P-value

1Detailed Explanation

The p-value quantifies the evidence against the null hypothesis in frequentist statistics. A p-value < 0.05 traditionally indicates statistical significance, though this threshold is arbitrary. Important caveats: p-values do not measure effect size, clinical importance, or the probability that the finding is true. Multiple comparisons inflate Type I error, requiring adjustment (Bonferroni, false discovery rate). 'P-hacking' (selectively reporting favorable results) leads to inflated false positives. The American Statistical Association (2016) and Nature Human Behaviour (2019) have published statements cautioning against overreliance on p-values. Effect sizes and confidence intervals should always accompany p-values.

2Examples

• A.A p-value of 0.03 means there is a 3% probability of observing such results if the null hypothesis is true
• B.P-value adjustment using Bonferroni correction when testing 20 hypotheses at alpha = 0.05/20 = 0.0025

3Why It Matters in Research

P-values are ubiquitous in medical research but are frequently misinterpreted. Understanding their limitations is essential for rigorous research and responsible reporting.

4Related Terms