Validity and Reliability

1. Kuder Richardson method

1.1. Determines if items in one half of the test have as much in common with one another as they do with items in similiar form

2. Split half method

2.1. The test is split into two equal halfs to determine the correlation between them

3. Test Retest Reliability

3.1. The test is administered twice

3.2. The scores from the first and second test are compared to see if there is any difference in scoring

3.3. The problem with this method is administering the two test to close together because students can rely on memory.

4. Alternate Forms

4.1. Two forms of a test are given to determine the reliability of the scores

4.2. This method eliminates the problem of memory that the test retest method can cause

4.3. The problem with this method is being able to produce two good tests

5. Internal Consistency

5.1. The test is designed to measure one concept and if one item is answered correctly then the other items should be answered correctly also.

6. Predictive validity

6.1. How well the test will predict future behavior.

6.2. Works well for aptitude and personality tests

7. Concurrent criterion related validity

7.1. Two well known established tests are compared with the new test

8. Content Validity

8.1. Test questions match learning objectives

8.2. Measures if the test is in line with acadamic standards

9. Criterion Related Validity

9.1. Scores from a test are matched with an outside test which is done in two ways.

10. Construct Validity

10.1. Resutls from a test should match expectations