NWB2023_Challenges of causality in Open Science

Open Science has been steadily on the rise in the past decade. This includes Open Science practices, infrastructures, and policies, and touches on virtually all aspects of the research cycle and the academic system. The impacts of Open Science are often in the spotlight, but we should clearly distinguish it from the effects of Open Science on impacts. In the PathOS project, we are interested in understanding the causal effects of Open Science. Causal effects are difficult to infer in general. A fundamental problem is that many of the concepts of interest are not directly observable. Scientometrics studies typically use various metrics as indicators of the (unobserved) concepts of interest, but this also raises some thorny questions around causality. We will review some of the challenges around causality that emerge when studying open science and indicators and suggest ways forward.