A Visual Approach to SPSS for Windows 2nd Edition by Leonard Stern

Exercise 1

For this exercise, use the file 04Cars.sav available on the publisher's website for this text at the address http://www.pearsonhighered.com/stern2e. The data set is based on information obtained from Edmunds.com and is reprinted here with permission. Perform a multiple linear regression analysis to predict a car's highway gas mileage (variable HiwayMPG ) from the following variables: Engine , Cylinders , HP , Weight , Wheelbase , and Length. Before performing the analysis, omit the data of the hybrid vehicles (case numbers 69, 70, and 94). Enter all the predictor variables simultaneously. From your printout, answer the following questions.
1. Does it appear that the distributional assumptions of the residuals have been met
2. What is the numerical value of the simple correlation between the dependent variable and each predictor variable
3. Considering just the predictor variables, which two are most strongly related
4. What proportion of variance of the variable HiwayMPG is explained by all the predictors combined Is that proportion statistically significant
5. What is the equation that predicts HiwayMPG from the (unstandardized) predictors
6. Using semipartial correlation values, list, in order of magnitude, the three variables in the equation most effective in predicting the value of HiwayMPG.
7. Are there any cars in the data set that produce residuals that could be considered outliers If so, which car(s)