Model: SBP - b0 + bl (SVmR) + bz(SVmR) 2. Data: The square of SVmR would be calculated. The values for SBP, SVmR, and SVmR 2 would be entered. Interpretation: The final predictive equation is SBP = -2.92 +

289.00(SVmR) - 414.42(SVmR) 2. The fit is shown in the figure. The p-values for tests of both models are less than 0.001, which is significant.

The modeled prediction accounts for the majority of variability in both models; SVmR is a major predictor in either model. Moving from the simple model to the parabolic model increases the R 2 from about 53 % to about 68%, indicating that the curved model is the better fit.