Specific Judgment Errors in Clinical and Laboratory Research: Type III, Type IV, Type M, and Type S Errors

Reliable clinical and laboratory research requires clearly defined methodological processes, appropriate statistical evaluation, and careful interpretation of findings. Null hypothesis significance testing (NHST), derived from Fisher’s concept of statistical significance and the Neyman–Pearson decision framework, traditionally focuses on controlling type I and type II errors. However, these conventional errors do not capture all sources of misleading inference. Type III and type IV errors address situations in which a study answers the wrong question, identifies an effect in an unexpected or opposite direction, or misinterprets a statistically supported result. More recently, type S and type M errors have been proposed to assess the reliability of statistically significant findings by focusing on the direction and magnitude of the estimated effect. These errors are particularly relevant in low-powered or noisy studies, including large clinical laboratory datasets, method-comparison studies, reference interval studies, and analytical validation studies, where significant results may have the wrong sign or substantially exaggerate the true effect size. This review summarizes the conceptual basis and practical implications of type III, type IV, type S, and type M errors in clinical and laboratory research, emphasizing their relevance for study design, analysis, interpretation, reproducibility, and decision-making.