Kaplan Premier CAT tests vs real ones
This topic has expert replies
-
- Newbie | Next Rank: 10 Posts
- Posts: 3
- Joined: Wed Oct 29, 2014 11:43 am
I got 3 scores in a row of 50 on the quant section for the Kaplan tests and i'm concerned they are inflated with how Kaplan calculates the number. I also did the Princeton tests, untimed, and I was in the mid 40s. I know this subject has been beaten to death but thats a pretty big percentile swing of around 20%. Any one take multiple Kaplan tests and score way lower?
GMAT/MBA Expert
- [email protected]
- Elite Legendary Member
- Posts: 10392
- Joined: Sun Jun 23, 2013 6:38 pm
- Location: Palo Alto, CA
- Thanked: 2867 times
- Followed by:511 members
- GMAT Score:800
Hi mikep741852,
Based on what you've described, there are a number of obvious factors that would affect your score (as well as a number of other factors that you might not have considered):
1) Different CATs from different companies will have their own "built-in biases", so comparing one Scaled Score to another can be problematic.
2) Taking a CAT UNTIMED is NOT realistic, so I would find any score earned under those conditions to be questionable.
3) The big benefit of taking a CAT is that it allows you to measure your OVERALL PERFORMANCE. Since you admit to have taken sections untimed, I have to ask if you took the ENTIRE CAT? Skipping the Essay and IR sections is NOT realistic and you would likely score better in the Quant section if you did not complete the first hour of the Test.
4) Percentile data can be dubious because the pool of Test Takers who take any given exam will likely be different.
If you're looking for the MOST realistic assessment of your abilities (beyond taking the actual GMAT), then you should take a FULL GMAC CAT under realistic conditions.
GMAT assassins aren't born, they're made,
Rich
Based on what you've described, there are a number of obvious factors that would affect your score (as well as a number of other factors that you might not have considered):
1) Different CATs from different companies will have their own "built-in biases", so comparing one Scaled Score to another can be problematic.
2) Taking a CAT UNTIMED is NOT realistic, so I would find any score earned under those conditions to be questionable.
3) The big benefit of taking a CAT is that it allows you to measure your OVERALL PERFORMANCE. Since you admit to have taken sections untimed, I have to ask if you took the ENTIRE CAT? Skipping the Essay and IR sections is NOT realistic and you would likely score better in the Quant section if you did not complete the first hour of the Test.
4) Percentile data can be dubious because the pool of Test Takers who take any given exam will likely be different.
If you're looking for the MOST realistic assessment of your abilities (beyond taking the actual GMAT), then you should take a FULL GMAC CAT under realistic conditions.
GMAT assassins aren't born, they're made,
Rich
-
- Newbie | Next Rank: 10 Posts
- Posts: 3
- Joined: Wed Oct 29, 2014 11:43 am
Rich, I agree with the timed vs untimed, but I didnt expect to score higher on timed vs untimed, like I did. Seems like the Kaplan one calculates a higher score then what I would have expected to see if it was a princeton test....by about 4-5 points id say. I also find it hard to believe that I got a 50, with getting 12 wrong....
GMAT/MBA Expert
- [email protected]
- Elite Legendary Member
- Posts: 10392
- Joined: Sun Jun 23, 2013 6:38 pm
- Location: Palo Alto, CA
- Thanked: 2867 times
- Followed by:511 members
- GMAT Score:800
Hi mikep741852,
You've brought up a number of important issues, but some aspects of what you assume about the situation are not correct. You seem to think that all CATs are "created equal" and that the Princeton results are more realistic (But why? Because you scored lower?). If you have not yet done so, you should do a full question-by-question review of each of your CATs. Ask yourself how many times you got a question wrong because of a silly/little mistake and how often you got a question wrong because it was too hard. How many times did you "guess" and get lucky? A score drop can usually be traced back to too many silly mistakes.
You also seem focused on the number of correct answers, but that is NOT the only factor in how a CAT (or the GMAT) calculates your score. There are a number of factors that go into calculating a score, and on Test Day, there are Experimental questions that do NOT even count.
