Statistical power of articles published in three health psychology-related journals

Power was calculated for 8,266 statistical tests in 187 journal articles published in the 1997 volumes of Health Psychology (HP), Addictive Behaviors (AB), and the Journal of Studies on Alcohol (JSA). Power to detect small, medium, and large effects was .34. .74. and .92 for HP; .34, .75, and .90 for AB; and .41, .81. and .92 for JSA. Mean power estimates are .36, .77, and .91, giving a good estimation for the field of health psychology. J. Cohen (1988) recommended that power to detect effects should be approximately .80. Using this criterion, the articles in these journals have adequate power to detect medium and large effects. Intervention studies have much less power to detect effects than nonintervention studies do. Results are encouraging for this field, although studies examining small effects are still very much underpowered. This issue is important, because most intervention effects in health psychology are small.