Quality & Quantity 33: 157–168, 1999.
© 1999 Kluwer Academic Publishers. Printed in the Netherlands.
Measuring Career Mobility: An Empirical
Comparison of Six Mobility Indexes
TOON W. TARIS
and JAN A. FEIJ
Psychology & Health/Utrecht University;
Kurt Lewin Institute/Free University, Amsterdam
Abstract. The current paper presents six indexes that can be used to characterize the course of a
career during a particular time interval, respectively, (1) the total number of transitions during that
interval; (2) the number of positive transitions; (3) the number of negative transitions; (4) the sub-
traction of the number of negative transitions from the number of positive transitions; (5) the relative
uncommonness of the transitions; and (6) the subtraction of the number of negative transitions from
the number of positive transitions, weighted by their uncommonness. Advantages and disadvantages
of these six indexes are discussed. Further, an empirical example is presented that draws on data from
a sample of 357 employed Dutch youth. Finally, our approach is compared to previous approaches
(event-centered methods, such as survival analysis, and career-centered methods, such as clustering
techniques). It is concluded that our simple approach complements these other approaches well.
Key words: longitudinal data analysis, event history data, career development.
The last two decades have witnessed a rapid increase in the interest in longitudinal
data collection and analysis (Dijkstra & Taris, 1995). The data often take the form
of event histories (e.g., Blossfeld, Hamerle & Mayer, 1989; Yamaguchi, 1994),
consisting of sequences of states occupied by the sampling units during a particular
observation period, as well as the timing of transitions from one state to another
(e.g., from married to unmarried, or vice versa). Although event history data can
be collected on many topics and for many types of sampling units (e.g., persons,
business ﬁrms, research projects), for simplicity we refer only to event histories of
As regards the analysis of event history-data, one may distinguish between
event-centered and career-centered methods. In event-centered analysis, researchers
focus on the occurrence and recurrence of one particular type of event, e.g., from
employment to unemployment, from married to single, etc. The sequence of tran-
sitions is conveniently split in its constituent episodes, during which the person
belonged to the state of interest. Two typical modes of analysis are survival analysis
Author for correspondence: T.W. Taris, Utrecht University, Department of Social and Organiza-
tional Psychology, P.O. Box 80.140, NL-3508 TC Utrecht, The Netherlands. Telephone +31-30-253
4781, fax +31-30-253 7584, email: firstname.lastname@example.org. This study was partly supported by the Dutch
Organization for Scientiﬁc Research NWO, Grant 580-02.207, as part of the research program
‘Fatigue at Work’.