This is not my area of expertise and I suspect this is fairly basic.

I have two independent groups of experts that have rated two different sets of applications.

I have calculated the intraclass correlations (single measure, one-way random effects) for both groups. Now I want to test the difference for statistical significance. What is the best method for doing that?

Here are some numbers:

Group 1:

• ICC (1,2) = 0.36 (0.28–0.43 95% CI)
• Numbers of experts: 19
• Number of paired ratings: 1.085

Group 2:

• ICC (1,2) = 0.33 (0.25-0.40 95% CI)
• Number of experts: 22
• Number of paired ratings: 1.304