A histogram can reveal an unusual observation on one variable. However, sometimes a point stands out not on a single variable but instead on a combination of values on two or more variables. For this reason, approaches such as those outlined previously are not sufficient to identify unusual observations. Here, we will see how to use IBM SPSS Statistics to detect multivariate outliers.
Here is a scatterplot of simulated data following a bivariate normal distribution:
The points generally trend from the lower left to upper right, but there is one point up and to the left by itself. Note that this point would not ...
