topic Re: intraclass correlation coefficient (ICC2,1) in Discussions

intraclass correlation coefficient (ICC2,1)

Junjun — Thu, 08 Jun 2023 16:31:31 GMT

Hi, I would like to know how to perform intraclass correlation coefficient (ICC2,1, two-way random effect model).

The data I have is from 150 people answering to a test repeatedly measured twice, 2 weeks after the initial test.I would like to know the ICC of this test.

Sorry, if I have not made myself clear enough to answer to this question.

Thank you very much.

Re: intraclass correlation coefficient (ICC2,1)

MRB3855 — Mon, 06 Mar 2023 10:28:10 GMT

Welcome to the community. Yes, more info would be helpful; so, you say you have a repeated measures design with two-way random effects? Recognizing potential proprietary issues, can you lay it out more clearly/explicitly for us? How are the subjects allocated to the random factors? Fixed effects? etc...

Re: intraclass correlation coefficient (ICC2,1)

Junjun — Tue, 07 Mar 2023 06:11:32 GMT

Thank you for your reply.

After I asked my question, I found the paper below.

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4913118/]

I think what i wanted to do was a two Two-way random effects, absolute agreement, single rater/measure.

I can now see that JMP will calculate ICC (with bias and interaction), so I think this will correspond to ICC(2,1).

Thank you for the help.