Interpreting Polynomial Regression

This paper focuses on the interpretational difficulties that confound polynomial regression analysis. While polynomial regression is statistically sound, it produces awkward equations which "describe" a curve with a series of linear slopes. An example illustrates the technique and the awkward properties of regression coefficients in the quadratic case. Finally, the paper briefly outlines an algebraic manipulation that transforms polynomial regression equations into a format that is readily interpretable.