English Français
The effect of casual language on causal understanding: The need for unconfounded effect size benchmarks
Effect size benchmarks are used for guiding theory, interpreting practical significance, and reviewing scientific progress. However, effect size estimates that are correlational typically violate the definition of an “effect” because they do not capture a cause-and effect relationship. We begin the current work with a review of the state of the literature including four key challenges in creating effect size benchmarks and establishing evidence of causal inference strength. Next, to illustrate the limitations and
opportunities in current practice, we present a systematic review of the leadership literature that highlights four themes related to causally identified effect sizes–or lack
thereof–from experiments that meet (or fail to meet) standards of rigor. We conclude this work with a blueprint that provides a meaningful redirection of the conversation so
that future meta-analytic studies can provide accurate, specific, and unconfounded effect size benchmarks.