Hi Hicola,
Two questions, two answers.
On 01/03/2008, Davies, Nicola <[log in to unmask]> wrote:
> Hi Everyone,
>
> I am wondering if you could help me with a statistical query.
>
> What would be the best way to examine differences in questionnaire
> responses at two points in time?
If those responses are continuous, it's a repeated measures t-test, if
they're ordinal, a Wilcoxon text, if they're categorical, a Sign test
or McNemar test.
> Furthermore, what would be the best way to examine whether a particular
>variable can account for any differences?
>
Usually the best way is some sort of regression, use the second
measure as output, control for the first measure, and see if another
variable predicts the second measure, controlling for the first.
> Thank you for your help.
>
No problem.
J
> Kind Regards,
>
>
> Nicola Davies,
> BSc; MSc Comm.; PhD Researcher
> Researcher at the University of Oxford
>
--
Jeremy Miles
Learning statistics blog: www.jeremymiles.co.uk/learningstats
Psychology Research Methods Wiki: www.researchmethodsinpsychology.com
|