Feature Request: Chi-Square Using Multinomial Test

[ToddDaniel]

I am creating an assignment on chi-square using the multinomial test in JASP. In the first of my examples, the expected values are random (equal) and in the second example, the expected values are from a comparison population. I would like to request the following additions for the chi-square results in Frequencies > Multinomial Test

1. In the Multinomial Test table, add a column with Phi and Cramer’s V effect size option such as the option that is already in the Frequencies > Contingency Tables > Statistics option.
2. In the Descriptives table, (Additional Statistics > Descriptives) add a column for the residual values, O – E
3. At the bottom of the Descriptives table, add a row that totals each observed and expected column. You need the N of Observed for writing up the results in APA style.
4. Beneath the Descriptives table, add a note for the percentage of the cells in which the Expected values are less than 5. This will help with checking the assumptions of the chi-square test. For example, in SPSS, the notation reads: “0 cells (0.0%) have expected count less than 5. The minimum expected count is 24.77.” Thank you so much for all that you are doing with JASP. This is excellent software.

[EJWagenmakers]

All good suggestions. Guys, who can handle this?

[FransMeerhoff]

Hi @ToddDaniel

Sorry for the late reaction. We are quite busy at the moment releasing JASP 0.12 with all kind of new features. Anyway I've put your suggestions on the backlog, so that we can handle your feature request in the version after 0.12

Kind regards Frans Meerhoff

[ToddDaniel]

Thank you for letting me know.

I am excited to see the new version. I expect that we will see some great additions to JASP. We appreciate your hard work on some incredible software.

Any speculation on when the new version will be released? I am getting ready to make a video tutorial, but I would wait to use the newest version.

Todd

On Feb 21, 2020, at 2:21 AM, FransMeerhoff notifications@github.com<mailto:notifications@github.com> wrote:

Hi @ToddDanielhttps://github.com/ToddDaniel

Sorry for the late reaction. We are quite busy at the moment releasing JASP 0.12 with all kind of new features. Anyway I've put your suggestions on the backlog, so that we can handle your feature request in the version after 0.12

Kind regards Frans Meerhoff

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHubhttps://github.com/jasp-stats/jasp-issues/issues/566?email_source=notifications&email_token=AB4MSQ5JXQMFMDSO4X2XCODRD6FJHA5CNFSM4JU54NV2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEMR42QI#issuecomment-589548865, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AB4MSQY255KJU2DBSPAKK6TRD6FJHANCNFSM4JU54NVQ.

[joefitness]

I am bumping this up for @ToddDaniel suggestions of adding Phi and Cramer’s V effect size and residuals to the Multinomial Test.

JASP has been fantastic! Thank you!

[EJWagenmakers]

I'm assigning this to @FBartos under the general heading "more love for contingency tables"

[brandmaier]

+1 for "more love for contingency tables"

I also miss effect sizes (Cohen's omega) for the multinomial test.

[tomtomme]

related https://github.com/jasp-stats/jasp-issues/issues/2108

Also please add standardized residuals for the multinomial test, as discussed here: https://github.com/jasp-stats/jasp-issues/issues/2801