Intraclass Correlation Coefficient (ICC) Calculator
Measure the reliability of continuous ratings or repeated measurements. All six ICC forms are reported so you can pick the model that matches your design.
Updated · By Bibin XavierData 60 × 7
| Casetext · 60 levels | Rater_Anumeric · coded | Rater_Bnumeric · coded | Rater_Cnumeric · coded | Judge_1numeric | Judge_2numeric | Judge_3numeric |
|---|---|---|---|---|---|---|
| C01 | 1 | 1 | 1 | 58.8 | 59.8 | 53.4 |
| C02 | 1 | 1 | 1 | 68.1 | 74.1 | 69.3 |
| C03 | 1 | 1 | 1 | 74 | 78.5 | 73.6 |
| C04 | 3 | 4 | 3 | 68.5 | 60.9 | 63.2 |
| C05 | 3 | 3 | 3 | 68.8 | 79.7 | 73.2 |
| C06 | 3 | 3 | 3 | 47.2 | 58.8 | 47.9 |
| C07 | 4 | 4 | 4 | 72.2 | 81.8 | 81.7 |
| C08 | 1 | 3 | 1 | 61.9 | 65.5 | 62.2 |
Variables and options
Results
60 rows · example dataIntraclass correlation coefficients
| Form | Model | ICC | 95% CI | F | df | p |
|---|---|---|---|---|---|---|
| ICC(1,1) | One-way random, single rater | .741 | [.636, .825] | 9.58 | 59, 120 | < .001 |
| ICC(1,k) | One-way random, average of raters | .896 | [.840, .934] | 9.58 | 59, 120 | < .001 |
| ICC(2,1) | Two-way random, absolute agreement, single rater | .745 | [.603, .841] | 11.88 | 59, 118 | < .001 |
| ICC(2,k) | Two-way random, absolute agreement, average | .898 | [.820, .941] | 11.88 | 59, 118 | < .001 |
| ICC(3,1) | Two-way mixed, consistency, single rater | .784 | [.692, .856] | 11.88 | 59, 118 | < .001 |
| ICC(3,k) | Two-way mixed, consistency, average | .916 | [.871, .947] | 11.88 | 59, 118 | < .001 |
Note. Koo and Li (2016): below .50 poor, .50–.75 moderate, .75–.90 good, above .90 excellent. ICC(2,1) is the usual choice for inter-rater reliability with raters sampled from a larger population.
Interpretation
Using a two-way random-effects model for absolute agreement, reliability is moderate, ICC(2,1) = .745.
How to cite this tool
Using these results in a thesis, journal article, report or class assignment? Please cite this free tool so others can find it. It keeps the site free.
APA 7
Xavier, B. (2026). Intraclass Correlation Coefficient (ICC) Calculator [Web application]. FreeStatisticalTools.com. https://freestatisticaltools.com/icc-calculator/
BibTeX
@software{xavier_intraclasscorrelation_2026,
author = {Xavier, Bibin},
title = {Intraclass Correlation Coefficient (ICC) Calculator},
publisher = {FreeStatisticalTools.com},
year = {2026},
url = {https://freestatisticaltools.com/icc-calculator/},
note = {Web application}
}Suggested sentence for your methods section: “The analysis was carried out with the free Intraclass Correlation Coefficient (ICC) Calculator tool on FreeStatisticalTools.com (Xavier, 2026).”
What is the intraclass correlation (ICC)?
The intraclass correlation coefficient measures the reliability of continuous ratings from several raters, or of repeated measurements. All six forms are reported so you can choose the one that matches your design.
When to use the intraclass correlation (ICC)
Use the ICC for continuous ratings by several raters, or test-retest reliability. Choose the form: one-way if each subject has different raters, two-way random if raters are a sample, two-way mixed if these raters are the only ones of interest.
How to set up your data
One row per subject and one column per rater or occasion.
How to use this tool
- Enter your data. Paste columns from Excel, Google Sheets or SPSS with variable names in the first row, or upload a CSV or Excel file. The example data shows the expected layout.
- Choose the variables. Select the rater or occasion columns.
- Read the results. Results update automatically. Read the plain-English interpretation, then use Copy table to paste APA-style tables into Word and Copy for the results sentence.
Frequently asked questions
Which ICC should I use?
For inter-rater reliability with raters drawn from a larger pool, ICC(2,1) for absolute agreement is the usual choice. Use ICC(3,1) when these raters are the only ones of interest.
How is ICC interpreted?
Koo and Li (2016): below .50 poor, .50–.75 moderate, .75–.90 good and above .90 excellent.
Single or average measures?
Use single measures if one rater’s score will be used in practice, and average measures if the mean of all raters will be used.
Related tools
Browse all categories
Data cleaning and preparation · Describe and explore · Compare means · Nonparametric tests · Correlation and regression · Bayesian statistics · Categorical data · Reliability and agreement · Multivariate analysis · Structural equation modeling · Econometrics and time series · Panel data and causal inference · Marketing and decision research · Survival, clinical and meta-analysis · Study planning and calculators · PLS-SEM