Veritas Vs Princeton Review

Question

Which one of the two is more accurate when compared to the GMAT ? I seem to be getting a lot of variation in my scores and I don't know where I stand. Who has a better algorithm ?

My scores were as follows:

Princeton Review Test 1: 600 (Q 38/ V 35)
Princeton Review Test 2: 500 (Q 35/ V 25)
Princeton Review Test 3: 550 (Q 37/ V 30)
Veritas Prep Test 1: 630 (Q 43/ V 33)
Princeton Review Test 4: 540 (Q 35/ V 30)

Are Test 1's from each company easier ? I seem to be doing better in them. The impression I get is that PR does not improve their difficulty level unless you get a good number of answers correct in a row. I've had 8 questions correct in a row with no increase in difficulty (They show difficulty level). I felt Veritas is slightly more adaptive, but I may be wrong. Any one out there with good input ?

Note: I know that the number of correct questions don't count, but I seem to be getting only 8-10 incorrect answers on PR, with a score of 540. While in Veritas I got ~12-15 wrong and still scored higher. So I don't understand how their algorithms can be significantly different.

KarishmaB · Answer

Here is a bit about our tests:

Veritas Prep designed its practice tests to simulate the official GMAT as closely as possible, both in terms of question quality and how the computer-adaptive algorithm determines your ability level and chooses which questions you see. Our GMAT practice tests draw from a pool of hundreds of realistic GMAT questions, each of which is measured on multiple attributes that go far deeper than simply “easy / medium / hard.” Given the adaptive nature of our practice tests, no two test instances will be exactly the same.

Veritas Prep’s practice tests are built on a platform that employs Item Response Theory (IRT), the same system underlying the real GMAT. Unlike more basic approaches that simply count correct and incorrect responses and then generate a score, IRT recognizes that no two GMAT questions are exactly the same. What question you receive at any given point on a test will be dictated by how well you have performed up to that point, and how you answer that question will impact your ability estimate and help determine which question you see next. The system also takes “content balancing” into account, ensuring that you see a mix of question types that are representative of what you will see on the real GMAT, all while gathering as much information about your ability level as possible.

Every one of our GMAT questions must be validated across thousands of responses from hundreds of students before it is added to a Veritas Prep GMAT practice exam. Veritas Prep has collected more than one million responses from tens of thousands of GMAT students, via both the GMAT Question Bank and by using experimental questions in our computer-adaptive tests (just as the real GMAT collects data). In fact, Veritas Prep collects more than 50,000 new responses each week! Each item is not just validated once, but is constantly recalibrated based on new data. Veritas Prep also regularly measures its practice tests’ scores against the Official GMAT by collecting student feedback, ensuring that its practice exam scores are very close to those generated by the real test.

EMPOWERgmatRichC · Answer

Hi Mirj626,

The scoring algorithm on the Official GMAT is far more complicated than most people realize. Since that algorithm is proprietary, no GMAT company has an exact match for it, thus CAT scores can vary a bit based on the 'biases' involved in their respective designs. Since you're skeptical of these 3rd-party CAT scores, have you taken any of the GMAC CATs? How did you score on THOSE CATs (including the Quant and Verbal Scaled Scores)?

GMAT assassins aren't born, they're made,
Rich

Veritas Vs Princeton Review

Prep Toolkit

Top 5 GMAT Debrief Videos

615 to 715 in 15 Days (Ria's Strategies)

More than Quant/Verbal Abilities: Speed vs. Accuracy

745 with Only Self-Prep and GMAT Club

From 555 to 765: 210-point Improvement

Perfect 805 Score Debrief by Julia

My Rewards