Building a Regression Model as a Screening Tool
You are on the admissions board of a university, in charge of creating a screening process that will predict academic success (university grade point average (GPA)) for newly entering freshman. At your disposal is a variety of data from hundreds of current university students including:
- High School GPA
- SAT score (from high school)
- ACT score (from high school)
- Socio-economic status
- Freshman and sophomore university cumulative GPA
- Parents' education
- Intelligence Quotient
Describe how you would build a regression model that could be used as a screening tool, with mention of the high correlation (colinearity) between SAT and ACT, and possibly IQ.