECD Database: retrieved on 2012/12/16 from http://stats.oecd.org/.

Paas, L.J., Vermunt, J.K. and Bijmolt, T.H. (2007). Discrete-time discrete-state latent markov modelling for assessing and predicting household acquisitions of financial products. Journal of the Royal Statistical Society, Series A, 170(4), 955-974.

Pavlopoulos, D., Muffels, R. and Vermunt, J.K. (2012). How real is mobility between low pay, high pay and non-employment. Journal of Royal Statistical Society, Series A, 175(3), 749-773.

Pischke, J.-S. (1995). Measurement error and earnings dynamics: Some estimates from the PSID validation study. Journal of Business and Economic Statistics, 13(3), 305-314.

Poulsen, C.S. (1990). Mixed markov and latent markov modelling applied tobrand choice behaviour. International Journal of Research in Marketing, 7(1), 5-19.

Rendtel, U., Langeheine, R. and Berntsen, R. (1998). The estimation of poverty dynamics using different measurements of household income. Review of Income and Wealth, 44(1), 81-98.

Rodgers, W.L., Brown, C. and Duncan, G.J. (1993). Errors in survey reports of earnings, hours worked, and hourly wages. Journal of the American Statistical Association, 88(3), 345-368.

Sels, L., and Van Hootegem, G. (2001). Seeking the balance between flexibility and security: A rising issue in the low countries. Work, Employment and Society, 15(2), 327-352.

Shorrocks, A.F. (1976). Income mobility and the markov assumption. Economic Journal, 86, 566-578.

Skinner, C.J., Holt, D. and Smith, T.M.F. (1989). Analysis of Complex Surveys. Wiley.

van der Pol, F., and Langeheine, R. (1990). Mixed markov latent class models. Sociological Methodology, 20, 213-247.

Vermunt, J.K. (1997). Log-Linear Models for Event Histories. London: SAGE publications.

Vermunt, J.K., and Magidson, J. (2008). LG - Syntax User’s Guide: Manual for Latent GOLD 4.5 Syntax Module. Belmont Massachussetts: Statistical Innovations Inc.

Vermunt, J.K., Langeheine, R. and Böckenholt, U. (1999). Discrete-time discrete-state latent markov models with time-constant and time-varying covariates. Journal of Educational and Behavioral Statistics, 24, 178-205.

Vermunt, J.K., Tran, B. and Magidson, J. (2008). Latent class models in longitudinal research. In Handbook of Longitudinal Research: Design, Measurement, and Analysis, (Ed., S. Menard). Burlington, MA: Elsevier, 373-385.

Previous