Permutation-based analyses and effect size

[PaulGinns]

Could you incorporate permutation-based analyses, e.g. for 2-group comparisons of experimental data? Unlike traditional statistics, these involve few assumptions to generate a p value.

It would also be great to build in a permutation-based effect size - for experimental designs, I would recommend one developed by Berry and Mielke (1992) - http://epm.sagepub.com/content/52/1/41.short

Abstract: A new measure of association is introduced which measures the degree of association between a nominal independent variable and nominal, ordinal, or interval dependent variables. An extension to multivariate problems provides analyses of multiple dependent variables, including mixtures of interval, ordinal, and/or nominal dependent variables. A commensuration technique is given for standardizing dependent variables. A permutation test of significance is provided for the new measures.

Hopefully,

Paul

[boutinb]

@EJWagenmakers Can you also answer this (quite old) request?

[EJWagenmakers]

(particularly relevant for @JohnnyDoorn) Sorry for the very tardy response. This is certainly interesting! We've been busy with a lot of other issues but this should definitely go on our agenda.

Cheers, E.J.

[PaulGinns]

Hello,

I'm wondering if this suggestion I made a few years ago is going to be considered soon? I think this "minimal assumptions" effect size for experimental data would be a very useful addition to JASP.

Regards,

Paul Ginns

[EJWagenmakers]

We have been implementing a bootstrap test recently, should be in the developers' version -- I'll go and check whether it was for the t-test or something else. E.J.

[EJWagenmakers]

OK, I learned that we are now implementing bootstrap tests for linear regression coefficients and post-hoc tests for ANCOVAs. E.J.

[tomtomme]

While we made great progress implementing bootstraping CIs, it is still missing in

• all t-Tests
• AN(C)OVA: for CIs of contrasts
• Linear Mixed Models: CIs of for slopes, contrasts, marginal means