1
$\begingroup$

I have one DV: performance on research knowledge test (pretest and postest, so repeated measures--interval level measurement). I have two IV: 1) Online textbook or traditional textbook and 2) Onsite campus versus remote campus. Both IV were only measured once, at the beginning of the study.

I think I should be running a repeated-measures ANOVA but I usually only see those where the same group of people was exposed to both interventions, which is not the case here. Is this okay for me to run? Are there other analyses I should be exploring instead? Thank you in advance for any help.

Our main research question is whether Online textbook was as good as or better than regular textbook in terms of the change in scores over time--which one produced more change over time in students' knowledge (interval level).

Alternatively, we could look at it using a regression framework where the posttest is the DV and the pretest is a control variable and we include the two IV as dummy variables. I know how to do this, but I'm wondering if it isn't better to do a repeated measures ANOVA as it accounts better for the linkage between the pretest and the posttest.

Improve this question
$\endgroup$
1
  • $\begingroup$ How is performance on research knowledge test measured, Susan? In other words, what type of variable is performance? Also, are your two independent variables measured just once for each student? Finally, what are your actual research questions for this study? If you could add this information in your original post, that would be helpful to those wishing to answer. $\endgroup$
    – Isabella Ghement
    Commented Aug 14, 2019 at 15:21

1 Answer 1

Reset to default
0
$\begingroup$

I think your confusion arises around the point that you consider pre- vs. post-test as the dependent variable which is not wrong but this is how a MANOVA would conceptualize it not how a rmANOVA does it. A rmANOVA would consider knowledge as the one and only dependent variable with 3 predictors: time point (pre vs. post), textbook type, and campus type - where the first is a within subject predictor (or factor if you want to call it like that) and the other two are between subject predictors.

So, yes, you may run a three-way mixed ANOVA with pre- vs. post as within factor and the other as between factors. If you want to learn a more general technique that helps you in a lot more situations, you may also want to consider linear mixed models (https://ourcodingclub.github.io/2017/03/15/mixed-models.html). In this case, your rmANOVA should actually be equivalent to a mixed model with a random intercept across subjects.

Does this answer your question?

Improve this answer
$\endgroup$
2
  • $\begingroup$ Yes, thank you. I know I should learn GLM, or whatever it's called in SPSS which I'm using, but this is such a tiny study and GLM seems so huge! Thank you! $\endgroup$
    – Susan S
    Commented Aug 14, 2019 at 17:21
  • $\begingroup$ SPSS calls it mixed models, I think. The generalized linear model (GLM) is on another page. Happy to hear it helped! Please feel free to upvote and accept the answer so other people know that this question already has a final answer. $\endgroup$
    – StoryTeller0815
    Commented Aug 14, 2019 at 17:29

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.