Statistical Significance Testing:Problems and Reflection

Statistical significance testing(SST) is a commonly used way of statistical inference.Owing to its overenthusiastic adoption and an exaggerated view of its role in research,however,SST has long been abused and misused so much that its own validity has been undermined.After defining such basic pairs of concepts in SST as sample and population,this paper discusses some misconceptions of statistical inference based on statistical significance and non-significance,replication fallacy as well as confusion of statistical significance.To address the problems,it is essential to apply with effect size test,statistical power test,confidence interval estimation or repetitive research.