The data in Table 9.26 do not arise from a factorial experiment but illustrate how a multiple regression model and a lack of fit test can be used. Thirty-six pine seedlings were randomly divided into...

The data in Table 9.26 do not arise from a factorial experiment but illustrate how a multiple regression model and a lack of fit test can be used. Thirty-six pine seedlings were randomly divided into nine treatment groups. Each group received a different concentration of a complete nutrient solution (in ppm) for a period of several weeks. The response is growth in millimeters during a two-week period. Note that the levels of nutrient are not equally spaced. A curve showing the response to the nutrient solution may be used to determine an economic optimum amount to use.