next »

[annmarie]

I know that this should go on the research questions thread, but I am frustrated and few are reading the research questions on a Sunday night.

I am analyzing data with SPSS statistics software.  I have used other software and am trying to find the features i need in SPSS.  I did a nested analysis before

I have a mixed within and between subjects design. 
8 subjects are in Group 1
8 subjects are in Group 2
8 subjects are in Group 3

There are two within subjects independent variables.  I believe that the different groups will  respond differently to one of the independent varaibles.

Can anybody help me do the nested ANOVAs?

[mirandaf]

I might be able to help tomorrow when I'm back in the office and have access to SPSS (which I don't here at home).

If I'm understanding you correctly, you've done nested ANOVAs in other software (e.g., SAS or STATA) and now you want to do it in SPSS?

So you want to compare the means of group 1 against group 2 against group 3? You essentially want to run a series of T-tests, only all at once? If I remember......... I think it's something like Analyze-Compare Means - ANOVA (maybe?). And there's something about the grouping variable. I've used ANOVA (and ANCOVA, analysis of covariance) when analyzing evaluation results, i.e., comparing Group 1 that got 2 years of the Treatment with Group2, that only got 1 year of Treatment; and with group 3, which was the no-treatment Control group. Into the 'grouping variable' box I clicked over the various treatment levels.

I've only really worked with SPSS, and my impression is that it's a little user friendly than other statistical packages because you (generally) don't have to write code. It's basically all point-click, point-click, point-click. If you do want to insert some code, you can do this through File-New-Syntax, and then Run (I think).

Of course, if I've misunderstood what you're trying to do, then I probably haven't helped at all.

[annmarie]

Mirandaf,
Thank you.  I do think that you understand.  I have done this in other software but have to do it in SPSS.  I think that the  Analysis>Mixed might be where to go on the menu. 

Like you thought, I have Groups 1,2, and 3.  I also have Factor A and Factor B that are fixed factors for all 3 groups. 

AnnMarie

[msparticularity]

On my version, you want Analyze - General Linear Model - Multivariate. I've got 12 or 13, but I don't think it's changed all that much.

[mirandaf]

Annmarie: What version of SPSS are you working with? 15.0?

[oldfullprof]

Hi.  You have to use repeated measures for the within subjects, even if there is no time difference.  Get the excellent SPSS manual for more details.

[annmarie]

Thanks to all of you who looked into this for me. 

Mirandaf,
I am using SPSS Version 16.  I was able to dowload it for 14 days.  The statistics packages that I used for my dissertation are less common and I wanted to do this work in SPSS.
AnnMarie