What is the difference between ICC(2,1) and ICC(3,1)?

ICC type depends on rater selection and generalisability. ICC(1,1): random raters, one measurement; ICC(2,1): random raters from a population, generalise to all raters; ICC(3,1): fixed raters, results apply only to these specific raters. ICC(3,1) = Pearson's r for two raters.

When should I use ICC instead of Cohen's kappa?

Use ICC for continuous or interval-level measurements (blood pressure, test scores, grip strength); use kappa for categorical or nominal data (diagnosis present/absent, classification into groups). For ordinal ratings with many categories, weighted kappa and ICC(3,1) often give similar results.

Intraclass Correlation Coefficient (ICC)

Calculate the Intraclass Correlation Coefficient (ICC) from ANOVA variance components. ICC measures both inter-rater reliability and test-retest reliability for continuous measurements. Koo and Mae (2016) thresholds: <0.50 poor, 0.50–0.75 moderate, 0.75–0.90 good, >0.90 excellent.

Data set — comma or newline separated

Statistical Results — n = 8

Count (n)8

Sum (Σx)40

Min2

Max9

Range7

Q14

Q35.5

IQR1.5

Mean (μ)5

Median4.5

Mode4

Pop. Variance (σ²)4

Pop. Std Dev (σ)2

Sample Variance (s²)4.571428571

Sample Std Dev (s)2.138089935

Skewness0.65625

Excess Kurtosis-0.21875

How to Use

Enter the values for the scenario you are assessing. Calculate the Intraclass Correlation Coefficient (ICC) from ANOVA variance components. ICC measures both inter-rater reliability and test-retest reliability for continuous measurements. Koo and Mae (2016) thresholds: <0.50 poor, 0.50–0.75 moderate, 0.75–0.90 good, >0.90 excellent. Use the icc calculator result to inform your calculation.

Intraclass Correlation Coefficient (ICC)

How to Use

Frequently Asked Questions

What is the difference between ICC(2,1) and ICC(3,1)?

When should I use ICC instead of Cohen's kappa?