Refer to Table 12.1, in which the relationship between x = mathematics achievement test score and y = final calculus grade was studied.

1. Enter the values for x and y into the first two columns of a MINITAB worksheet. 2. The main tools for linear regression analysis are generated using Stat Regression Regression. (You will use this same sequence of commands in Chapter 13 when you study multiple regression analysis.) The dialogue box for the Regression command is shown in Figure 12.19(a). 3. Select y for the "Response" variable and x for the "Predictor" variable. You can now generate some residual plots to check the validity of your regression assumptions before you use the model for estimation or prediction. Choose Graphs to display the dialogue box in Figure 12.19(b). We have used Regular residual plots, checking the boxes for "Normal plot of residuals" and "Residu- als versus fits." Click OK to return to the main dialogue box.

4. If you now choose Options, you can obtain confidence and prediction intervals for either of these cases: A single value of x (typed in the box marked "Prediction intervals for new observations"). Several values of x stored in a column (say, C3) of the worksheet. 5. Enter the value 50 in Figure 12.20

(a) to match the output given in Figure 12.13. When you click OK twice, the regression output is generated as shown in Figure 12.20(b). The two diagnostic plots will appear in separate graphics windows.

Transcribed Image Text:

Student Mathematics Achievement Test Score (x) Final Calculus Grade (y) 39 43 65 1234567 8. 9 10 78 21 52 64 82 57 92 47 89 2232 28 73 75 98 34 56 52 75