In line with the Leiden Manifesto , we promote the use of quantitative bibliometric analyses as a means to support, not replace, qualitative expert assessment. Research intelligence should be driven by questions, not indicators. At the library, we help to develop suitable sets of indicators and methods that are tailored to the particular goals of the evaluation.

As bibliometric indicators are prone to ambiguity and biases, we encourage a combination of multiple indicators to provide a robust and pluralistic picture. We strongly discourage the use of unweighted citation-based indicators, such as the h-index (or Hirsch index) and its variants , as these are inconsistent, insufficient for measuring changes in performance over time, and unsuited to compare individuals across age groups and disciplines.

We support the use of weighted indicators that are normalized for document type, year of publication and subject area, such as the category normalized citation index (CNCI) used by InCites to which Maastricht University subscribes.

As many other institutions, like VSNU, KNAW, NWO, NFU, and ZonMw, Maastricht University has signed the San Francisco Declaration on Research Assessment (DORA) . In line with the DORA recommendations, we do not support the use of journal-based indicators, such as journal impact factors, in the evaluation of individual researchers. Like many scientometrics experts, we share in the idea that citation-based metrics are better suited for evaluating larger research units, the higher the aggregate, the better.

We support the idea that excellence and performance are multidimensional concepts and that only some aspects of excellence and performance can be captured by bibliometrics.

Like the social and local relevance of a piece of scientific work, also individual researchers’ excellence in teaching, mentoring, and administration cannot be assessed on the basis of citation counts.

We, therefore, suggest a combination of qualitative and quantitative evaluations for research and researchers.