Validity and Reliability

1. Predictive VE - this is a predicative test to determine future achievement on tests and/or success in an area. Both concurrent and predictive CRVE produce a numerical index of validity (Kubiszyn & Borich, 2013, p. 328). In the classroom, teachers may utilize this evidence to predict future growth and progress.

2. Reference

2.1. Kubiszyn, T. & Borich, G. (2013). Educational testing & measurement: Classroom application and practice (10th ed.). Hoboken, NJ: John Wiley & Sons, Inc.

3. Test-Retest

3.1. This type of reliability allows teachers to test students twice on the same materials, which allows them to compare scores. If there is a drastic difference between scores, then instruction may need to be re-evaluated. It is best to wait a considerable amount of time between retests so that accurate comprehension and information retention is demonstrated, rather than remembering what was on the first test (Kubiszyn & Borich, 2013, p. 339).

4. Alternative Form

4.1. Alternate form allows teachers to administer two similar, equivalent tests so that scores may be compared and the reliability of scores may be evaluated. Conditions must remain similar for both test administrations, but eliminates the possibility of students remembering questions from the test-retest strategy. (Kubiszyn & Borich, 2013).

5. Internal Consistency

5.1. Internal consistency methods of reliability include split-half and Kuder-Richardson, both involving the strategy of taking similar test items and correlating responses to see if students got one question right that they would also respond correctly to similar test questions. These methods alleviate memory and practice efforts, but can only be used if a test includes "similar items measuring a single concept" (Kubiszyn & Borich, 2013, p. 342).

6. Reliability

7. Construct

7.1. Construct Validity Evidence simply measures if test results correspond to a theory or what a teacher expects from what has been taught. Ex. a math test would correlate to math instruction, etc. (Kubiszyn & Borich, 2013, p. 329).

8. Criterion-Related

8.1. Criterion-related validity evidence (CRVE) deals with relating scores from a test to an external criterion. Within this type of validity, there are two different types:

8.1.1. Concurrent CRVE - this evidence basically administers two tests at the same time in order to check for correlation which is called a validity coefficient between the two tests. This type of validity evidence allows the teacher to compare student test results with an already well-established test (Kubiszyn & Borich, 2013).

9. Content

9.1. Content Validity Evidence is basically a "logical judgment", when a test is inspected to assess that specific content is demonstrated on a test. This type of validity evidence does not check for aspects such as item construction or reading level appropriateness of the test. At a minimum, the test measures instructional objectives (Kubiszyn & Borich, 2013, p. 327).

10. Validity Evidence