Statistics/SPSS please help!

[Cas_P]

I'm not sure this is in the right place but wanted as many people as possible to see it before I pull my hair out and give up

Any Psychology/people who know stats I need to pick your brains!

Basically I have replicated the McGurk effect and have hypothesised that with background noise present the MG effect will be more apparent due to increased reliability in visual perception.

I have 2 groups of 10. 10 people did audio cond. and also audio visual cond. with no background noise (control group). And the other 10 people did the same however with background noise (Experimental group). I'm measuring the McGurk effect in both groups (correct answers in AV cond. vs Audio) but I'm also measuring the difference between control and experimental groups.

How would I input this into SPSS as:
A: 2 seperate repeated measures (within participants?) T-tests, to then evaluate the variance between the groups.
or
B:An ANOVA, which I can put all the data into in one go

Cheers,
Cassie

[captainzep]

So, what precisely were you doing when the early morning research methods lectures were happening Cas?

It involved bed didn't it?

Was it, the McGherkin' effect?

[_rubinho_]

captainzep said:

So, what precisely were you doing when the early morning research methods lectures were happening Cas?

It involved bed didn't it?

Was it, the McGherkin' effect?

She must have had a MANOVA!



http://en.wikipedia.org/wiki/Multivariate_analysis...

[DrTre]

captainzep said:

Was it, the McGherkin' effect?

It's the look of complete incomprehension when you ask at Mcdonalds for "No gherkin"

I imagine. I've never set foot in there. F&M only for me. Foie gras, cognac, those little fish eggs, that sort of thing.

Edited by DrTre on Sunday 29th November 20:02

[Cas_P]

Think I've worked it out, when I get the results I'll post them up and you smart people might feel nice enough to have a quick look

Edit : How do I do a boxplot with 2 D.V's and 2 I.V's?

Edited by Cas_P on Sunday 29th November 20:31

[Four Cofffee]

Samples of 10 and you want to run stats on it? I know often uni assignments are there to test your knowledge rather than produce proper work but that is hardly a viable sample, especialy for T tests. I am less well versed in non parametrics.

I recall having a flow chart for the correct statistical test but this does the same job:

http://www.whichtest.info/index.htm which says 2 groups, interval data I guess if you have scores of some sort (although technicaly I think you could argue that it is really Ordinal): T tests.

Edited by Four Cofffee on Sunday 29th November 20:36

Edited by Four Cofffee on Sunday 29th November 20:38

[Cas_P]

Four Cofffee said:

Samples of 10 and you want to run stats on it? I know often uni assignments are there to test your knowledge rather than produce proper work but that is hardly a viable sample

I recall having a flow chart for the correct statistical test but this does the same job:

http://www.whichtest.info/index.htm

Yes, this is more about ability to write up a report/do the stats than the results IYKWIM?

This is the only sample I could aquire in such a short time. I have been spent a lot of time in Birmingham because my mom was ill thus missed the initial meeting for this miniproject in which people were grouped up into 5-6, of which everyone would collect 10-15 samples each. However I do not want to submit extenuating circumstances and get an extension as I have exams in week 10, week 9 (as of tomorrow) is for revision and would rather be doing that than this report so I want it done and in. I emailed the tutor about this and he said he was happy for me to use a small sample, and well, I have had a week to do this and tested 20 people (10 with BG noise, 10 Without) so considering, I think I've done OK. I'm just at the point now where I'm tired and fed up with it thus arnt thinking straight and it's all a bit of a jumble, also having never covered MANOVA's before it's a dark area.

Hope this explains why I have such a rubbish sample size, and thanks for the site I'll look at it now

Cheers
Cas

[baz1985]

I found this to be useful.

Field, A. P. (2009). Discovering statistics using SPSS: and sex and drugs and rock 'n' roll (Third Edition). London: Sage.

The 2005 2nd edn is pretty much the same.