hi,all,
One more question on the Anova analysis of multi-level conditions.
I have three task conditions A1, A2, A3, each condition was one level of a co-variate A, and a baseline condition.
If I want to get the main effect of co-variate A, i.e., to see any difference between any task conditions, how can I do it? Is it possible to define the main effect of A directly in the first-level analysis, and then enter to the second-level analysis using one-sample t-test?
I know one possible way is to get the contrast file for each task condition, i.e., A1-base, A2-base, A3-base, and enter them to the second-level one-way anova analysis. Which test, t or f, should be used? And, what is the difference between this way and the upmentioned individual way? Which one is better?
Thanks for any suggestion.
best
hengyi
----------------------------
>1.) Why do you need a 2-way ANOVA? The order of the 1st level regressors is 1=pre hrf, 2=pre temporal derivative(TD), 2=post hrf, 4=post TD. So, [1 0 -1 0] tests hrf and [0 1 0 -1] tests TD. If so, you get two con images. Why would you need a 2-way ANOVA for this? Isn't it a oneway within-subjects ANOVA with 2 levels?
>2.) Is it true that the 1st level t-contrasts test only for greater activations before than after (ie pre minus post)? If one wanted to test both tails (ie pre>post or post>pre), would it be correct to run F contrasts on the first level and take these to the second level?
>
>Best,
>Stefan
>
>
>Mirta -
>
>Simply create T-contrasts for [1 0 -1 0] (canonical: before - after) and [0 1 0 -1]
>(derivative: before - after) and enter the resulting contrast images into a (two-way)
>ANOVA (no constant term; no within subject effects), and evaluate the F-contrast
>for the effects of interest.
>
>Rik
>
>
>----- Original Message -----
>From: "Mirta Villarreal" <[log in to unmask]>
>To: <[log in to unmask]>
>Sent: Wednesday, December 15, 2004 7:08 PM
>Subject: [SPM] 2nd level analysis with F test?
>
>
>Hi all
>
>I want to perform a pair-t-test within a group of subjects before and after some
>treatment. The problem is that for the individual analysis I get better results when I see
>at the [1 0, 0 1] F contrast (model = hrf + first derivative).
>In previous e-mails I was recomend to evaluate the T-contrasts [1 0] and [0 1] at the
>individual level first and take them to an ANova model. I did it and it was ok.
>
>But how can I take this two contrast for each group to a pair t-test? I should choose 4
>pairs? (pair 1= hrf before tretment, pair 2= TD before, pair 3= hrf after, pair 4= TD
>after) or I can use directly the [1 0, 0 1] F contrast and 2 pairs?
= = = = = = = = = = = = = = = = = = = =
|