[Bug]: how exactly is a differing sample size per group handled?

[renatajorge78-dotcom]

JASP Version

0.95.2

Commit ID

No response

JASP Module

Power

What analysis are you seeing the problem on?

Power analysis comparing two proportions

What OS are you seeing the problem on?

macOS Intel

Bug Description

I use JASP to teach dental undergraduates and I'm very pleased with it (thank you all!) but I always get stuck on the power analysis for comparing two proportions regarding the sample size input in relation to p1 and p2. I initially assumed that p1 should match N1 and p2 should match N2 but that doesn't seem to always add up and it's also not clear what to input in Sample size per group (N) when the two groups have different sample sizes - I usually start with N1 (and match that with p1) but end up having to fiddle a bit with N and the proportions to finally get them to match.
On the independent sample t-test, it's again not super clear what the Sample size per group (N) means when the two groups have different sample sizes - I believe we need to report N1, but I always have to go by trial and error, although it's never as complicated as the proportions one.
I'm not sure if this is a bug or I might be missing something obvious, apologies if it's the latter, but if you could clarify it for me, I'd be very grateful.

Expected Behaviour

1. p1 should match N1 and p2 should match N2
2. It should be clearer what Sample size per group (N) is supposed to be, particularly when the two samples are not equal

Steps to Reproduce

I can't really reproduce this, sorry!

Log (if any)

No response

More Debug Information

No response

Final Checklist

• I have included a screenshot showcasing the issue, if possible.
• I have included a JASP file (zipped) or data file that causes the crash/bug, if applicable.
• I have accurately described the bug, and steps to reproduce it.

[github-actions]

@renatajorge78-dotcom, thanks for taking the time to create this issue. If possible (and applicable), please upload to the issue website (#3728, attaching to an email does not work) a screenshot showcasing the problem, and/or a compressed (zipped) .jasp file or the data file that causes the issue. If you would prefer not to make your data publicly available, you can send your file(s) directly to us, issues@jasp-stats.org

[tomtomme]

Can you clear up which test exactly you mean? This one?

Regarding your question: p is for proportion and n for sample size, so no - those would not match, why would they?

The other question: I agree, that "sample size per group" is not clear, if those differ in size

[renatajorge78-dotcom]

Yes, that's the one. If you are asked to provide p1 (proportion of group 1) and p2 (proportion of group 2) and then you are asked to give the sample size proportions as N1/N2, yes one would expect p1 to match N1 and p2 to match N2, just by simple logic. Why would you consider group 1 for proportion to be group 2 for sample size and vice-versa?

[tomtomme]

ah sorry - I misunderstood the word "match" in this context - not a native speaker here. Of course n1 and p1 belong together and n2 p2. I thought you were saying that they are identical things....

[tomtomme]

the remaining issue then is that "sample size per group" is not clear, if those differ in size - correct? And you probably ask this for the post hoc scenario to calculate the power