Thank you for this useful video. I have a question please : What to do in case of ordinal variables when checking for these outliers ? what method is the adequate one? Mahala Distance or Cook's Diastance ? Does it have sense to apply this method when my data is only composed with ordinal variables and not continuous ones ?
Univariate outliers don't make any sense on an ordinal scale because there can't be any extreme values. Multivariate outliers can occur though. For these, I strongly recommend Cook's distance instead of Mahalanobis (which is too strict).
Thank you for this very informative video, Dr. Gaskin!
Dr James, the best. Sir.
Thanks Prof Gaskin..I did my analysis in my thesis to learn from your video of Smart PLS 3.0..
I learnt alot from this. Thanks sir
Thank you for this useful video. I have a question please : What to do in case of ordinal variables when checking for these outliers ? what method is the adequate one? Mahala Distance or Cook's Diastance ?
Does it have sense to apply this method when my data is only composed with ordinal variables and not continuous ones ?
Univariate outliers don't make any sense on an ordinal scale because there can't be any extreme values. Multivariate outliers can occur though. For these, I strongly recommend Cook's distance instead of Mahalanobis (which is too strict).
Is it possible to use the cook distance for more than 1 dependent variables?
I don't think so. Not in SPSS. SPSS simple linear regression handles only one DV at a time.
Sir wt if we have 2 dependent variables and a mediator.then which of the variables should be put as dependent variable while checking cooks distance??
You can try them both separately.