Regression Within & Between Subjects - Long Format

Whirly123 · Post by **Whirly123** » Thu Jul 11, 2019 2:39 pm

I am just wanting to double check that my Long Format data isn't counting data points more than once.

To keep explanation simple, I am running a Regression and my study design has a repeated measure where participants get to donate to different charities and there are 9 charities. So for each participant, there are 9 donation measures and charity type is the IV.

Each participant also filled in some personality inventories but because its long format their score is repeated 9 times. I just wanted to double check that Jamovi is taking this into account and not counting this 9 times or there is some other way I should be doing this?

Post by **jonathon** » Sat Jul 13, 2019 4:06 am

hi,

that is one of the drawbacks of 'long format'. you'll need to convert it to wide format to avoid the issue you describe (unfortunatly, there's no way to do this in jamovi at present).

cheers

jonathon

Whirly123 · Post by **Whirly123** » Tue Jul 16, 2019 9:28 am

Oh, I have it as wide format but then I can't do a regression with my repeated measures?

Post by **jonathon** » Tue Jul 16, 2019 10:58 am

ah, so you can't use a 'normal' regression (regression -> linear regression) if you've got repeated measures. you can either use an RM ANOVA (and add covariates), or use the mixed models in the GAMLj module.

does that make sense?

jonathon

Whirly123 · Post by **Whirly123** » Tue Jul 16, 2019 2:15 pm

It does, thank you

jamovi