Dominique van Roon, Statistics Netherlands
Carel Harmsen, Statistics Netherlands
Many couples start living together or decide to break up without an official marriage or divorce. However, today’s statistics based on the Dutch population register only present legal marriages and divorces. A database containing information derived from registers on start and end dates of all cohabitation relationships is not readily available yet. Using register-based data instead of sample surveys has major advantages, including the possibility to study small subgroups and relationships over time at the individual level. Furthermore, by linking this file to other registers, an abundance of additional research opportunities will be created. One disadvantage is that married and unmarried couples that do not cohabit still remain concealed. Besides being able to present data with the annual total of new and ended cohabitation relationships in the Netherlands, we aim to provide a sound basis for researchers to study, for instance, the effects of ending a cohabitation relationship, regardless of whether the couple was married or not; the chance of a new relationship within subpopulations; and the average time interval between ending one relationship and starting a new one. We have therefore committed ourselves to building such a database, assuming it is possible by using address and family information as well as data on changes in household composition. We would here like to discuss the methods we have used and present thorny outstanding problems, one of the largest being how to deal with the fact that not all people running a household together are in a cohabitation relationship (e.g. students living together in a dormitory). How do we distinguish ‘romantic’ relationships from purely economic ones? Of course we will also go into the solved problems, such as how to deal with administrative noise and how to divide households with two or more generations into the relevant couples.
Presented in Session P1. Poster Session 1