VALIDITY

Validity is the degree to which a test measures what it purports to measure.

Types of validity include:

1. Content validity: the purpose is to see if the test items match a set of goals and objectives. Content validity is established by demonstrating a similarity or relationship between what is being measured on the test and what is being taught in a targeted program. This is achieved by comparing tests with school courses or objectives and by using experts in content areas. Types of tests are criterion referenced tests, skill tests, licensure tests, aptitude tests, and survey achievement tests.

2. Criterion validity: this is when test scores are used to draw conclusions about the person's performance on another measure. There are two types of criterion validity: concurrent and predictive.

a. Concurrent validity: this is when the relationship between the test scores and the criterion scores is established at the same time using ratings, observation, or another test. Types of tests include aptitude tests, ability tests, personality tests, and employment tests.

b. Predictive validity is used to determine if there is a relationship between test scores and the performance in that area in the future (i.e. the extent the test scores are predictive of actual performance). This is done by correlating test scores with the criterion measure after a period of time. Types of tests include scholastic aptitude tests, general aptitude tests, readiness tests, personality tests, and intelligence tests.

3. Construct validity: this determines the test's ability to measure a hypothetical, theoretical construct, such as anxiety, creativity, intelligence, etc. This is done by conducting statistical analysis such as factor analysis or multivariate analysis of variance. Types of tests include intelligence tests, aptitude tests, and personality tests.

When a test has low reliability, its validity is also low. A highly reliable test may or may not be valid.