Much of the confusion surrounding interpretation and application of th
e coefficient of determination, R2, can be alleviated if it is defined
explicitly as a comparison of a given model to the null model EY = be
ta0. The model-comparison definition allows R2 to be easily generalize
d, and standard extensions such as coefficients of partial determinati
on are seen to be special cases of this generalization. Formulas becom
e simpler, more unified, and more easily understood. Commonly cited pr
oblem areas such as R2 for the no-intercept model and model comparison
s using different values of R2 are also clarified by this perspective.