A Guide to Business Statistics

Chapter 12Multiple Regression

In Chapter 11, we used regression analysis to estimate the relationship between grades and attendance rates. The results suggested that attending more classes will lead to higher grades on average. However, attendance only explained a fraction of the differences observed in students' grades. There must be more to the story. There are of course other variables that may affect the performance, and if we can gather data on those variables, we should include them in our analysis. In this chapter, we analyze regression using multiple independent variables. A simple linear regression is really just a special case of multiple regression, and outside of undergraduate classes in statistics they are both just called regression.

In this chapter, we will dig deeper into the question of what affects students' course grades in my undergraduate statistics courses. Some of the additional variables we will include are more continuous in nature, including scholastic aptitude test (SAT) scores, hours spent studying, and a score on a logical thinking test. Other variables will be categorical (or qualitative) in nature, including the gender and year of study. We will discuss the differences in interpretations between continuous and binary data in regression. We will also consider how variables may interact with each other. The overarching objectives are the same from Chapter 11. The goal is to estimate the relationship between grades, and the independent variables then ...