From the linked Associated Press article:

Rather than award credit for an introductory course to incoming students who got the highest score on the AP test, the department gave those students a condensed version of the Dartmouth course's final exam. Ninety percent failed, Tell said. And when those students went on to take the introductory class, they performed no better than those who did not have the high AP test scores.

http://www.npr.org/templates/story/story.php?storyId=1696110...

You know what I'd like to see - is a second study, in which students who got the highest scores on the Dartmouth course, with the same interval of delay also take that same final exam.

I was surprised at the number of fellow students at my University who openly admitted that within a month or two of the final exam, they couldn't recall much of anything about some of the courses in which they had scored an A or higher in.

It's entirely possible that the issue isn't the AP courses, but the significant delay taking (and doing well in the course) - and retesting again.

Alternatively - take a student from another Ivy League, Yale, Harvard - and see how they would do on a Dartmouth Final exam with some delay between taking the course and testing out again.

Indeed. A lot of education in the western world is remarkably shallow. Blitz through a lot of "material" and busy work, then cram so you can pass the test the next day, then forget everything. It's a horrible way to "learn" anything.

I found the issue was just building on my education. You could teach me concept X until you were blue in the face, and I wouldn't remember a whole lot a few months later. But give me a half-decent education on concept X and then soon after move to something that builds on concept X, and I will retain a far greater understanding of concept X for far longer.

I think it is related to a similar effect I have noticed- when I teach something, I again retain that information much better.

In defense of AP tests, they're meant to replace college level courses ... not university level courses. They're equivalent to what you'd encounter at a local community college, not what you'd encounter at the top 100 research universities in the country.

That's not what "college" means in the United States. "College" is a synonym with "University." There is sometimes some word usage where the connotations can be different, but rarely, and in particular, there's no difference in the notion of "university level courses" and "college level courses."

This is doubly confusing in the case of Dartmouth College, which is actually a university (I got my physics Ph.D. there - and they have a medical school, a prominent MBA school, etc.) but is called a "College" for interesting historical reasons involving Daniel Webster.

Generally the difference is that a college will offer undergraduate programs and universities offer both undergraduate and graduate degrees.

Interesting. I wonder how much the trend of "teaching to the test" has infected the AP program.

Anecdotal evidence, I tested out of a bunch of courses with AP credits. The only ones I actually went directly to the next level were Economics. I scored 5's on micro and macro and skipped the 100 level of both. I did the intermediate and it went just fine (I got A's). Could I pass the final exam for the previous level with a 90% right when I got there? I am really not sure. Presumably I would have been studying and preparing for that weeks in advance and the content would be fresh in my mind (not after a summer of not studying). I don't feel like I missed anything and saved a lot of time (and money).

Dartmouth has a very strong teaching tradition. Certainly not all the professors are good at it. But freshman etc. courses aren't taught by teaching assistants.

I was a physics graduate student at Dartmouth. The undergraduate lectures were given by professors, but most of the homework and exam grading, and all the lab instruction, was by us, and doing this was a requirement for the Ph.D. I discovered in teaching my first lab that 1/3 of the undergraduates at this Ivy League school, mostly premeds in this case, did not know how to add fractions.

The main difference between AP tests and college courses is not the material that is covered (roughly comparable), but rather the format in which the knowledge of the material is evaluated.

Like most standardized tests, the AP tests are incredibly game-able. As a student, I began taking random AP tests without taking the class and just studying a guide the week before the test. I passed every test (most with a 5), but that was because I had learned to become good at the test, rather than learning the material behind it. (There are some exceptions, like art portfolios)

But myself and the above commenter (InclinedPlane) are the exception. There is, frankly, no better way for a student to demonstrate prior knowledge of a subject than through a standardized test.

Dartmouth, and all colleges, are welcome to set their policies to reflect their trust in the system, but there should be a way, whether internal or standardized, for a student to be placed based on prior knowledge.

"just studying ... the week before the test. I passed every test"

I'm not sure this supports your conclusion. My guess is that many of us could do the same for the final exams of the classes those AP tests translate to credit for. In fact I'd guess most of us either did this or know of many people who did the same for even more advanced classes.