Validity and Reliability

1. Validity - Does the test measure what is says it measures?

1.1. New node

2. Reliability - A test is reliable if it yields basically the same results after repeated use.

3. 3 methods used to ensure reliability are: test-retest, alternative form, and internal consistency

4. Content Validity - Ensures that the test questions correspond to the users decisions of what should be covered by the test

5. Test-Retest- Test is given twice and both scores determine the correlation

6. Alternative Form - Uses two tests that are equivalent for an estimate

7. Internal Consistency - Items within the test should be correlated & internally consistent

8. Criterion-Related Validity Evidence - Two types of Criterion-Related validity evidence is the concurrent and the predictive

9. Concurrent - Determined by giving both new and established tests with a correlation between the two test scores

10. Predictive Validity Evidence - Tells us how well our test predicts the future behaviors of the test-takers

11. Construct Validity Evidence - This type of evidence is certain if the test's relationship to other information corresponds well with some theory