Statistical significance using p-values has well-known flaws in assessing the reliability of research findings. A new study puts forth an innovative statistic called the Robustness Index to quantify the fragility of results. By measuring how changing sample size impacts significance, this index provides an intuitive metric of a study’s robustness against violations of assumptions. The Robustness Index adjusts for sample size and prevalence, overcoming limitations of existing fragility measures dependent on sample size. As the reproducibility crisis grows, this research highlights the need for improved statistical tools. Adopting the Robustness Index could greatly strengthen the validity and transparency of medical research beyond the standard p-value.
Citation: Heston TF. The robustness index: going beyond statistical significance by quantifying fragility. Cureus. 2023 Aug 30;15(8):e44397. https://doi.org/10.7759/cureus.44397