0
$\begingroup$

I have run a psychophysical experiment where there are three conditions. All participants undergo each condition. The task is to search for a target on a display and button press when found. Eye movements are tracked. The target was red, blue and green in the conditions respectively. There were 100 trials for each condition and these were done in blocks, although the ordering of the blocks was random. I am measuring time taken to the find the object, and I will process this to be Gaussian.

  • Should I analyse the results using an ANOVA or a repeated measures ANOVA?
Improve this question
$\endgroup$

2 Answers 2

Reset to default
1
$\begingroup$

It will be repeated measures because each subject has 3 observations (you’ve randomized blocks to control for order effects, so I assume you’ll average over them).

That’s true for the behavioral data and also for the eye tracking data, but it’s not clear how you’re dealing with that (good luck - it’s not trivial!)

Improve this answer
$\endgroup$
1
  • $\begingroup$ Thank you so much for your answer. What do you mean by average over them? There will be three means, one for each condition - is that what you mean? If not then I may be missing something important $\endgroup$
    – keyserSoze
    Commented Apr 11, 2018 at 1:44
1
$\begingroup$

If the participants are sampled more than once, ie: they undergo each condition, you have to run a repeated measures ANOVA since polling a subject multiple times means you no longer can make the assumption of independent sampling, so you have to correct per the subject. At least this is what I was told since I'm dealing with this very same issue at the moment. Where I have 9 subjects tested over the course of 10 days with 8 varying stimuli and I measure the response to each stimuli from each subject. Since I'm using the same population each day and for each stimuli I need to run a repeated measures ANOVA.

Improve this answer
$\endgroup$

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.