Test Evaluation

1. Reliability

1.1. Test-Retest

1.1.1. Give test twice

1.1.2. Correlation should occur

1.2. Alternative Form

1.2.1. Give two equal tests to same group

1.2.2. Then correlate scores

1.3. Internal Consistency

1.3.1. Split half methods

1.3.1.1. AKA odd-even

1.3.1.2. Divide test into 2 equal halves then correlate scores

1.3.2. Kuder-Richardson Method

1.3.2.1. Items from one test have as much in common with items from another test

1.4. Importance

1.4.1. Does the test yield the same/similar score rankings consistently

2. Validity

2.1. Content

2.1.1. Simplest Evidence

2.1.2. Inspect Questions

2.2. Concurrent Criteron Related

2.2.1. Good measurement to compare new test

2.3. Predictive

2.3.1. Predicts future behavior of examinees

2.4. Construct

2.4.1. Relationship to other information corresponds with a theory

2.5. Importance

2.5.1. Ensures the test measures what its suppose to measure