Repeated Measures ANOVA Problem

Discuss statistics related things

by Xenophon » Tue Apr 09, 2019 2:57 pm

I'm trying to run a RM-ANOVA with 8 levels (Repeated Measures Factors) but in the assumptions of sphericity there is an indication reporting that "the repeated measures has only two levels. the assumption of sphericity is always met when the repeated measures has only two levels". However, when I'm running the same test with 7 levels this indication doesn't appear (in the same database). Is this a problem of JAMOVI? If yes, could you fix it?

In case that you need the database, I'd be happy to provide it to you.

Thanks in advance.
Xenophon
 
Posts: 7
Joined: Fri Mar 29, 2019 3:22 pm

by reason180 » Sat Apr 13, 2019 3:53 am

In ANOVA, the number of levels a repeated-measures factor has isn't the same as the number of repeat-measures factors there are. For example, a design might have a single repeated measures factor--week--with the eight levels of that factor being Week 1, Week 2, Week 3, Week 4, Week 5, Week 6, Week 7, and Week 8. In contrast, the ANOVA might contain eight repeated-measures factors, with only two levels per factor. Example:

Factor 1: Year (2017 and 2018)
Factor 2: Time of Day (morning and evening)
Factor 3: Nutrient Type (fat and carbohydrate)
[ . . .]
Factor 8: Amount Consumed (low and high).

To do a sphericity test on a particular factor, that factor must have more than two levels.
reason180
 
Posts: 27
Joined: Mon Jul 24, 2017 4:56 pm

by Xenophon » Tue Apr 23, 2019 7:56 am

I'm running a post-hoc analysis in the RM-ANOVA and the SE and df remain the same in all comparisons. Is this normal?

However, when I'm running the same tests in SPSS or JASP the SE and df are different (JASP and SPSS results in postdoc analysis coincide).
Attachments
Στιγμιότυπο 2019-04-23, 10.52.03 πμ.png
Στιγμιότυπο 2019-04-23, 10.52.03 πμ.png (95.74 KiB) Viewed 687 times
Xenophon
 
Posts: 7
Joined: Fri Mar 29, 2019 3:22 pm

by jonathon » Wed May 01, 2019 8:42 am

hi xenophon,

yes, that's normal. the post-hoc tests in jamovi are based on estimated marginal means:

https://cran.r-project.org/web/packages ... asics.html

cheers

jonathon
User avatar
jonathon
 
Posts: 896
Joined: Fri Jan 27, 2017 10:04 am


Return to Statistics