Lifetime Data Analysis

, Volume 9, Issue 4, pp 311–330 | Cite as

The Perpetual Student: Modeling Duration of Undergraduate Studies Based on Lifetime-Type Educational Data

  • Aglaia G. Kalamatianou
  • Sally McClean


It is important to educational planners to estimate the likelihood and time-scale of graduation of students enrolled on a curriculum. The particular case we are concerned with, emerges when studies are not completed in the prescribed interval of time. Under these circumstances we use a framework of survival analysis applied to lifetime-type educational data to examine the distribution of duration of undergraduate studies for 10,313 students, enrolled in a Greek university during ten consecutive academic years. Non-parametric and parametric survival models have been developed for handling this distribution as well as a modified procedure for testing goodness-of fit of the models. Data censoring was taken into account in the statistical analysis and the problems of thresholding of graduation and of perpetual students are also addressed. We found that the proposed parametric model adequately describes the empirical distribution provided by non-parametric estimation. We also found significant difference between duration of studies of men and women students. The proposed methodology could be useful to analyse data from any other type and level of education or general lifetime data with similar characteristics.

duration of studies lifetime-type educational data survival analysis non-parametric and parametric estimation modified goodness-of-fit procedure 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. P. D. Allison, “Disrete-time methods for analysis of event histories,” in Sociological methodology, (S. Leinhardt, ed.) Jossey-Bass: San-Francisco, pp. 61–98, 1982.Google Scholar
  2. R. W. Blischke and D. N. P. Murthy, Reliability, Wiley: New York, 2000.Google Scholar
  3. H.-P. Blossfeld, A. Hamerle, and K. U. Mayer, Event History Analysis. Statistical Theory and Application in the Social Sciences, Lawrence Erlbaum Associates, Inc.: Hillsdale, New Jersey, 1989.Google Scholar
  4. L. A. Booth and S. E. Satchell, “The hazards of doing PhD: An analysis of competion and withdrawal rates of British PhD students in the 1980s,” J. R. Statist. Soc. A. vol. 158, part2 pp. 297–318, 1995.Google Scholar
  5. N. E. Breslow, “A generalized Kruskal-Wallis test for comparing K samples subject to unequal patterns of censorship,” Biometrica vol. 57 pp. 579–594, 1970.Google Scholar
  6. D. Collett, Modelling Survival data in Medical Research, Chapman & Hall, 1994.Google Scholar
  7. D. R. Cox and D. Oakes, Analysis of Survival Data, tChapman & Hall, 1984.Google Scholar
  8. A. P. Dempster, N. M. Laird, and D. B. Rubin, “Maximum likelihood from incomplete data via the EM algorithm (with discussion),” J. R. Statist. Soc. B, vol. 39 pp. 1–38, 1977.Google Scholar
  9. L. Fahrmeir, “Dynamic modelling and penalized likelihood estimation for discrete time survival data,” Biometrica vol. 81 pp. 317–330, 1994.Google Scholar
  10. M. Greenwood, The Natural Duration of Cancer. Reports on Public health and Medical Subjects, Her Majesty's stationery office: London, vol. 33 pp. 1–26, 1926.Google Scholar
  11. J. O. Gribbin and S. I. McClean, “Modelling the duration of spells of withdrawal from a female dominated labour force,” Journal of Applied Statistics vol. 17 pp. 369–381, 1990.Google Scholar
  12. A. Hamerle and G. Tutz, Diskrete modelle zur analyse von verweildauern und lebenszeiten, Frankfurt Campus Verlag: New York, 1989.Google Scholar
  13. M. Hollander and F. Proschan, “Testing to determine the underlying distribution using randomly censored data,” Biometrics vol. 35 pp. 393–401, 1979.Google Scholar
  14. J. M. Jamshidian and R. I. Jennrich, “Acceleration of the EM algorithm by using quasi-Newton methods,” J. R. Statist. Soc. B, vol. 59 pp. 569–587, 1997.Google Scholar
