How to identify multivariate outliers in SPSS using squared Mahalanobis distance

[@zohalh14]

Thanks for the video! Can you use Mahalanobis distance if your IVs are categorical in a mixed anova?

[@jochan0]

Dr Crown, thanks for the tutorial. if there are covariates. are they to be checked as IV as well? your kind support is appreicated.

[@talzabidi1569]

why we don't check the outliers for Depend variable?

[@mikecrowson2462]

Hi there Talal. Thanks for your question. You still need to check for outliers on the dependent variable in multiple regression (usually via inspection of residuals - e.g., standardized or studentized). The squared Mahalanobis distance is used in the context of regression diagnostics to identify potential outliers on the set of independent variables (basically those cases that are higher leverage points) that may be influential with respect to the regression parameters. The reason I do not go into the topic of inspection for outliers on the DV (via use of residuals) is because the aim is to focus on the more general approach to obtaining squared Mahalanobis distance in SPSS through the regression module. However, if you are wanting to learn more about identification of outliers related to the dependent variable in the context of multiple regression, go feel free to visit my video on multiple regression here: kzbin.info/www/bejne/jWGlqJ-Qq5uIopI I hope this clarifies things for you! Cheers!