1
$\begingroup$

I currently have data that looks at participants receiving 10 stimuli successively, but the order the stimuli are presented changes with each participant. What type of analysis should be used to look at this. What do you do with the repeated measures ANOVA when the stimuli presented are in a random order?

$\endgroup$

1 Answer 1

2
$\begingroup$

Not much changes after you randomize the stimuli, the only thing it could affect is your covariance structure (which it probably doesn't unless you specified that manually in your statistical program, because it is usually compound symmetric otherwise known as exchangeable by default). This is because a repeated-measure ANOVA tests the null hypothesis $(\mu_1 = \mu_2 = \mu_3 = ...=\mu_n)$, where $n$ is your number of stimuli, and this equality is invariant after any permutation of the hypotheses. So you could do a repeated measures ANOVA (after testing the sphericity assumption), a mixed model or MANOVA. The difference between these approaches have been addressed, for example, in this question: Differences between MANOVA and Repeated Measures ANOVA?.

$\endgroup$

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.