The Harvard faculty announced a major change in its advanced placement exam policy last week, deciding in a unanimous vote to no longer give credit or accelerated placement for anything less than a 5 on an advanced placement test. High school counselors and Ivy League applicants around the country are now waiting to see if other top colleges will follow suit.

As Yale administrators debate a potential shift in the University’s advanced placement policy, they should seriously consider the significant disparity Harvard’s faculty found between students scoring a 4 on the exam and those who earned a 5. But they should also avoid the temptation to paint important academic policy in overly broad strokes. Instead of jumping at Harvard’s move with a similar across-the-board change, Yale should urge a re-examination of advanced placement policy through its current, well-conceived framework of independent departmental decision.

In spirit, Harvard’s decision is the right one. In a recent study, Harvard College Dean Harry Lewis found that students who earned 4’s on the chemistry and microeconomics exams performed significantly worse in their first Harvard classes in the subjects than their classmates who earned 5’s.

This disparity should come as no great surprise. In fact, the simple reality may be that advanced placement exams do little to prepare students for college-level work at all. As anyone who has ever taken a high school advanced placement class knows, the teaching and material is often geared solely toward producing high exam scores, often at the cost of substantive learning. Treating advanced placement classes as “test prep” for the types of questions most likely to appear on the exam does little to impart the critical thinking skills necessary to succeed in a college class.

One could easily make the argument that top colleges should not grant acceleration or better placement for advanced placement test performance at all. The problem with this approach — and any other across-the-board move, like Harvard’s last week — is that it fails to account for the substantial difference between advanced placement subjects.

While advanced placement exams may be inappropriate for awarding credit or deciding placement in some — maybe even most — subjects, there are others in which even a grade of 4 may reflect a level of achievement sufficient for Yale’s purposes. Deciding which exams meet such standards is the job of the individual departments, which possess the vast knowledge and subject-specific judgment needed to evaluate the merits of advanced placement tests within their field.

This discretion is useless, though, unless departments take the time to rigorously research the relevant tests and examine the freshman-year performance of students who accelerate or gain higher placement through advanced placement scores.

While changes in some cases may not be necessary, everyone involved will benefit from the scrutiny. And when departments do not believe the tests meet Yale’s high academic standards, they should not be reticent to discontinue accepting advanced placement scores altogether.