To construct a residual plot, we use the same xaxis as the scatterplot, but we use a vertical axis of residual values. For an individual sample point, the residual is the value of the y coordinate minus the predicted value of y that is obtained by substituting the x value into the regression equation. The residual is the vertical “error” or difference between the y coordinate that is observed and the y coordinate predicted from the regression equation.
When analyzing a residual plot, look for a pattern in the way the points are configured, and use these criteria:
