Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: st: reliability with -icc- and -estat icc-
From
Nick Cox <[email protected]>
To
[email protected]
Subject
Re: st: reliability with -icc- and -estat icc-
Date
Wed, 27 Feb 2013 02:03:18 +0000
Psychometrics is not my field but my gut impression is that you really
should use the scores. The context you cite is no doubt correct, but
for comparing raters with each other the scores they gave are
essential. Why is skewness implied to be a problem here? It's just one
of the facts.
On Wed, Feb 27, 2013 at 1:56 AM, Lenny Lesser <[email protected]> wrote:
> Thank you for you comments and yes you may use my example.
>
> I'm guessing both of you believe that I should be using the score
> rather than the ranking of the apps.
> Some have thought I should use the rank as the scale was from 0-100
> and all were ranked very low, resulting in a skewed distribution. We
> talked about either ranking the apps or using a transformation of the
> scores.
>
> If you think we are ok using the straight scores, then I will stick
> with that. I'm not sure of the statistical arguments for or against.
>
> On Tue, Feb 26, 2013 at 5:49 PM, JVerkuilen (Gmail)
> <[email protected]> wrote:
>> On Tue, Feb 26, 2013 at 8:47 PM, Nick Cox <[email protected]> wrote:
>>> I believe you on the first point. Are you agreeing/disagreeing on the second?
>>
>> I was vehemently agreeing with you. That rater was crAZy!
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/faqs/resources/statalist-faq/
* http://www.ats.ucla.edu/stat/stata/