Tilburg University | Tilburg School of Social and Behavioral Sciences is looking for a PHD candidate PhD candidate in Machine Learning-Informed Formal Theory Construction (Methodology and Statistics)
Department: Methodology and Statistics
Location: Tilburg
Contract size: 1.0 fte (40 hours per week)
Full-time gross monthly salary: €2.901 – €3.707
Contract duration: 12 months with a possible extension of 36 months.
Desired starting date: September 2025
Do you want to shape the future of social scientific research? Then join our team as a PhD student to co-develop machine learning-informed methods for theory construction.
Theories describe scientists’ understanding of phenomena. Ideally, theories are used to derive hypotheses, and continuously updated based on new insights. In practice, many scientific fields focus near-exclusively on conducting empirical studies, skipping the important step of revising theory based on the ressults. In this PhD project, you will develop methods to help applied (social) scientists construct theories based on data patterns identified with causal discovery and interpretable machine learning.

YOUR PROFILE

We are looking for a PhD candidate with a strong background in machine learning, data science, or applied/mathematical statistics and interest in philosophy of science. Interdisciplinary candidates, especially with strong quantitative skills and a background in social science or philosophy of science, can also apply.

Other requirements include:

• A (nearly) completed (research) master’s thesis which, preferably, involves machine learning, a simulation study, and/or pertains to theory development.
• Programming skills in R, Python, or C++
• Research skills and data analytical abilities.
• Communication and cooperation skills and the willingness to work in a team.
• Project management and organization skills.
• Interest in open science and team science.
• Proficiency in English, including academic writing.
• Interest in providing small-scale education, such as teaching working groups or bachelor thesis supervision.

YOUR RESPONSIBILITIES

• Conducting empirical and methodological research
• Developing user-friendly open source research software (in R and, optionally, Python)
• Preparing scientific articles for publication in international journals (ideally open access), present key findings at national and international scientific conferences, and write a dissertation that connects your scientific articles.
• Active participation in the Department of Methodology and Statistics, Theory Methods Society, and optionally, your choice of other professional organizations, including the Tilburg Meta-Research Center, Open Science Community, Paul Meehl Graduate School, and the Interuniversity Graduat
• School of Psychometrics and Sociometrics (IOPS).
• Participating in collaborations with applied researchers in the areas of (mental) healthcare, climate science, adolescents’ emotional development and substance (ab)use, cooperation, and morality, or new collaborations established based on your interests.
• Contributing to education and supervision (e.g. supervising bachelor’s theses and research skills groups; not more than 10% of your time will be devoted)

As a PhD candidate, you will conduct independent empirical and methodological research in the areas of interpretable machine learning and causal discovery. You will work towards the goal of developing and validating a workflow for constructing formal theories from patterns in empirical data. Each year, you will address a different milestone in pursuit of this goal:

• Benchmarking causal discovery methods on social science data
• Comparing the performance of different interpretable machine learning methods on social science data
• Developing the basic functionality for machine learning-informed theory construction as open source software
• Extending the basic functionality in one of several possible directions, based on your interest