where SSE is defined as the unweighted sum of squared prediction errors for all of the conditions and TSS is the sum of squared deviations around the mean of the observed data (averaged across conditions). Considering our example, SSE is defined by Equation 4.4 with the weight View the MathML sourceand TSS is also defined by Equation 4.4 with View the MathML source except that the prediction, p G(t ), from the PVL model is replaced with the mean