3
$\begingroup$

I am doing a study with one variable with continuous data; the measurement involves measurements done by three people. I wish to do an inter-rater reliability test, e.g. with Bland-Altman analysis, but cannot find information about how it works with multiple measures. Would some other test be better?

Sample data

Rater A: 380 404 381 374 377 392 373 369 371 378 392 331 366 418

Rater B: 379 400 380 372 366 390 374 364 370 378 392 326 364 424

Rater C: 379 402 380 372 367 392 374 361 373 378 392 332 364 420

Improve this question
$\endgroup$

1 Answer 1

Reset to default
4
$\begingroup$

Usually the intraclass-coefficient is calculated in this situation. It is sensitive both to profile as well as to elevation differences between raters. If all raters rate throughout the study, report ICC(2, k); if only one rater rates everything and the other only rate, say, 20% to check the interrater agreement, then you should report ICC(2, 1).

Improve this answer
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.