Seven myths about causal inference in ecology: clarifications and common ground

Laura E Dee , Paul J Ferraro, Katherine Johannet Siegel, Steve J. Miller, Jarrett E. K. Byrnes, Hannah E. Correia, Tyler L. McIntosh, Suchinta Arif

Identifying and quantifying causal relationships from data is a central goal in ecology and conservation. To achieve this goal, ecologists are increasingly applying frameworks and methods from the broader, multidisciplinary causal inference literature. These applications have generated both enthusiasm and skepticism. Here, we address seven common sources of skepticism that we see as myths. Much of this skepticism treats causal inference as a competing research paradigm. However, it is better understood as a set of ideas and practices that can sharpen ecological reasoning and strengthen empirical evidence. The value of causal inference lies in clarifying the assumptions required to interpret statistical associations causally and in encouraging researchers to state these assumptions explicitly, justifying them using prior knowledge and biological reasoning. In this sense, causal inference does not replace what ecologists already do well. Rather, it helps reveal, refine, and communicate the logic already underlying empirical research with causal aims. Across the seven concerns that we examine, a common message emerges: debates about causal inference should focus less on labels and techniques and more on whether the assumptions behind a causal claim are credible, clearly stated, and fit for purpose when making causal claims in ecology.