  15. P. N. Jewell, C. A. Kimber, M.-L. Ting Lee, and A. G. Whitmore, eds., Lifetime Data: Models in Reliability and Survival Analysis, Kluwer Academic Publishers, 1996.Google Scholar
  16. A. G. Kalamatianou, C. A. Karmas, and T. P. Lianos, “Techical higher education in Greece,” European Journal of Education vol. 23 pp. 271–279, 1988.Google Scholar
  17. A. G. Kalamatianou, “Using life table method on a censored data set for estimating duration of studies in University Departments,” in Bulletin of the International Statistical Institute, 52ndSession. Contributed papers, Tome LVIII, Book2, pp. 101–102, 1999.Google Scholar
  18. A. G. Kalamatianou and S. McClean, “A stratified non-parametric survival model for describing the distribution of duration of studies in various university departments,” in Statistical Modeling, Proceedings of the 15 th International Workoshop on Statistical Modeling, New Trends in Statistical Modeling, (V. Nunez-Anton and E. Ferreira, eds.) The University of the Basque Country, pp. 441–452, 2000.Google Scholar
  19. A. G. Kalamatianou and S. McClean, “A cohort-stratified non-parametric survival model for duration of under-graduate studies,” in Applied Stochastic Models and Data Analysis, (G. Govaert, J. Janssen, and N. Limnios, eds.) Universite de Technologie de Compiegne: France, vol. 2/2 pp. 597–603, 2001.Google Scholar
  20. J. D. Kalbfleisch and L. R. Prentice, The Statistical Analysis of Failure Time Data, John Wiley & Sons Inc: New York, 1980.Google Scholar
  21. E. L. Kaplan and P. Meier, “Nonparametric estimation from incomplete observations,” J. Amer. Stat. Assn. vol. 53 pp. 457–481, 1958.Google Scholar
  22. C. A. Karmas, T. P. Lianos, and A. G. Kalamatianou, “Greek Universities: An overview,” European Journal of Education vol. 23 pp. 261–269, 1988.Google Scholar
  23. J. P. Klein and M. Moeschberger, Survival Analysis, Springer Verlang: New York, 1997.Google Scholar
  24. C. T. Le, Applied Survival Analysis, John Wiley and Sons, Inc.: New York, 1997.Google Scholar
  25. E. T. Lee, Statistical Methods for Survival Data Analysis, second edition, John Wiley and Sons, Inc.: New York, 1992.Google Scholar
  26. L. M. Leemis, Reliability Probabilistic Models and Statistical Methods, Prentice-Hall, Inc.: Englewood Cliffs, New Jersey, 1995.Google Scholar
  27. R. A. Maller and X. Zhou, Survival Analysis with Long-Term Survivors, John Wiley and Sons: Chechester, 1996.Google Scholar
  28. N. Mantel and W. Haenszel, “Statistical aspects of the analysis of data from retrospective studies of disease,” Journal of the National Cancer Institute vol. 22 pp. 719–748, 1959.Google Scholar
  29. E. Marubini and M. G. Valsecchi, Analysing Survival Data from Clinical Trials and Observational Studies, John Wiley & Sons, 1995.Google Scholar
  30. G. Molenberghs and E. Goetghebeur “Simple fitting algorithms for incomplete categorical data,” J. R. Statist. Soc. B, vol. 59 pp. 401–414, 1997.Google Scholar
  31. W. Nelson, Applied Life Data Analysis, John Wiley & Sons: New York, 1982.Google Scholar
  32. R. R. Regal and K. Larntz, “Likelihood methods for testing group problem solving models with censored data,” Psychometrica vol. 45 pp. 353–366, 1978.Google Scholar
  33. T. H. Scheike and T. Kold Jensen, “A discrete survival model with random effets: An application to time pregnancy,” Biometrics vol. 53 pp. 318–329, 1997.Google Scholar
  34. C. F. J. Wu, “On the convergence properties of the EM algorithm,” The Annals of Statistics vol. 11 no.1 pp. 95–103, 1983.Google Scholar

Copyright information

© Kluwer Academic Publishers 2003

Authors and Affiliations

  • Aglaia G. Kalamatianou
    • 1
  • Sally McClean
    • 2
  1. 1.Department of SociologyPanteion University of AthensAthensGreece
  2. 2.School of Information and Software EngineeringUniversity of UlsterColeraine, Northern IrelandUK

Personalised recommendations