Data Analysis with Microsoft Excel: Updated for Office 2007

Chapter 9 Multiple Regression 371

Summary of Calc Analysis

What main conclusions can you make about the calculus data, now that you

have done a regression, examined the regression residual fi le, and plotted

some of the data? With an R^2 of 0.37 and an adjusted R^2 of 0.320, the regres-

sion accounts for only about one-third of the variance of the calculus score.

This is disappointing, considering all the weight that college scholarships,

admissions, placement, and athletics place on the predictors. Only the al-

gebra placement score and whether calculus was taken in high school have

signifi cant coeffi cients in the regression. There is a slight problem with the

assumption of a constant variance, but that does not affect these conclu-

sions. You can close your workbook now, saving your changes.

Regression Example: Sex Discrimination

In this next example, you use regression analysis to determine whether a

particular group is being discriminated against. For example, some of the

female faculty at a junior college felt underpaid, and they sought statisti-

cal help in proving their case. The college collected data for the variables

that infl uence salary for 37 females and 44 males. The data are stored in the

Discrimination workbook.

To open the fi le:

1 Open Discrimination from the Chapter09 data folder.

2 Save the workbook as Discrimination Multiple Regression.

Table 9-1 shows the variables in the workbook.

Table 9-1 The Discrim Workbook

Range Name Range Description

Gender A2:A82 Gender of faculty member (F 5 female, M 5 male)

MS_Hired B2:B82 1 for Master’s degree when hired, 0 for no Master’s

degree when hired

Degree C2:C82 Current degree: 1 for Bachelor’s, 2 for Master’s,

3 for Master’s plus 30 hours, and 4 for PhD

Age_Hired D2:D82 Age when hired

Years E2:E82 Number of years the faculty member has been

employed at the college

Salary F2:F82 Current salary of faculty member