The Ongoing Tyranny Of Statistical Significance Testing

1066 Words5 Pages
Article by Stang, Poole and Kuss (2010) titled “The ongoing tyranny of statistical significance testing in biomedical research” describe common misuses and interpretation of statistical significance testing (SST). The authors point out fallacy understanding in interpretive the p-value and how it often mixed in measuring effect size and its precision. This misconception then they assert may impede scientific progress and furthermore become unintended harmful treatment. They also proposed an important way out of the significance fallacies in this article. Therefore, in this article review, all the finding that made by the authors will be summarized and review of it will be drawn based on other references. 1. Statistical Significance Test (SST) and P-value Stang, Poole and Kuss explain, in SST, P-value is an important part to decide the null hypothesis. The SST itself, they explain is analytical approach that developed based on two prominent statisticians, Fisher and the Neyman-Pearson. However, in present practical, SST is incompatible amalgamation of those two theories. In Fisher theory, P value represents the strength of evidence against the null hypothesis: the lower the P-value, the stronger the evidence. In this theory, they criticize lake of alternative hypothesis and concept of statistical power. In contrast Neyman and Person theory included the alternative for the null hypothesis, type I and II error and theoretical of effect size. This hybrid method leads to
Open Document