Welcome to the website of the Predicting Fertility data challenge (PreFer). The aim of the challenge is to measure current predictability of fertility outcomes in the Netherlands to advance our understanding of fertility

The challenge has now ended. We are analysing the results and will be sharing updates on this website. If you have questions about PreFer, please check the FAQ or contact us

About PreFer

Important dates

  • Application deadline: March 24, 2024. This deadline has passed and the application is now closed.
  • Start of the data challenge Phase 1: April 1, 2024
  • Selection into Phase 2: June 3 - June 17, 2024
  • Start of Phase 2: June 17, 2024
  • End of Phase 2 and Phase 1: September 16, 2024
  • Announcing the winners: November 2024

Deadlines for submissions:

  • April 22nd, 12:00pm (afternoon) CEST (intermediate leaderboard)
  • May 13, 12:00pm (afternoon) CEST (intermediate leaderboard)
  • June 3, 12:00pm (afternoon) CEST (intermediate leaderboard)
  • July 29, 12:00pm (afternoon) CEST (intermediate leaderboard)
  • September 16, 12:00pm (afternoon) CEST - final deadline for submissions based on survey data
  • October 25, 17:00 CET - final deadline for submissions based on register data

Why predict fertility outcomes in a data challenge?

Fertility is widely studied in diverse disciplines due to its importance to individuals and societies. A lot of factors have been identified that are related to fertility outcomes. Yet these important factors only explain a fraction of the variation in fertility outcomes and we are unable to explain even their short-term changes. What do we miss?

This data challenge can potentially advance our understanding of fertility behavior and improve social policies and family planning in several ways. Measuring how well different factors and models can predict fertility outcomes for new cases will show which factors are more important. It can narrow down a scope for potential interventions and help people reach their desired family size. Comparing and interpreting different models submitted to a data challenge (e.g. theory- and data-driven) can identify new factors currently overlooked by the theories of fertility and highlight the gaps in current knowledge (e.g. important interactions or non-linear effects).

Why participate

  • Use unique data from the LISS panel and the Dutch population registers, including a social network for the entire Dutch population*
  • Have fun learning new skills, competing with other teams, and discovering unique data while predicting an important life outcome
  • Contribute to research on fertility behaviour
  • Collaborate on scientific papers describing the results of the challenge
  • Win recognition and prizes

*Only a selection of participants will get an opportunity to work with the register data in Phase 2 of the data challenge. For these participants the costs of access to the register datasets will be covered by ODISSEI. See conditions for accessing the register data here.

How to participate

Apply

Fill out this form to apply for participation in the data challenge. You can find the details about applications here. The application deadline is March 24th.

Build a model

Use any method you’d like to train a predictive model using training data we provide. The models will be then evaluated on a holdout set based on the F1 score. More information about the phases of the data challenge can be found here.

Submit the model

Submit your model (in Phase 1) or predictions (in Phase 2). The link the submission platform with all the instructions how to submit will appear on this website soon.

Ethical approval

This research was approved by the ethical committee of sociology at the University of Groningen (SOC-2425-S-0002).