Three factor nested design ANCOVA

Moderators: statman, Analyst Techy, andris, Fierce, GerineL, Smash

friedrichshain
Posts: 1
Joined: Fri Apr 06, 2012 12:19 pm

Three factor nested design ANCOVA

Postby friedrichshain » Fri Apr 06, 2012 1:52 pm

Hi all,

I definitly need your help for evaluating my experiment:

I run a field experiment in which I designed 5 different groups:
1 Control group (label Treat, coded 0 for control, 1 for any kind of treatment)
2 Factorlevel A (coded 1 for S, 2 for R, and 0 for control)
2 Factorlevel B (coded 1 for L, 2 for H and 0 for control)

I measured post-level of a DV and treat pre-level of this DV as a covariate (v_555_0).

Because this is not a fully crossed design, a nested design approach must be appropriate.
First of all, do you think this will be appropriate?
And if so, how could I handle this in SPSS?
I ran:
ATASET ACTIVATE DatenSet1.
UNIANOVA v_555__1 BY Treat GruppeSR GruppeLH WITH v_555__0
/METHOD=SSTYPE(3)
/INTERCEPT=INCLUDE
/EMMEANS=TABLES(Treat) WITH(v_555__0=MEAN) COMPARE ADJ(BONFERRONI)
/EMMEANS=TABLES(GruppeSR) WITH(v_555__0=MEAN) COMPARE ADJ(BONFERRONI)
/EMMEANS=TABLES(GruppeLH) WITH(v_555__0=MEAN) COMPARE ADJ(BONFERRONI)
/PRINT=HOMOGENEITY
/CRITERIA=ALPHA(.05)
/DESIGN=v_555__0 Treat GruppeSR(Treat) GruppeLH(Treat).

I cannot run planned contrast, but Bonferroni works.
Do you think this is the right procedure? Or do I have to evaluate my design totally different?

Thanks a lot for your help and effort!
Benjamin

Who is online

Users browsing this forum: No registered users and 1 guest

cron