LOW-POWER, TYPE-II ERRORS, AND OTHER STATISTICAL PROBLEMS IN RECENT CARDIOVASCULAR RESEARCH

Frequently in biomedical literature, measurements are considered ''not statistically different'' if a statistical test fails to achieve a P value that is less than or equal to 0.05. This conclusion may be misle ading because the size of each group is too small or the variability i s large, and a type II error (false negative) is committed. In this st udy, we examined the probabilities of detecting a real difference (pow er) and type II errors in unpaired t-tests in Volumes 246 and 266 of t he American Journal of Physiology: Heart and Circulatory Physiology. I n addition, we examined all articles for other statistical errors. The median power of the t-tests was similar in these volumes (similar to 0.55 and similar to 0.92 to detect a 20% and a 50% change, respectivel y). In both volumes, similar to 80% of the studies with nonsignificant unpaired t-tests contained at least one t-test with a type II error p robability >0.30. Our findings suggest that low power and a high incid ence of type II errors are common problems in this journal. In additio n, the presentation of statistics was often vague, t-tests were misuse d frequently and assumptions for inferential statistics usually were n ot mentioned or examined.