GLM or one way ANOVA for withing groups data?

[pegah]

Hi,

i have problem with data analysis method.

my DOE:

 3 factors with 2 levels for 2 of them and 4 levels for the other one (i.e. 16 treatments in 4 groups)
group one:
1 1 1
1 1 2
1 1 3
1 1 4
group 2:
1 2 1
1 2 2
.
.
group 3:
2 1 1
2 1 2
.
.
group 4:
2 2 1
2 2 2
2 2 3
this treatment was unacceptable. no data recorded for that.

my goal:

  ANOVA for testing main effects and interactions for significance. i mean, i want to test all treatments data with each other.

my GLM output: 1- a Tests of Between-Subjects Effects table. 2- Estimated Marginal Means table. 3-Post Hoc Tests for the factor with 4 levels.

my problem:

 the GLM gave me an output which there were no post-hoc test within groups. i can't find how to analyse the significance within each groups data and one data with other data in another group. does it need to run a separate one-Way Anova?

[Maguin, Eugene]

I'm struggling with your English syntax so I'm not sure I completely understand what you want to do. I think this would be your GLM/Unianova statement:
GLM DV by A B C.   or  Unianova DV by A B C.  Of course you might add a print command get means and SDs, effect size, coefficients, etc.

Have you tried the Emeans command?

This part '... i can't find how to analyse the significance within each groups data and one data with other data in another group.' Is what is not understandable to me. In terms of my syntax above what specific comparisons do you want to make. Assume A and B have two levels each (A1, A2, B1, B2) and C has four. What are the comparisons?
Gene Maguin


-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of pegah
Sent: Friday, August 02, 2013 5:46 AM
To: [hidden email]
Subject: maybe simple problem!

Hi,

i have problem with data analysis method.
my DOE: 3 factors with 2 levels for 2 of them and 4 levels for the other one (i.e. 16 treatments in 4 groups) my goal:  ANOVA for testing main effects and interactions for significance my problem: in GLM, i can't find how to analyse the significance within each groups data and one data with other data in another group. does it need to run a separate one-Way Anova?



--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/maybe-simple-problem-tp5721455.html
Sent from the SPSSX Discussion mailing list archive at Nabble.com.

=====================
To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[pegah]

hi
 thank you for answering and sorry about my English,

you mentioned "Emeans command" and it simply helped me.

it was about "simple effects" pairwise comparison that spss glm does not do it by default.

thanks again.

Pegah