Epidemiological 2 × 2 Table Analyzer

Scope and design

This calculator analyzes one unstratified table of independent observations. Cohort and randomized-trial modes report risk measures; cross-sectional mode reports prevalence measures; unmatched case-control mode reports an exposure odds ratio and does not estimate risks.

Formula and confidence intervals

Group risks or prevalences use Wilson score intervals. The risk or prevalence ratio uses a conventional log-Wald interval, the absolute difference uses Newcombe's unpaired hybrid-score method 10, and the odds ratio uses a log-Woolf interval. Pearson's chi-square test is calculated without Yates correction and the conventional two-sided Fisher exact test conditions on the observed margins. No automatic continuity correction is applied.

Assumptions and limitations

• Each observation is independent and appears once.
• The exposure/intervention and outcome are genuinely binary and correctly oriented.
• Displayed estimates are crude and unadjusted.
• Case-control sampling does not identify outcome risks, risk ratios, risk differences, or NNT/NNH.
• Cross-sectional associations do not establish temporal order.
• Approximate ratio intervals can be unreliable with sparse data and are unavailable at some zero-cell boundaries.
• Matched, stratified, clustered, weighted, repeated, person-time, survival, and adjusted analyses are outside this release.

Interpretation and safety

Association does not establish causation. Statistical significance does not establish clinical or public-health importance, and a non-significant result does not establish equivalence or no association. Odds ratios and risk ratios are different measures and can differ materially when the outcome is common.

Validation

Version 1.0.0 was checked against prespecified hand calculations and named R reference implementations for Wilson intervals, log-Wald ratios, Newcombe differences, Pearson chi-square, and Fisher exact tests. Boundary, design-applicability, interface, privacy, and responsive-layout tests are included. Version 1.1.0 retains the same statistical methods with a harmonized interface; owner review, native PDF inspection and screen-reader listening checks remain pending. Availability does not constitute regulatory validation.

Privacy

All calculations use aggregate counts in browser memory. The app does not transmit, upload, or persist entered counts, labels, or results. Do not enter identifiable or confidential information.

References

1. Centers for Disease Control and Prevention. Analyzing and Interpreting Data. Field Epidemiology Manual.
2. Centers for Disease Control and Prevention. Epi Info StatCalc: Tables (2 × 2, 2 × n).
3. Newcombe RG. Two-sided confidence intervals for the single proportion: comparison of seven methods. Stat Med. 1998;17:857–872.
4. Newcombe RG. Interval estimation for the difference between independent proportions: comparison of eleven methods. Stat Med. 1998;17:873–890.
5. Agresti A. Categorical Data Analysis. 3rd ed. Wiley; 2013.