On Wed, 24 Jan 2007 12:49:08 +0000, Paul Mcgeoghan
<
[hidden email]> wrote:
>Hi,
>
>I have a student who has 10 questions where the responses are Yes, No and
Don't Know.
>She has 9 people who answer these questions on 2 seperate occasions.
>
>She expects if someone says Yes, the first time, they should say Yes the
2nd time to the same
>question so people should answer the questions the same both times the
questionnaire was given to
>them in general.
>
>She wants a test of agreement which will calculate a statistic for each
question and has been
>advised to use Cohen's Kappa (but this is looking at how 2 or more raters
agree on a set of
>questions so I don't think this is appropriate personally).
>
>In SPSS, have been looking at 2 related samples test and marginal
homogeniety as each question is
>nominal with 3 categories so I think this is the way to approach it. Is
this the best method in this
>case?
>
>Paul
>
>
>==================
>Paul McGeoghan,
>Application support specialist (Statistics and Databases),
>University Infrastructure Group (UIG),
>Information Services,
>Cardiff University.
>Tel. 02920 (875035).
I think I have found what I am looking for Reliability Statistics and
Intraclass Correlation Coefficient and Absolute Agreement?
Locked
