Specification uncertainty: what the disruption index tells us about the (hidden) multiverse of bibliometric indicators

Abstract Following Funk and Owen-Smith (Manag Sci 63:791–817, 2017), Wu et al. (Nature 566:378–382, 2019) proposed the disruption index (DI 1 ) as a bibliometric indicator that measures disruptive and consolidating research. When we summarized the literature on the disruption index for our recently published review article (Leibel and Bornmann in Scientometrics 129:601–639, 2024), we noticed that the calculation of disruption scores comes with numerous (hidden) degrees of freedom. In this Letter to the Editor, we explain based on the DI 1 (as an example) why the analytical flexibility of bibliometric indicators potentially endangers the credibility of research and advertise the application of multiverse-style methods to increase the transparency of the research.