Raising AP enrollment can lower average exam scores without lowering teaching quality.

Schools often track Advanced Placement, or AP, using different numbers: who takes the class, who sits for the test, and who passes. These steps come from separate administrative systems and measure very different things. To understand this pipeline, researchers tracked 276 New Jersey public high schools over several school years. They compared course enrollment, test taking, and test scores. When a school increased formal AP class enrollment by 1 percentage point, its overall successful attainmentsuccessful attainmentthe overall share of students who end up passing an AP exam—meaning students passing exams—rose by 0.34 percentage points. Higher class enrollment was also strongly tied to more students taking exams. However, it was negatively linked to conditional attainmentconditional attainmentamong students who take the exam, the share who pass, which is the pass rate among test takers. Offering more AP subjects and higher enrollment intensity showed this same pattern. As access expands, pass rates among test takers dip, likely because the mix of participating students changes, not because teaching got worse.

Why it matters

Schools and policymakers often treat AP participation and pass rates as measures of success. This study shows why those numbers can move in opposite directions: when more students take AP courses, the average exam score may fall simply because the group of test-takers is larger and more diverse. That means a school can be doing a better job of expanding opportunity while its published scores go down.

Questions answered