It is often held that causal inference is a kind of statistical inference, to be carried out by estimating effect sizes using randomised controlled trials or meta-analyses, or by means of model-based approaches such as structural equation modelling or graphical causal modelling. I argue that this view is both mistaken and partly responsible for the replication crisis.

That observed associations are not replicated by subsequent studies is a part of normal science. A problem only arises when those associations are taken to establish causal claims: a science whose established causal claims are routinely overturned is indeed in crisis. Purely statistical methods for causal inference draw causal conclusions from patterns of associations and thus can lead to such a crisis. If, however, we understand causal inference as involving mechanistic inference alongside statistical inference, as Evidential Pluralism suggests, then we can avoid these fallacious inferences from association to causation. Thus, Evidential Pluralism can prevent the drama of normal science from turning into a crisis.