ABSTRACT
One of the goals of scientific research is to try to explain and predict phenomena. For example, if you are a teacher you know that some students in your classes excel, while others just can’t seem to grasp the concepts you are teaching. In order to explain these differences, you might look for explanatory variables such as IQ, motivation, amount of time spent on homework, reading speed, socio-economic status, or any other number of variables. Regression is a technique that allows you to look at a number of explanatory variables and decide which ones have independent power to explain what’s going on with the variable you have measured, for example scores on a TOEFL test. We’ll call this property you measure the response variable, and the variables you think might explain what’s going on the explanatory variables. The explanatory variables are also sometimes called independent variables, but we will avoid that since these variables are often correlated with the response variable, and thus not very “independent.”
