Given the long history of effect size (ES) indices (Olejnik and Algina
Contemporary Educational Psychology, 25, 241–286 2000) and various attempts by APA
and AERA to encourage the reporting and interpretation of ES to supplement findings from
inferential statistical analyses, it is essential to document the impact of APA and AERA
standards on ES reporting practices. In this paper, we investigated the impact by examining
findings from 31 published reviews and our own review of 451 articles published in 2009
and 2010. The 32 reviews were divided into two periods: before and after 1999. A total of
116 journals were reviewed. Findings from these 32 reviews revealed that since 1999, the ES
reporting has improved in terms of its rate, variety, interpretation, confidence intervals, and
fullness. Yet several inadequate practices still persisted: (1) the dominance of Cohen’s d, and
the unadjusted R2/η2, (2) the mere labeling of ES, (3) the under-reporting of confidence
intervals, and (4) a lack of integration between ES and statistical tests. The paper concludes
with resources on Internet and recommendations for improving ES reporting practices.
附件列表