Question

You build several models predicting job performance (sales dollars in thousands) as a function of years of industry experience. The parameter estimate is .50 (p < .0001) for a simple linear regression and .50 (p<.0001) when including three other predictors (education level, SAT score, and age) in a multiple linear regression. What does the .5 mean in the simple linear regression? What does the .5 mean in the multiple linear regression?

Answer 1

The parameter estimate of 0.50 in a simple linear regression is the slope of the linear relationship between the criterion variable and the predictor variable. So, each unit change in the industry experience changes the job performance by 0.50 in the model.

The parameter estimate of 0.50 in a multiple linear regression is the slope of the linear relationship between the criterion variable and the predictor variable,independent of all the other predictor variables, i.e., with all other predictor variables remaining constant. So, each unit change in the industry experience changes the job performance by 0.50 in the model, with all the other predictors (education level, SAT score, and age) remaining constant.