Sum of Squares (SS)

The total or corrected sum of squares for a set of data is calculated by taking the difference between each data point from the mean, squaring this difference, and then summing the squared values. In a regression analysis, this total sum of squares is partitioned into portions attributable to the model and to random error.