Bonferroni Adjustment Overview
The Bonferroni adjustment is a simple guard against false discoveries. It is useful when one study runs many tests. Each test creates a chance of a false positive. More tests increase that combined chance. This calculator applies a strict correction to every raw p value. It multiplies each p value by the number of planned comparisons. The adjusted value is then capped at one. A result is significant when the adjusted p value is at or below the selected alpha.
Why Multiple Testing Matters
A single test at alpha 0.05 allows a five percent error rate. Ten tests create more room for accidental findings. Bonferroni control keeps the family wise error rate close to the chosen level. It is conservative. That means it may miss weak effects. Yet it is trusted because it is clear, transparent, and easy to audit.
Practical Research Use
Use this tool when comparing several groups, outcomes, models, or survey questions. Enter one p value per line, or separate values with commas. Add labels when you want a cleaner report. Set comparisons to the number of tests in the family. Use the planned number, not only the number that produced small p values. This avoids selective correction.
Reading The Results
The corrected alpha shows the raw p value threshold. The adjusted p value shows the corrected evidence level. Both views lead to the same Bonferroni decision. The summary also counts significant and non significant tests. Export the table when you need documentation for reports, classes, manuscripts, or review notes.
Good Practice
Plan the test family before analysis. Keep related hypotheses together. Do not mix unrelated projects in one correction unless they share one decision family. Report raw p values, adjusted p values, alpha, and the comparison count. State that the Bonferroni method was used. Also discuss its conservative nature. For exploratory work, combine this method with effect sizes and confidence intervals. Statistical significance should not replace scientific judgment.
Limits and Alternatives
Bonferroni works best when errors are costly and the test set is not huge. When tests are highly correlated, it can be too strict. Holm adjustment is often more powerful. False discovery rate methods may suit broad screening studies as well.