I mentioned earlier that taking an Official GMAC CAT would provide you with arguably the most accurate information, so you should plan to take one. Make sure to take the ENTIRE CAT though, otherwise you'll end up with data that won't tell you anything about your pacing, fatigue problems or endurance issues.
GMAT assassins aren't born, they're made,
Rich
You've brought up a number of important issues, but some aspects of what you assume about the situation are not correct. You seem to think that all CATs are "created equal" and that the Princeton results are more realistic (But why? Because you scored lower?). If you have not yet done so, you should do a full question-by-question review of each of your CATs. Ask yourself how many times you got a question wrong because of a silly/little mistake and how often you got a question wrong because it was too hard. How many times did you "guess" and get lucky? A score drop can usually be traced back to too many silly mistakes.
You also seem focused on the number of correct answers, but that is NOT the only factor in how a CAT (or the GMAT) calculates your score. There are a number of factors that go into calculating a score, and on Test Day, there are Experimental questions that do NOT even count.
I mentioned earlier that taking an Official GMAC CAT would provide you with arguably the most accurate information, so you should plan to take one. Make sure to take the ENTIRE CAT though, otherwise you'll end up with data that won't tell you anything about your pacing, fatigue problems or endurance issues.
GMAT assassins aren't born, they're made,
Rich
-
- Newbie | Next Rank: 10 Posts
- Posts: 3
- Joined: Wed Oct 29, 2014 11:43 am
Valid points....Also, could be that it just seemed easier to me because I have done more studying- as far as why I think one may be more realistic.
- dabral
- Master | Next Rank: 500 Posts
- Posts: 134
- Joined: Fri Jan 23, 2009 1:23 pm
- Location: California
- Thanked: 59 times
- Followed by:15 members
@mikep741852
I will second Rich's advice on taking a full length GMATPrep or Exam Pack 1 tests to really find where you currently stand. One cannot rely on any of the third party CATs for realistic prediction of GMAT scores. There are a whole of host reasons for that, one of the main ones being the underlying statistics for each individual question that the official GMAT tests contain. Anyway, try to take the test under conditions that are close to the real exam, say in a library or after hours at office.
You will have the answer to your question.
Cheers,
Dabral
I will second Rich's advice on taking a full length GMATPrep or Exam Pack 1 tests to really find where you currently stand. One cannot rely on any of the third party CATs for realistic prediction of GMAT scores. There are a whole of host reasons for that, one of the main ones being the underlying statistics for each individual question that the official GMAT tests contain. Anyway, try to take the test under conditions that are close to the real exam, say in a library or after hours at office.
You will have the answer to your question.
Cheers,
Dabral
Free Video Explanations: 2021 GMAT OFFICIAL GUIDE.
- David@VeritasPrep
- GMAT Instructor
- Posts: 2193
- Joined: Mon Feb 22, 2010 6:30 pm
- Location: Vermont and Boston, MA
- Thanked: 1186 times
- Followed by:512 members
- GMAT Score:770
It is certainly true that not all practice Tests are created equal. And it is certainly true that if I was only going to take 4 practice tests those would be the two free GMATPrep Tests and the Two GMATPrep Exam Pack 1 tests for purchase.
However, it is also true that not all third-party tests are created equal, either. As stated above:
So by all means we encourage everyone to use the official tests! But if you need more than 4 tests you might want to try the free Veritas exam! https://www.veritasprep.com/gmat/free-gm ... tice-test/
However, it is also true that not all third-party tests are created equal, either. As stated above:
The Veritas Prep practice tests do have statistics for each question. The Veritas exams are based on Item Response Theory as are the official GMAT exams. Each item on the Veritas exams is rated not according to someone's idea of the difficulty of the question but based on actual student responses. At this point there are over 8.7 million responses to the Veritas tests and question bank!One cannot rely on any of the third party CATs for realistic prediction of GMAT scores. There are a whole of host reasons for that, one of the main ones being the underlying statistics for each individual question that the official GMAT tests contain.
So by all means we encourage everyone to use the official tests! But if you need more than 4 tests you might want to try the free Veritas exam! https://www.veritasprep.com/gmat/free-gm ... tice-test/