Hi Helen
A simple rule of thumb is that if your test was significant then you
have enough power. So if all your hypotheses were confirmed by the
statistical tests used you shouldn't need to worry ;-)
If you had some non-significant results post-hoc power calculations
are not really advisable (Russ Lenth's page is helpful
http://www.cs.uiowa.edu/~rlenth/Power/ ). Although if you think that
for some tests non-significant results may have been due to low power
(e.g. the effect size was relatively large) you might want to think
about the detectable effect size your sample gives rather than the
sample size you should have had. For example, for an independent
t-test with 30 cases in each arm you would only be able to detect
standardized mean differerences (i.e. chohens d) greater than .73, or
for 60 in each arm d>.51 (for p=.05 & power=.8)
Just to reiterate what others have said, it's unlikely that examiners
will want to spend much time discussing the intricacies of your
statistical analyses.
Best wishes with your viva
Sam
On 22 October 2010 12:25, Helen Mann <[log in to unmask]> wrote:
> Hello (yes me again, im revising for my viva next week and getting paranoid)
>
> I never did any sample size power calculations before I ran my
> test....mainly because I never knew about them at the time and thought you
> just went with how many people you could get to take part in your
> experiment.....anyway, do you think I need to go through all the papers
> where I got my scales from and work out the sample size I needed
> for completing each scale to get power = 0.8???!!
>
> Or can I use my data to work out the sample size power and then say whether
> I needed more or less participants?!!?
>
> Has anyone ever been asked about sample size in a viva??!?!
>
> HELP!!
>
> Helen
>
--
Sam Norton
Centre for Lifespan & Chronic Illness Research
University of Hertfordshire
www.go.herts.ac.uk/samnorton
|