It is easy to assume that science is more flawed than in the past, given widespread coverage of the reproducibility crisis, perverse incentives and P-value hacking, alongside a proliferation of corrective measures (see, for example, C. K. Gunsalus et al. Nature 566, 173–175; 2019). But it could be that we are now seeing more problems simply because we are more alert to them.
Consider related lessons from medicine. When a new treatment method is invented for a particular disease, there is typically an increase in the number of people who might benefit from that treatment, because conditions that were previously undetectable or ignored are now uncovered systematically.
Science could be the same. Today, we study in minute detail the dynamics of discovery, careers, teams and institutions, sometimes exposing issues that we didn’t know existed (see S. Fortunato et al. Science 359, eaao0185; 2018). That, to me, is grounds for optimism: such intense scrutiny should markedly improve the quality of science.
Nature 567, 311 (2019)