Adjusted standardized residuals to chi square tests

[andremonezi]

It would very interesting if JASP could provide standardized/adjusted standardized residuals for a posteriori comparisons when running chi-square tests.

This feature is quite important mainly in those cases with contingency tables 3X3 or more. Regards,

[boutinb]

@EJWagenmakers What do you think? Is is something for Jiashun?

[andremonezi]

Dear all, what do you think about that?

[EJWagenmakers]

I'm seeing some other options we could add to the contingency table. So it's worth an upgrade. About the standardized residuals, this document describes a series of options with some really nice plots as well: http://www.sthda.com/english/wiki/chi-square-test-of-independence-in-r

[NenaRoncevic]

Dear colleagues, this is just to support idea to add options on standardized/adjusted standardized residuals in chi-square tests in future iterations of JASP. Thank you.

[tomtomme]

@andremonezi We have those residuals now, but we do not have the p-values for those post hoc tests.

@Kucharssim If we want to add those p-values, here the references from the almost duplicate #1442

• https://cran.r-project.org/web/packages/chisq.posthoc.test/vignettes/chisq_posthoc_test.html
• https://github.com/jasp-stats/jasp-issues/files/8752026/chisq.posthoc.test.pdf
• https://stackoverflow.com/questions/52656990/residual-analysis-based-on-the-results-of-chi-squared-test-of-independence-in-r