Home > Statistics Every Writer Should Know > The Stats Board > Discusssion
intraclass correlation coeficient
Message posted by mandy on October 23, 2000 at 12:00 AM (ET)
I'm trying to run a test-retest on some data to show relaibility. Is this the correct choice in stats??
READERS RESPOND:
(In chronological order. Most recent at the bottom.)
Re: intraclass correlation coeficient
Message posted by sof on October 24, 2000 at 12:00 AM (ET)
Test-retest is one way of checking for reliability. There are other ways depending on the nature of your data, problem, and what assumptions you are willing to make about the population of interest. Any statistical procedure has to be checked for validity and reliability. This involves the data, how the data was collected, what assumptions you are willing to make about the data collection process and about the population of interest from which the data is presumably a random or probabilistic sample.
Re: intraclass correlation coeficient
Message posted by Bill on October 25, 2000 at 12:00 AM (ET)
Mandy, The intra-class correlation coefficient is usually used to access inter-rater agreement, a special case of reliability. A Pearson Correlation is usually used to measure test-retest reliability (depending on the nature of your measurement scale).
Re: intraclass correlation coeficient
Message posted by George (via 144.136.176.130) on May 10, 2001 at 6:49 AM (ET)
Should Intraclass correlations be used when you have Cluster Sampling?How do you control for Cluster Sampling Bias/Error when conducting a Manova?
Your $5 contribution helps cover part the $500 annual cost of keeping this site online.