ABSTRACT
In multiple regression we are interested in predicting a dependent variable from a set of predictors. In a previous course in statistics, you probably studied simple regression, predicting a dependent variable from a single predictor. An example would be predicting college GPA from high school GPA. Because human behavior is complex and influenced by many factors, such single-predictor studies are necessarily limited in their predictive power. For example, in a college GPA study, we are able to improve prediction of college GPA by considering other predictors such as scores on standardized tests (verbal, quantitative), and some noncognitive variables, such as study habits and attitude toward education. That is, we look to other predictors (often test scores) that tap other aspects of criterion behavior.
