When the issue of grading tests comes up, there are two general approaches: norm reference and criterion reference. In this article, I hope to shed some light on the differences between norm referenced tests and criterion referenced tests.
What is a norm referenced test?
Norm referenced tests are tests that the compare test takers with each other. Many times you see results of norm referenced tests as a percentile. For example, you may get test results back that put you in the 90% percentile meaning that you score 90% higher than other test-takers.
Norm referenced tests are usually used when the purpose of the test is to select the best test-takers. Norm referenced tests are great for situations where there are more applicants than positions available. Administering a norm referenced test lets the administrator see how participants stack up against each other.
The SAT is a norm referenced test.
What is a criterion referenced test?
A criterion referenced test is a test that measures test takers against a specific set of criteria. Many licensing exams are criterion referenced. That is, as long as the test taker scores above a certain grade, a 70% for example, the test taker can pass the test.
Norm referenced and criterion referenced tests compared.
Norm referenced assessments are usually used when the administrator needs applications to compete against each other. A norm referenced exam would be a good choice if the exam was to see who would receive a scholarship, for example.
A criterion referenced exam would be ideal if the administrator needs test takers to compete against a certain set of requirements. An exam for an insurance license would be a criterion based assessment. It doesn’t matter if any one specific insurance agent is better than another; what matters is that the insurance agent has the correct skills to do his or her job correctly.
When designing an assessment it’s important that one of the first decisions you make is if you want your exam to be norm referenced or criterion referenced.
What to read next:
If you are interested in learning how to design assessments and tests, you should learn about the three things that every assessment should have: practicality, reliability, and validity.
