Abstract
<jats:p>Rigorous scientific evidence requires both “transparency in action” about the research process and “transparency in thought” about the assumptions and logic underlying decisions. The open science movement has emphasized transparency in action, including open data and code standards. In parallel, causal inference literature has emphasized transparency in thought, particularly around the assumptions needed to make causal claims and the limits on our inferences. These two topics, however, are rarely discussed together in ecological and conservation research. Here, we demonstrate how research practices that increase transparency around causal inferences complement the open science agenda. To illustrate, we use a case study on the direct and indirect effects of drought on dryland plant productivity through soil moisture. We use two expert-elicited causal graphical hypotheses to highlight how causal inference practices and workflows help reveal true scientific uncertainties versus simply differences in analytical choices that may muddle transparency in thought. Cross pollination between the open science and causal inference communities can expand both the reproducibility and credible of causal claims in ecology and conservation.</jats:p>