|| Checking for direct PDF access through Ovid
The present paper responds to defenses of statistical significance testing offered by Levin and Robinson. First, some inaccurate perceptions of contemporary criticisms of statistical tests are noted. Second, areas of disagreement are explored. For example, it is noted that all nine empirical studies of reporting practices since 1994 show that “encouraging” (per the 1994 APA style manual) authors to report effect sizes has not worked; two reasons for this failure are explored. Finally, two important areas of agreement regarding needed improvements in contemporary practices are noted.