Statistical Significance, Effect Size, and Confidence Intervals

Since its invention 90 years ago, the p-value has become the standard by which most quantitative research is judged; however, it was never intended for this purpose. Indeed, a 2016 joint statement by the American Statistical Association argued, “By itself, a p-value does not provide a good measure of evidence regarding a model or hypothesis.” A p-value alone is uninformative; it is prone to false positives and says nothing about an effect's magnitude (e.g., effect size) or range (e.g., confidence intervals). Additionally, over-reliance on p-values may even encourage unethical research practices. In this session, Dr. Taylor will discuss the concepts of, and interrelationships, among p-values, effect sizes, and confidence intervals. After completing this seminar, attendees will gain a solid foundation for interpreting and presenting “meaningful” research results spanning beyond the p-value alone.