GMAT Strategy

Does anyone know about how accurate PR practice test scores are. :

Hi,

I am a teacher and tutor with The Princeton Review. There have been a few questions and posts about TPR practice tests, so I thought I'd respond here to share my knowledge about them.

Our practice tests are based on extensive research into the test, including how the computer adaptive algorithm works. I took the GMAT recently and found that our tests are just a bit harder than the real thing, probably because we try to show a particularly broad variety of problem types.

Indeed, my scores reflect my impression. On TPR tests, my average verbal score is 41, my average math score is 45, and my avcerage composite score is 700. My real GMAT scores from April 2008 are 47V, 49M, 770. Granted I was more focused during the real thing, as opposed to taking the tests at home, but I do believe our tests are a bit harder but well within an acceptable range.

I have read that some say our tests are too easy or in some way not reflective. This is definitely not true. Quite the contrary: I have heard that some of the other test prep companies do not even have a computer adaptive algorithm, although I do not know for a fact.

Of course, I agree that any preparation must include the GMATPrep software as well.

Cheers,

David Stoll
The Princeton Revew

I took Princeton Review's tests and felt they were a very good indicator of the real thing.

The PR overall score tends to be lower than actual based on my experience. I also see the majority of posters who share their scores tend to have lower PR score than actual. This does not appear to be an isolated incident.

The PR quant is not representative of the actual exam (PR is much easier). In the real exam, there will be a lot more questions testing the student's knowledge at the application and correlation levels than what the PR test does.

If pressed, I could find specific examples from a PR sample test versus a GMATPrep sample test to support my hypothesis. I wonder if we should start doing that?

For Example, here are some sample questions I was given on my PR test (scored 50 on the quant on this test). These questions came from the middle of the exam and the preceding 5-6+ questions were correct (so this should be considered a 700+ questions?)

On a certain test, Bill scored 20 more points than John, but half as many points as Sue. If the sum of the scores of the three students was 180 points, how many points did Bill receive

How tall is Maggie?

(1) Last year, Maggie was 5 feet, 10 inches tall.

(2) If Maggie were 2 inches taller, she would be six feet tall.

In my experience, these are not representative of 700+ questions. On the real GMAT, you would be lucky to get 1-2 questions this easy if you expect to score a 50+ on the quant..

just my two cents...

I took PR tests and I agree with you that people generally do better on the actual test - which is a good thing!

I know PR penalize not completing the test far more than the actual GMAT.

Also, since PR isn't using the real CAT algorithms, the results are more linear. It means if you do poorly in the end, you will penalize more than what you would in the actual test.

Cheers,

Reza

How tall is Maggie?

(1) Last year, Maggie was 5 feet, 10 inches tall.

(2) If Maggie were 2 inches taller, she would be six feet tall.

Maggie is hot! And I don't have time for her right now because my test is coming up. Sorry Maggie, the answer is B but I am not interested in you right now