ABSTRACT
Dissatisfaction over the limitations of significance testing and over mischaracterization of these limitations has generated a great deal of research and commentary over the past 15 years (e.g., Cohen, 1994; Cortina & Dunlap, 1997; Schmidt, 1998). In addition, the American Psychological Association (APA) created a task force whose charge was to consider whether and how APA policy should be modified to reflect the state of the art regarding the reporting of empirical results. Among the recommendations of the task force was that indices of effect size be regularly reported in empirical research published in APA journals and that authors’ conclusions be influenced by them. The primary advantage of the adoption of this recommendation, in our opinion, has been that authors of large sample studies are no longer able to claim unfettered support for hypotheses simply due to p < .05. Instead, conclusions must be tempered with recognition that small effects, though perhaps not attributable to chance, were trivially small.
