1. What is ICC?

The Intraclass Correlation Coefficient (ICC) is a measure of reliability or agreement among raters or measurements. It assesses how strongly units in the same group resemble each other, ranging from 0 (no agreement) to 1 (perfect agreement).

2. How Does the Calculator Work?

The calculator uses the ICC formula:

Where:

• \( MS_R \) — Mean Square Between Rows (subjects)
• \( MS_C \) — Mean Square Between Columns (raters)
• \( MS_E \) — Mean Square Error
• \( k \) — Number of raters
• \( n \) — Number of subjects

3. Interpretation of ICC Values

Guidelines:

• ICC < 0.5: Poor reliability
• 0.5 ≤ ICC < 0.75: Moderate reliability
• 0.75 ≤ ICC < 0.9: Good reliability
• ICC ≥ 0.9: Excellent reliability

4. Using the Calculator

Tips: Enter mean square values from ANOVA table, number of raters and subjects. All values must be positive with at least 2 raters and 2 subjects.

5. Frequently Asked Questions (FAQ)

Q1: What's the difference between ICC and Pearson correlation?
A: ICC assesses agreement (absolute values), while Pearson assesses linear relationship (relative values).

Q2: Which ICC model should I use?
A: This calculator uses ICC(2,1) - two-way random effects model for absolute agreement.

Q3: How many raters/subjects do I need?
A: Typically 2-5 raters and 30+ subjects for stable estimates, but depends on expected ICC.

Q4: Can ICC be negative?
A: Yes, but this usually indicates problems with the data or model specification.

Q5: What are alternatives to ICC?
A: Cohen's kappa (for categorical data), Bland-Altman plots, or Krippendorff's alpha.