| Date: | Fri, 16 Aug 2002 06:31:55 +0100 |
|---|
| Reply-To: | Chris Evans <chris1@psyctc.org> |
|---|
| Sender: | "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU> |
|---|
| From: | Chris Evans <chris1@psyctc.org> |
|---|
| Organization: | PSYCTC.org |
|---|
| Subject: | Re: Reliability analysis |
|---|
| In-Reply-To: | <sd5cc927.020@osca.latrobe.edu.au> |
|---|
| Content-type: | text/plain; charset=US-ASCII |
|---|
On 16 Aug 2002 at 9:42, Sue Cotton wrote:
> I was wondering whether someone would know the difference between
> single and average intraclass correlation coefficients that are
> produced in output of SPSS after running reliability analysis.
>
> Which one is the best estimate to use?
Depends what you want to use your data for. If you have multiple
rater data and will only use one rater at a time in future then the
single ICC is the estimate of the population single rater
reliability. If you have (say) five raters and will always use five,
or data from all 21 items from a measure and will always use the
whole measure, then the average ICC is the estimate of the population
reliability of all five raters or the full scale.
Cheers,
Chris--
Chris Evans <chris@psyctc.org>
Consultant Psychiatrist in Psychotherapy,
Rampton Hospital; Associate R&D Director,
Tavistock & Portman NHS Trust;
Hon. SL Institute of Psychiatry
*** My views are my own and not representative
of those institutions ***
|
