0
$\begingroup$

I am currently checking for independence on the characteristics of the finger, based on gender, finger and hand. My lecturer recommended using anova test, but how do you implement it on outputs that are alphanumeric? Are there more appropriate tests? I have used the chi squared test for testing independence of genders and hands seperately but was told it was not appropriate. Could someone explain why? Any resources for any recommendations would be greatly appreciated.

$\endgroup$
3
$\begingroup$

Log-linear models may be useful. They are used like multi-dimensional chi-square tests of association. Also, it sounds like you have a three-dimensional table of counts; the Cochran–Mantel–Haenszel test may be applicable.

$\endgroup$
0
$\begingroup$

ANOVA is not recommended, because like you said, it is a discrete data anda one of ANOVA's assumption is data normality. I dont understand your experiment at all, bu i think you shoul use multinomial logistic regression.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.