Using Anchor-Based Methods to Determine the Smallest Effect Size of Interest

Effect sizes are an important outcome of quantitative research because they allow researchers to communicate the practical significance of the findings. However, there are few guides for psychological researchers that explain how they can quantify which effect sizes are practically meaningful, and which are not. Being able to specify the smallest effect size that is considered practically meaningful facilitates the design of well-powered experiments and enables researchers to falsify predictions by rejecting effects larger than this effect size. We illustrate two methods used in clinical research to determine the minimally detectable difference: the smallest effect size that is associated with a subjectively noticeable change at the individual level. These anchor-based methods provide one possible approach for researchers to determine their smallest effect size of interest, and to interpret the effect size estimates observed in research lines. We discuss the limitations of applying anchor-based methods in psychology, and point out there is ample room for statistical and methodological improvement, but also highlight the potential for anchor-based methods to provide one approach to establish a smallest effect of interest in lines of research.