The highest ranking journals were selected for review as published research in them was thought to represent the better designed studies available to American criminologists. From the review of 129 articles, 53 were selected as they contained tests of statistical significance. Three statistical power determinations were made for each of the 3,689 reported tests of significance, and the average power for detecting a range of possible treatment effects was calculated for each of the studies included in the analysis. The survey showed appreciably higher power levels than have been revealed in other disciplines. The median sample size was approximately 248 and fully 66 percent of the articles had less than a 50-50 chance of detecting a small treatment effect. The study concluded that the minimum information required to be delineated in articles should be standardized to enable replication and independent evaluation of all published results and thus upgrade criminological research. Tables and 33 references. (Author abstract modified)