Validity: What is the truth?

Validity is the property of an assessment tool that indicates whether the tool measures what it is intended for. In other words, validity refers to how well a test is capable of meeting its intended purposes. For example, if an intelligence test is supposed to measure a certain kind of intelligence as it is defined by the test’s creators, then it should do just that.

The validation process begins with a complete and clear description of what is to be measured (the construct), the purpose of the test, the claims to be made about test takers, the intended interpretations of the scores or other results, and the population for which the test is designed.

Types of validity

Just as there are different types of reliability, so there are different types of validity. We will present three of the most frequently used categories of validity in Table 2 (attached).

WHAT SHOULD YOU DO WHEN VALIDITY IS UNSATISFACTORY?

Generally speaking, if you do not get the validity you want, it is because the measurement instrument is not accomplishing what it should.

If you designed an achievement test and you seek a satisfactory level of content validity, you need to rewrite the questions of the test to make them more consistent with the opinions of the experts you consulted.

In terms of criterion validity, you would need to re-examine the items of the test in order to answer the following question: How well do responses to the questions relate to the criterion? This procedure implies that the criterion you are using (college success, critical thinking ability, etc.) make sense.

In terms of construct validity, you should take a deeper look at the theoretical arguments that underlies the test you developed. Perhaps the definition and conceptualization of the investigated concept need some critical rethinking. Your task is to find consistency between the theory and the test items based on the theory.