Validity and Reliability
by Keri Lamb
1. A test is considered valid if it can prove that it measures what it implies it will measure (Kubiszyn & Borich, 2010).
2. There are three different types of validity (Kubiszyn & Borich, 2010).
2.1. Content validity evidence is used to determine if the test items match what the teacher wants them to cover on the test (Kubiszyn & Borich, 2010).
2.2. Criterion-related validity evidence is when the scores of tests are compared to another outside standard. There are two types of criterion-related validity evidence (Kubiszyn & Borich, 2010)
2.2.1. Concurrent criterion-related validity experience is when a new test and a valid, and already established test are given to students at the same time, so that their date can be compared (Kubiszyn & Borich, 2010).
2.2.2. Predictive validity evidence is used to determine how well a student will do on future tests (Kubiszyn & Borich, 2010)
2.3. Construct validity evidence is when a test is considered valid because it relates to information that matches a well-known theory (Kubiszyn & Borich, 2010).
3. Validity and reliability are important in learning and assessment because they allow the teacher to be sure that her test items are appropriate; a student cannot show that he has knowledge of the material if the test items are not valid and reliable (Kubiszyn & Borich, 2010). When a teacher is going to use the information gathered from tests, she needs to be sure that it is accurate; an unreliable test will not have accurate test scores, and those scores could be the difference between a student passing a class or failing a class.
4. Reference: Kubiszyn, T. & Borich, G. (2010). Educational testing & measurement: Classroom application and practice (9th ed.). Hoboken, NJ: John Wiley & Sons, Inc.
5. Keri Lamb EDU 645
6. A test is considered reliable if the same person takes the test more than once and has consistent results (Kubiszyn & Borich, 2010).
7. There are three methods of determining the reliability of a test (Kubiszyn & Borich, 2010).
7.1. Test-retake is used to give a student a test two times to determine if the answers between the first test and the second test are the same (Kubiszyn & Borich, 2010).
7.2. Split-half methods are used to give students two similar parts of a test, but during different times to determine their grade (Kubiszyn & Borich, 2010).
7.3. Kuder-Richardson methods evaluate the internal consistency of tests and are used to determine if the test items are similar to other tests items (Kubiszyn & Borich, 2010).