Quality & Quantity 33: 229–242, 1999.
© 1999 Kluwer Academic Publishers. Printed in the Netherlands.
Why Do Cohabiting Couples Marry? An Example
of a Causal Event History Approach to
, KATHARINA POHL
Fakultät für Soziologie, Bielefeld University, Germany;
Economic Commission for Europe,
United Nations, Geneva, Switzerland;
Bundesinstitut für Bevölkerungsforschung, Wiesbaden,
Fakultät für Sozialwissenschaft, Ruhr University of Bochum, Germany
Abstract. The purpose of this paper is to demonstrate a causal approach to interdependent systems
based on two empirical investigations. These examples demonstrate (1) the study of two highly
interdependent processes: entry into ﬁrst marriage as the dependent process and the process of ﬁrst
birth/ﬁrst pregnancy as the explaining one; (2) an interdependence occurring mainly in a very speciﬁc
phase of individuals’ lives (i.e. during the period of ﬁrst family formation); (3) the involvement of
time lags between cause and its effect (e.g. time until detection of conception); and (4) the highly
dynamic character of an unfolding effect over time (i.e. the effect of ﬁrst pregnancy/ﬁrst birth on ﬁrst
marriage strongly depends on the progress of pregnancy and the time since the birth has taken place).
Key words: interdependent processes, simultaneity problems, causal analysis, event history analysis,
longitudinal data, fertility, nuptiality.
The analysis of parallel processes is an important but difﬁcult scientiﬁc endeavour
in the social sciences. In these studies the methodological issue of reverse cau-
sation often arises. Reverse causation means that one process, considered from a
theoretical point of view as the dependent one, has (direct or indirect) effects on
another process, considered as the independent one.
In this paper, we study the effects of ﬁrst pregnancy/ﬁrst childbirth on the rate of
entry into ﬁrst marriage for couples living in consensual unions in Germany. Both
processes form a highly interdependent dynamic system. In the analysis of this
interdependent system we are using a causal approach as described in the book by
Blossfeld and Rohwer (1995). This approach provides a straightforward solution
to (1) the simultaneity problem of interdependent processes, (2) the identiﬁcation
of lags between causes and their effects, and (3) the study of temporal shapes of