Suppose we want to test some property of a particular construct. Then we must design a test or a metric which measures the property. For example to study the intelligence of a person we may devise an IQ test. Here the person is the ‘construct’ and ‘intelligence‘ is the property we want to study. But here we need to make sure that our testing method is both reliable and valid.
Types of Validity
A test is said to be valid if the performances measured by accurate and comparable to those measurements obtained independently by experts. Some types of test validity are:
- Predictive Validity- A test has high predictive validity if it is able to predict future efficiency and performance of a candidate. This is used when selecting people for jobs or training.
- Concurrent validity- This type of test is used to obtain information about the candidate’s current status.
- Content Validity- In this article we will focus on content validity.
This type of test measures how far the test covers all aspects of the phenomena which is under study and whether it is actually measuring the property under study. Note that content validity is subjective. There may be a difference of opinion among experts as to whether a test has content validity or not. But as a general thumb rule if many experts agree then a test can be said to have content validity.
Some examples of tests which have content validity are:
- A comprehensive calculus test that measures the relevant skills and calculus related knowledge of a student.
- An IQ test as a preliminary measure of certain kinds of intelligence.
Some examples of tests which do not have content validity are:
- A spelling test as a measure of intelligence.
- Merely counting the number of push-ups a person can do as a overall measure of health and fitness.