0
$\begingroup$

I am designing an uncontrolled, prospective study (one group) with continuous outcome variables, which will be measured at 3 points in time (i.e., baseline, timepoint_1, timepoint_2).

If the null hypothesis for a simple pre/post study is that there is no difference between the mean "pre" score and the mean "post" score, what is the null hypothesis for a study with more time points?

Improve this question
$\endgroup$

1 Answer 1

Reset to default
1
$\begingroup$

The null is that the means are the same at all three time points, and this is tested with a repeated measures ANOVA.

Improve this answer
$\endgroup$
5
  • 1
    $\begingroup$ Thank you. If I conduct a repeated-measures ANOVA and the p-value is significant, what test would I conduct to determine where the difference is? $\endgroup$
    – Maldini
    Commented Sep 13, 2021 at 17:33
  • $\begingroup$ @Maldini That might warrant its own question. $\endgroup$
    – Dave
    Commented Sep 13, 2021 at 20:51
  • $\begingroup$ @Dave Is there a resource to help me determine whether a question should be its own question (v. a follow-up question)? I'm a newbie here and don't want to just post a multitude of individual, but related questions. Thanks! $\endgroup$
    – Maldini
    Commented Sep 13, 2021 at 22:36
  • $\begingroup$ On the contrary, I would say that the way Cross Validated (and probably most or all of the other Stacks) works is that it is beat to ask multiple questions. // There’s a way to take the tour that introduces you to Cross Validated, which you might consider doing. I think you even get a badge for it. $\endgroup$
    – Dave
    Commented Sep 13, 2021 at 22:41
  • $\begingroup$ As always, if this answers your question, please mark the answer as accepted, and optionally upvote it. $\endgroup$
    – Eoin
    Commented Sep 14, 2021 at 8:48

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.