Design of Experiments > Predictive Validity

## What is Predictive Validity?

A research study is useless unless it has some kind of predictive value. Predictive validity tells you how well a certain measure can predict future behavior. One of the most common uses for predictive validity is in University Admissions. Grade Point Average, SAT/ACT scores and other criterion are used to predict a student’s likely success in higher education. To test this theory, university students have been studied by their thousands and it’s been confirmed by hundreds of studies that there is a correlation between GPA/ACT/SAT and educational success. Therefore, the predictive validity is high for these measurements.

Other Examples:

- A
**depression outcome scale**is a tool to predict future behaviors of depressed people, such as social isolation, inability to hold down a job or maintain fulfilling relationships. Depressed patients can be followed to see if there is a correlation between their scores on a particular scale and their future behaviors. **Ability in mathematics**can predict future success in the sciences. This can be measured by giving tests to established scientists to see how they perform on math tests and whether there is a correlation between test scores and how well they do in their profession.**Pre-employment tests**can predict whether a person is likely to be successful in that career path. The pre-employment tests have often been tested for predicitve validity by following employees after employment to see if a correlation exists between test scores and career success.

Predictive validity is a subset of criterion validity. Criterion validity is an umbrella term for measures of how variables can predict outcomes based on information from other variables. To measure the criterion validity of a test, the test is sometimes calibrated against a known standard. In other cases, the test is measured against itself.

If you prefer an online interactive environment to learn R and statistics, this *free R Tutorial by Datacamp* is a great way to get started. If you're are somewhat comfortable with R and are interested in going deeper into Statistics, try *this Statistics with R track*.

*Facebook page*and I'll do my best to help!

i need the book

Nice