How is d from paired t-test calculated?

Discuss the jamovi platform, possible improvements, etc.

by RickCarlsson » Tue Apr 09, 2019 6:38 pm

First: Amazing job. Jamovi is absolutely fantastic. We are retiring SPSS department wide.

Now to my question: How is d calculated for paired sampe t-tests? There are several ways and knowing which is important.

In general it would, especially for teaching, be great to learn the mathematical formula behind different tests. Is there some simple way to access that? It’s not detailed in the R package.
RickCarlsson
 
Posts: 6
Joined: Thu Apr 04, 2019 10:04 am

by jonathon » Wed Apr 10, 2019 12:57 am

hi rick,

great you guys are able to move beyond spss.

at present, the easiest way is to dive into the scripts underlying each analysis. here's the relevant section for cohen's d for the paired samples t-test:

https://github.com/jamovi/jmv/blob/mast ... .R#L54-L66

longer term we want to provide good documentation of course.

with thanks

jonathon
User avatar
jonathon
 
Posts: 1729
Joined: Fri Jan 27, 2017 10:04 am

by RickCarlsson » Thu Apr 11, 2019 6:28 pm

This works for me :) Thanks!
RickCarlsson
 
Posts: 6
Joined: Thu Apr 04, 2019 10:04 am

by Nat » Sat Mar 20, 2021 7:25 am

Hi, I would like to know is there any way to calculate effect size relative to raw score instead of difference score, in paired-sample t-test.

on page 272 of the latest version of user-guide:
" To the extent that you care about the practical consequences of your research, you often want to measure the effect size relative to the original variables, not the difference scores (e.g., the 1% improvement in Dr Chico’s class over time is pretty small when measured against the amount of between-student variation in grades), in which case you use the same versions of Cohen’s d that you would use for a Student or Welch test. It’s not so straightforward to do this in jamovi; essentially you have to change the structure of the data in the spreadsheet view so I won’t go into that here, but the Cohen’s d for this perspective is quite different: it is 0.22 which is quite small when assessed on the scale of the original variables."

Firstly, could you please specify what do you mean by "against the amount of between-student variation in grades"?

Secondly, I can't find the jamovi file mentioned in note 12: "If you are interested, you can look at how this was done in the chico2.omv file" (I would like to see how you "change the structure of the data")

Thanks a lot!
Nat
 
Posts: 5
Joined: Sat Mar 20, 2021 7:16 am

by jonathon » Sat Mar 20, 2021 7:53 am

hi, could you start a new thread?

i'm unfamiliar with this user guide i'm afraid -- i'm not actually sure what you're refering to.

cheers

jonathon
User avatar
jonathon
 
Posts: 1729
Joined: Fri Jan 27, 2017 10:04 am

by Nat » Sun Mar 21, 2021 5:11 am

oh sorry about that...

sure, let me make a new post
Nat
 
Posts: 5
Joined: Sat Mar 20, 2021 7:16 am


Return to General