Ottenbacher K J, Barrett K A
School of Health Related Professions, State University of New York, Buffalo 14214.
Am J Phys Med Rehabil. 1990 Apr;69(2):102-7. doi: 10.1097/00002060-199004000-00011.
The findings of 100 data-based studies exploring the effectiveness of rehabilitation procedures were reviewed and analyzed. The statistical conclusion validity was determined by computing post hoc power coefficients for the statistical hypotheses tests included in the review studies. The data analysis revealed that the median power to detect a small effect size in the studiesreviewed was 0.08. The median power to detect medium and large effect sizes was 0.26 and 0.56, respectively. The results suggest the possibility of a high rate of type II experimental errors in the rehabilitation research. A type II error represents a false-negative conclusion and occurs when a researcher supports a null hypothesis that was false and should have been rejected.
