0
$\begingroup$

I have three experimental and one control group in my study. I have two observations for them (pre and post test). I want compare four group in those 2 observations.

I choose to do "Repeated Measures-ANOVA" analysis. However I have two questions:

enter image description here

  1. Which contrast should I choose here? Polynomial is okay?
  2. Which post-hoc test should I use? I did an example with Bonferroni and it seems okay. Although some researchers suggest Tukey Hsd, can I compare all four groups in Repeated Measures menu / Post-hoc / choose Tukey or Bonferroni. Or should I perform Tukey test manually?
$\endgroup$
1
  • 2
    $\begingroup$ Repeated measures ANOVA is no longer considered good statistical practice. Take a look at generalized least squares or mixed effects models. $\endgroup$ Sep 9, 2020 at 11:58

1 Answer 1

1
$\begingroup$

The POSTHOC subcommand applies only to between-subjects factors, because many of the methods available there are only for independent samples.

With only two time points, the only thing changing the contrast type on the repeated factor does is to change scaling and/or the sign of a contrast coefficient, because there is only one comparison that can be made between two points.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.