weight, with the a (intercept) and b (slope) parameters having been left out of the model. This is a

reminder that although the goal is to evaluate the SBP = weight model conceptually, in reality, this

relationship will be numerically different with each data set we use.

Evaluating residuals

The residual for a point is the vertical distance of that point from the fitted line. It’s calculated

as

, where a and b are the intercept and slope of the fitted straight

line, respectively.

Most regression software outputs several measures of how the data points scatter above and below the

fitted line, which provides an idea of the size of the residuals (see “Summary statistics for the

residuals” for how to interpret these measures). The residuals for the sample data are shown in Figure

16-5.

© John Wiley & Sons, Inc.

FIGURE 16-5: Scattergram of SBP versus weight, with the fitted straight line and the residuals of each point from the line.

Summary statistics for the residuals

If you read about summarizing data in Chapter 9, you know that the distribution of values from a

numerical variable are reported using summary statistics, such as mean, standard deviation, median,