October 29, 2025 | News | Marriage or Separation

How Age and Cohabitation Duration Modulate Partnerships Transitions

New statistical model helps understand transitions from partnerships to weddings or breakups

Scientists have developed a model to analyze transitions from non-marital cohabitation to either marriage or separation considering two-time scales simultaneously: the age of the individuals and the duration of the cohabitation.

A man and a woman are standing in a kitchen cooking together. They are talking. They are a couple.

Using a new statistical method, researchers at the MPIDR investigated how transitions to marriage or separation occur. For the first time, they considered two time scales. © Novak – stock.adobe.com

At what point do couples decide to either get married or break up? Does the length of the relationship alone suffice, or does the age of the partners also play a role? Scientists at the Max Planck Institute for Demographic Research (MPIDR), the Leiden University Medical Center, and Erasmus University Rotterdam Medical Center applied a two-time-scales hazard model to answer this question.

"We investigated how transition rates from cohabitation to marriage or dissolution of the partnership depend on two-time scales: age and cohabitation duration. Additionally, we discuss the analysis of time-to-event data with multiple time scales in sociodemographic processes more generally," explains Angela Carollo, an MPIDR researcher and the study's first author. The research team used data from the German Family Panel (pairfam), a panel analysis of intimate relationships and family dynamics.

The study analyzes transitions from non-marital cohabitation to marriage or separation, over the age of the individuals and the duration of their cohabitation simultaneously. Additionally, it makes a distinction according to the gender of the individuals and their place of residence eastern vs. western Germany. "We find that transitions from cohabitation to marriage depend on age and cohabitation duration in complex ways. In contrast, transitions from cohabitation to separation depend much more simply on the age of the individuals and the duration of cohabitation," says Carollo.

Figure: Hazard of marriage by age and duration of the cohabitation. East German men (top)  West German men (bottom)

© MPIDR

West German men who enter into cohabitation between the ages of 25 and 30 experience an increasing marriage rate during the first years of cohabitation, peaking approximately five years after the start of cohabitation. Women and men living in eastern Germany, on the other hand, marry much less frequently, and with different rates over age and cohabitation duration, than those living in western Germany.

The transition rates from cohabitation to dissolution of the partnership are more uniform. They are highest among people who start cohabiting before the age of 20 and within a few years of starting to live together. As the duration of cohabitation increases and the age of the individuals rises, the dissolution rates steadily decline.

Previous research has typically modeled marriage or dissolution rates over the duration of cohabitation. Age then is predominantly included in coarsened form only. “With our approach, we can improve current models by treating these time scales flexibly and assigning equal importance to both in the risk model," Carollo explains. "Examples of processes that develop over more than one time scale are abundant in sociodemographic research. The ability to consider the effects of multiple time scales for the same process flexibly makes it possible to model these processes with fewer restrictions and opens up the opportunity for new insights."

Original Publication

Angela Carollo, Hein Putter, Paul H.C. Eilers, Jutta Gampe: Analysis of Time-to-Event Data With Two Time Scales. An Application to Transitions out of Cohabitation in Sociological Methods & Research (2025); DOI: 10.1177/00491241251374193

Carollo, A.; Eilers, P. H. C.; Gampe, J.:
Software. The Comprehensive R Archive Network: CRAN. (2024)

Keywords

Time-to-event data; Bivariate hazard smoothing; P-splines; Cohabitation; Marriage; Separation

Contact

Head of Public Relations and Communications

Silvia Leek

E-Mail

+49 381 2081-143

Science Communication Editor

Christine Ruhland

E-Mail

+49 381 2081-157

Science Communication Editor

Silke Schulz

E-Mail

+49 381 2081-153

What next?

To the Home Page

Max-Planck-Gesellschaft - Logo
The Max Planck Institute for Demographic Research (MPIDR) in Rostock is one of the leading demographic research centers in the world. It's part of the Max Planck Society, the internationally renowned German research society.