nav[aria-label="Primary Navigation"] { padding: 0; & ul { list-style: none; width: 100%; display: flex; flex-direction: row; justify-content: start; align-items: start; gap: 30px; padding: 0; & li { margin: 0; } & ul li { list-style: none; } } }

Cramer Correlation

flo_kussener

Hey everybody,

A JMP Pro user would like to use the Cramer's V correlation (http://en.wikipedia.org/wiki/Cramér's_V). I didn't find it in the JMP documentation (neither File E.

Is anybody know the specification of this correlation and how we could do something similar in JMP Pro?

Thank you,

Florence

Accepted answers

All comments

jiancao

JMP doesn't have it, but it is fairly easy to calculate it using JMP Formula or by hand since all the ingredients are available in Fit Y by X.

This is Cramer's V formula as shown in the wiki

$V = \sqrt{\frac{\varphi^2}{\min(k - 1,r-1)}} = \sqrt{ \frac{\chi^2/n}{\min(k - 1,r-1)}}$

where $\chi^2$ is Pearson's Chi square, n is total # of obs, k and r are the # of columns and # of rows, respectively. Using JMP's Big Class, run Fit Y By X with Y=Age and X=Sex, Cramer's correlation is

V=Sqrt(2.554/40/1) =0.252686.

No need for JMP Pro, but i agree that it'd be nice to add it to list of measures of association.

Chris_Kirchberg

Thanks Jian,

When would someone use Cramer's V over Chi Squared? What are the pros and cons?

Thanks

Chris Kirchberg

jiancao

Pearson's Chi square is a testing statistic, while Cramer's V is a measure of association varying from 0 to 1 (analogous to a F-statistic and a correlation coefficient, r, for continuous variables). The p-value for the significance of V is the same for the Pearson's chi-squared test.

Chris_Kirchberg

Thanks

Chris Kirchberg

flo_kussener

Thank you Jian for these details.

Yes Chris, it would be nice.

Florence

De : JMP User Community Administrator <jmpcommadmin@sas.com<mailto:jmpcommadmin@sas.com>>

Répondre à : "jive-1049333682-hwqt-2-4mjn@community.jmp.com<mailto:jive-1049333682-hwqt-2-4mjn@community.jmp.com>" <jive-1049333682-hwqt-2-4mjn@community.jmp.com<mailto:jive-1049333682-hwqt-2-4mjn@community.jmp.com>>

Date : jeudi 16 avril 2015 19:45

À : Florence Kussener <Florence.Kussener@jmp.com<mailto:Florence.Kussener@jmp.com>>

Objet : Re: - Cramer Correlation