Promotion to full professor data quality

A bit of personal news: I have recently changed jobs. I am actually still doing the same thing as before: The people at Utrecht University were so kind to appoint me as professor of data quality. I will stay at the dept. of Methods and Statistics , and will continue teaching, doing management and research. If you do not know what the department is doing, do check out the link above; I am fortunate to have been working with incredibly smart and kind colleagues over the past years, and the kinds of projects that everyone at my department are doing really matter for science, for society and for individuals.

Methods & Statistics may not the area that brings cheers of excitement on birthday parties, but especially in the world of today, having good data, and knowing what to do (and not do!) with these data is more important than ever before. Data are everywhere; society is increasingly becoming datafied because of the ongoing digitalization of everything. Data-driven policy making (or evidence-based policy making) is talked about a lot, and rightly so. I feel there are lots of topics and areas where more or better data can really improve society, and there are still a lot of areas where policies are designed based on anecdotal evidence and personal experiences of policy-makers. How should we increase equality of opportunity in our educational system? How do we use customer feedback to really make service better? How can we produce statistics on e.g. inflation quicker so we can adapt our policies?

One critical issue is that having data alone is not enough. In the field of education, kids are nowadays tested more than ever. One cannot visit a hotel, or buy a product online, without receiving at least one feedback form. Prices can be tracked in real time; sensors can count the exact number of steps one takes on a day. But how do we make sure the data that we get are ‘good’ data, and how do we interpret, aggregate, and compare data in such a way that we can use these data?

While digitization brings about more and more data, there are reasons to worry about the quality of data. It is not only hotels that have trouble getting their customers to complete feedback forms. Nonresponse is increasingly a problem in social research generally. Missing data are pervasive in data streams. And while digital behavioral data from e.g. social media are incredibly rich, these data contain so much noise, it is hard to extract what you actually want to know about. Moreover, there is always the question of what data to use, and what data not to use in a world of abundant data.

These are the kinds of issues I work on. I am currently working on lots of cool projects that aims to tackle one, or several of these questions. For example, we are wrapping up a project on Smart Survey Implementation that aims to develop a methodology for integrating sensor data into surveys, including how to recruit respondent, use machine learning, UX/UI issues and data integration. Another ODISSEI SSHOC-NL project I am working on with Anne-Chiao Liu focuses on how to link data to eachother when you have problems with selection or measurement errors in one or both of the data sources.

I will also continue do more applied projects, on how to improve questionnaires, limit nonresponse, or draw efficient samples for social research. I do not do this alone: I strongly believe in team-based research and teaching, and am really happy to work with a group of great people in Utrecht .

If you want to know more about the ‘programme’ of the chair, do click on this link to read the full profile report for the chair in data quality . I will do an inaugural speech of the professorship on 2 April 2026.