ODISSEI LEZING: How predictable are fertility outcomes? Introducing the PreFer data challenge and its potential for fertility research

Op dinsdag 12 december tussen 14-15 uur zal Gert Stulp, universitair hoofddocent bij de afdeling Sociologie van de Rijksuniversiteit Groningen, presenteert de Predicting Fertility data challenge (PreFer). PreFer zal zich richten op de vraag of mensen kinderen hebben – met behulp van twee unieke datasets: het LISS-panel, een Nederlands longitudinaal onderzoek en gegevens uit het Nederlandse bevolkingsregister. Na de lezing is er ruimte voor vragen en discussie. Meer informatie over de lezing hieronder (Engels).

About this Lecture

In a data challenge, (teams of) researchers using the same dataset try to best predict a particular outcome. In several fields like computer vision, natural language processing, and healthcare, prediction data challenges led to substantial progress both in model development and problem understanding. However, data challenges and the focus on predictability are still rare in the social sciences. 

In this talk, we present the Predicting Fertility data challenge (PreFer), a collaborative effort between the University of Groningen, ODISSEI, Eyra, and Centerdata, and the first data challenge in demography. Its central goal is to establish the current predictability of fertility outcomes – whether people have children – using two unique datasets: the LISS panel, a Dutch longitudinal survey, and Dutch population registry data. 

We will discuss the design of the data challenge and how this data challenge can produce novel and robust insights into an important life outcome and advance fertility research. The challenge will be open to any interested (team) of researchers and during the talk we will also provide practical details (timeline, data, evaluation, etc.) for those who are interested in joining it.  

If you’d like to learn more about PreFer and to be notified when the registration for the challenge opens, please leave your email address here.


Gert Stulp is an associate professor at the department of Sociology at the University of Groningen. He studies causes of the variation in the number of children people have and would like to have, and employs diverse methods in his research including personal network data collection, simulation studies, and machine learning.


The ODISSEI Lunch Lectures highlight methodological issues and innovations in Social Science.

Photo by Lee Campbell on Unsplash