Page 1 of 1

Discrepancy Jamovi / R

Posted: Fri Sep 09, 2022 9:28 am
by fhauw
Hello everyone,

I just discovered this amazing Jamovi recently, and wanted to store all the R scripts resulting from the analyses made through the UI in a R script.
But when I launch these scripts on R, I get different stats than those I get using Jamovi, with exactly the same data, and without changing any code.
Has someone had the same issue, and if yes, how to solve it?
Thanks.

Best,
Fabien

Re: Discrepancy Jamovi / R

Posted: Wed Sep 14, 2022 12:09 pm
by Frank.M.LoSchiavo
Hi Fabien. I'm curious about what you're finding. Is it possible to share a bit of the code you're using or an example of the differences you're seeing?

Thx.

FML

Re: Discrepancy Jamovi / R

Posted: Fri Sep 16, 2022 2:24 pm
by fhauw
Hello,
Sorry for the late answer.
Trying to be simple: I'm comparing 2 populations for error rates and reaction time on multiple experiments.

When I get on syntax mode on jamovi, it gives me this code:

Code: Select all

gamlj::gamljGlmMixed(
    formula = Errors ~ 1 + Group_Id + Num_Lett + Group_Id:Num_Lett+( 1 | Item )+( 1 | Subj_Id ),
    data = data)
It gives me this model and those results on jamovi (screenshot):
https://postimg.cc/8JWtMSyp

Then on R:

Code: Select all

glme1 <- gamlj::gamljGlmMixed(
  formula = Errors ~ 1 + Group_Id + Num_Lett + Group_Id:Num_Lett+( 1 | Item )+( 1 | Subj_Id ),
  data = res_table)
Model and results:
https://postimg.cc/688R8N5y

Let me know if you would need more details,

Best,
Fabien

Re: Discrepancy Jamovi / R

Posted: Sun Sep 18, 2022 6:35 pm
by mcfanda@gmail.com
Num_Lett is defined as a factor in "res_table" and as a continuous variable in "data"