Bibliometric uncertainty: data quality, reliability, and validity
PhD defence
On 1 October 2026, Paul Donner successfully defended his PhD thesis at Leiden University, titled ‘Bibliometric uncertainty: data quality, reliability, and validity’.
For his thesis, Paul Donner studied the quality of bibliometric data and investigated how data errors can impact the outcomes of indicators that are used to assess research. While the existing literature has studied the limitations of bibliometric data to quite an extent, Paul’s PhD thesis goes a step further: he investigated in how far data errors lead to uncertainty in commonly used, sophisticated indicators.
For example, to compare the research outputs of academic organisations, the data needs to connect authors of publications to organisations in a reliable way. However, spelling variants, errors, or translations can make it difficult to clearly identify the correct organisation. Paul’s work shows what limitations like this mean for the reliability and validity of bibliographic data sources.
The implications of such uncertainty for the interpretation of bibliometric indicators are also discussed in a recent blog post by Paul.
In his laudation, Paul’s supervisor, Nees Jan van Eck, reflected on the thoroughness of Paul’s work, noting that it encouraged others to reconsider some of the basic assumptions underlying bibliometric research and practice. Through his careful analysis, Paul challenged assumptions that were often taken for granted and encouraged more critical thinking about the reliability of the indicators on which important decisions were based.
Paul Donner is a staff researcher at the German Centre for Higher Education Research and Science Studies and was an external PhD candidate at Leiden University, supervised by Paul Wouters and Nees Jan van Eck. His work includes fundamental, infrastructure, and contract research in science studies, primarily related to topics of indicator design and validation, research collaboration, and bibliometric data quality.