Test value for independent and paired t-test

[arcaldwell49]

Just a request, but would it be possible to add the "Test value" option for independent and paired t-tests?

This would allow for users to test hypotheses that are not zero (essentially the same as setting the mu argument/parameter in the t.test function of base R). I'm happy to help in this if the team could point me in the right direction. My guess is the biggest obstacle may be in calculating the Bayes Factor?

[jonathon-love]

hmm, my feeling is a test value for independent samples and paired samples might be too obscure. we try and limit ourselves to common settings to avoid clutter. a work around would simply to use the 'computed variable' system to offset one group of observations

[arcaldwell49]

Oh, I hadn't thought of that. That will definitely work for the paired samples t-test because you can then calculate a column of the numerator for the t-test (mu1-mu2-testValue). I think it would be a little complex for the independent samples t-test.

I think I figured out how to do this in my forked version. I will put in a pull request when I think I have it working. Worst case scenario you can just reject it. Just trying to do my part and help out jamovi!

[jonathon-love]

oh yeah, independent is a little trickier. try this:

[arcaldwell49]

Wonderful! That will certainly help in the immediate term. I put in a pull request for my modifications to the package.

Very happy to continue helping develop this extension if you all are interested!