Classical test theory is a foundational framework in psychometrics holding that any observed test score is made up of a person's true score plus random measurement error.
In equation form, it's simply: observed score = true score + error. If someone scores 72 out of 100 on a conscientiousness measure, classical test theory says that 72 reflects their actual standing on the trait, plus some amount of noise — from a poorly worded item, a distracted moment while answering, or simple chance. The theory doesn't try to eliminate that error; it tries to estimate how much of it there is, which is the basis for reliability.
Classical test theory treats every item on a test as roughly interchangeable and focuses on the test as a whole. Its main limitation is that an item's difficulty and a person's ability get tangled together — the same test can look easier or harder depending on who happens to take it.
This limitation is what item response theory was later developed to address, modeling individual items rather than total scores. Classical test theory is older and simpler, but it's still widely used because it's easy to compute and works well for many practical purposes.