1. What is ICC Sample Size Calculation?

The ICC (Intraclass Correlation Coefficient) sample size calculation determines the number of participants needed to achieve adequate power for reliability studies. It ensures your study can detect the expected level of agreement between raters or measurements.

2. How Does the Calculator Work?

The calculator uses the ICC sample size formula:

Where:

• \( N \) — Required sample size
• \( Z_{\alpha/2} \) — Critical value for Type I error (1.96 for α=0.05)
• \( Z_{\beta} \) — Critical value for power (0.84 for 80% power)
• \( ICC \) — Expected intraclass correlation coefficient
• \( k \) — Number of raters or repeated measurements

Explanation: The formula accounts for the expected reliability (ICC), number of raters, desired power, and significance level.

3. Importance of Sample Size for ICC

Details: Adequate sample size ensures your reliability study has sufficient power to detect the expected level of agreement, preventing underpowered studies that may miss true reliability.

4. Using the Calculator

Tips: Enter expected ICC (0-1), number of raters (≥2), select alpha level (usually 0.05) and desired power (usually 80%). All values must be valid.

5. Frequently Asked Questions (FAQ)

Q1: What is a good ICC value?
A: ICC >0.75 indicates excellent reliability, 0.6-0.74 good, 0.4-0.59 fair, and <0.4 poor reliability.

Q2: How many raters should I use?
A: Typically 2-5 raters. More raters reduce required sample size but increase study complexity.

Q3: What if my expected ICC is unknown?
A: Use a conservative estimate from similar studies or pilot data. Underestimating ICC leads to larger but safer sample sizes.

Q4: Can I use this for test-retest reliability?
A: Yes, when k represents the number of repeated measurements per participant.

Q5: What's the minimum sample size recommended?
A: Most reliability studies need at least 30-50 participants, even if calculations suggest fewer.