Cohen's Kappa Calculator
Cohen’s kappa summarises agreement between two raters or classification methods while adjusting for the agreement expected by chance. It needs observed and expected agreement values calculated from the same dataset.
How to use this calculator
- Enter the observed agreement (%) using the units shown beside the field.
- Enter the expected chance agreement (%) using the units shown beside the field.
- Review the calculated values and check that the assumptions match your scenario.
- Change one input at a time when comparing scenarios so you can see which assumption drives the result.
Formula and assumptions
κ=(Po−Pe)/(1−Pe), where observed agreement Po and chance-expected agreement Pe are proportions.
- Kappa depends on category prevalence and the way expected agreement was estimated. There is no universal interpretation threshold that applies to every field.
- The result is computed from the values you provide; it does not fetch external rates, live account data or local rules.
Worked examples
Worked example
Observed agreement 86% and expected agreement 55% gives κ≈0.689.
Example inputs: Observed agreement (%) = 86, Expected chance agreement (%) = 55.
Compare the displayed output with your own measurements before using it in a decision.
Understanding the result
Kappa depends on category prevalence and the way expected agreement was estimated. There is no universal interpretation threshold that applies to every field.
Results are estimates based on the stated model. If the situation depends on measured conditions or contractual specifications, use those source values rather than typical assumptions.
Important notes
Kappa depends on category prevalence and the way expected agreement was estimated. There is no universal interpretation threshold that applies to every field.
Frequently asked questions
What formula does this calculator use?
It uses the method described above: κ=(Po−Pe)/(1−Pe), where observed agreement Po and chance-expected agreement Pe are proportions.
What should I verify before relying on the result?
Kappa depends on category prevalence and the way expected agreement was estimated. There is no universal interpretation threshold that applies to every field.
Can I compare different scenarios?
Yes. Change the inputs while keeping units and definitions consistent, and record the assumptions used for each scenario.
Related calculators
- Cohen’s Kappa Calculator — Estimate chance-corrected agreement between two raters from a 2 by 2 agreement table.
- Cohen’s d Effect Size Calculator — Estimate standardized mean difference using the pooled sample standard deviation of two independent groups.
- 2×2 Matrix Calculator — Add, subtract and multiply 2×2 matrices and calculate each determinant with transparent row and column inputs.
- A/B Test Sample Size Calculator — Estimate per-variant sample size for a two-sided conversion-rate A/B test using a normal approximation, baseline rate and relative minimum detectable lift.
- Sampling Fraction Calculator — Calculate what percentage of a defined population is represented by a sample, and see the remaining population count.
- Circle Calculator — Calculate circle area, circumference, and diameter from radius.
Related explanations
Learn the method behind the number or compare the terminology before using the calculator.
- How to Calculate Probability — Learn basic probability formulas, complementary probability, and how to calculate simple event chances.
- How to Calculate Standard Deviation — Learn what standard deviation measures, the population and sample formulas, and how to interpret the result.
- Percent Difference — Understand percent difference, when it is appropriate, and how it differs from percentage change and percent error.
This page provides a calculation tool plus explanatory information so you can understand what the result means and when the underlying method may not fit your situation.