- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content
Hi everyone,
Given thee time points, is it possible to run repeated measure ANOVA without a control group? Basically I have a couple of variables (score1, score2) for the same group of observations and score1 and score2 were measured three times at three different time points after an intervention. I want to see if score1 and score2 show any change pattern over time and if the changes were significant.
Below is the table of the means at three time points:
Variable | N | Mean | ||
T1 | T2 | T3 | ||
score1 | 107 | 26.6 | 27.7 | 27.8 |
score2 | 107 | 19.5 | 20.2 | 19.7 |
I've done something similar before, but still not quite sure whether I should use PROC ANOVA, PROC GLM or PROC MIXED.
Thanks a lot!
Lizi
- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content
Here is an example using PROC GLM
This assumes you have the raw data, and not just the means you showed. The conversion to a logarithmic scale in the example probably isn't necessary for your situation.
Paige Miller
- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content
The thing is that my data does not have any comparison group fir CLASS statement. The purpose of my analysis is to see if the 107 observations experienced any significant change pattern over the three time points on their two measures. That means I only need within effects.
The example you showed actually compares the repeated measures between different groups.
- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content
Oh, I missed that.
I believe that this experiment now ought to be considered a multivariate repeated measures experiment, something I really have no experience with. However, you are correct that there should be no CLASS statement and score1 and score2 belong on the left hand side of the equal sign in the MODEL statement, while Time goes on the right side of the model statement.
But beyond that, I have no experience.
Paige Miller