That is, the total sums of squares (i.e., the overall variability of the outcome variable) can be decomposed into two parts: the variability associated with the model SS M , and the residual or leftov...That is, the total sums of squares (i.e., the overall variability of the outcome variable) can be decomposed into two parts: the variability associated with the model SS M , and the residual or leftover variability, SS R . However, it’s kind of useful to rearrange this equation slightly, and say that the SS value associated with a model is defined like this…
