As part of exploratory analysis of multivariate data, it can be very helpful to plot the data and look at the descriptive statistics to see patterns in the data and understand the variability of individual variables. It can also help detect if there are extreme or outlying samples in the data. Box plots of each variable are generated when doing descriptive statistics. This analysis is accessed by going to Tasks – Analyze – Descriptive Statistics.

In the dialog box, one defines the data for the analysis and can also choose to calculate the correlation matrix for the data. The output of descriptive statistics includes the main plots of the quantiles as well as mean and standard deviation for each variable. The quantiles plot is a box plot with whiskers where the bottom value is the low, middle the mean, and top – maximum value for the variable with the data set. The box limits represent the 25-75% percentile values for the variables in the sample set.