Classical test theory

Classical test theory (CTT) is a way of understanding how tests measure something you are interested in. CTT basically says that when you give someone a test, their score will depend on two things:

1. Their true level of knowledge or skill in the topic the test is measuring, and
2. Random error.

The true level of knowledge or skill is like how smart a person really is, or how skilled at some task they really are. Random error is how much their test score can be affected by things other than how smart they really are, like guessing, being tired or distracted, feeling anxious, or having a bad day.

So CTT's goal is to figure out how much of someone's test score should be explained by their true skill level, and how much should be explained by random error. CTT uses special formulas to figure this out. Once you understand how much of the score is because of the person's true skills, and how much is because of random things, you can figure out how good a test is for measuring what it is supposed to measure.