Big Difference between GMAT Prep1 & Prep2 scores

[Raj]

Hello,

I took my GMAT Prep1 on Aug 10, score 760 (V42, Q50).

I took my GMAT Prep2 on Aug 16, score 680 (v35, Q48).

Just for frame of reference my MGMAT scores started from 710 and ended up at 740 in the 4th test.

I was really unhappy with the 680 after the 760, so I am trying to find out what has happened and would like your input and see if any of you have had such an experience.

The only thing I have noticed is that in the 680-test, I missed quite a few questions in the first 10 questions in both Verbal and Quant and I am now forced to believe that the first questions do set the tone for the rest of the exam. I understand one shouldn't be spending inordinate amount of time on the frist 1/3 of the questions, but after this experience, I am going to spend more time (probably 2.5 mins each) on the first 1/3 of the questions.

I am sort of worried since in all the posts I have seen, the range of most people's GMAT Prep score is not greater than 20 points.

Thanks for any input,
-Raj.

[Guest]

Raj,

So if this is your first score below 700, I think you should not worry about too much. And also remember that once your verbal score falls below 40, you need a 49+ Quant score to bring your overall score up to 700+ again.

IMO you just did not concentrate as you have been in your previous trials and scored a little bit off your usual range. I would re-take the exam again.

Good luck,

EE

[Raj]

Thanks EE. Appreciate that. I will be resetting the two GMAT Preps so I can practice 2 more exams.

Thanks,
-Raj.

Raj,

So if this is your first score below 700, I think you should not worry about too much. And also remember that once your verbal score falls below 40, you need a 49+ Quant score to bring your overall score up to 700+ again.

IMO you just did not concentrate as you have been in your previous trials and scored a little bit off your usual range. I would re-take the exam again.

Good luck,

EE

[StaceyKoprince]

It looks like your quant score is what did the damage - the verbal dropped a little bit but the quant really dropped. Is that where you also noticed more early questions wrong? Did you notice anything else, like a string of questions wrong towards the end? Did you run out of time and have to rush at any point? Etc.

The early questions can "set the tone" as you say - but only if you don't then run out of time at the end because you spent too much time at the beginning. The penalty for getting a string of questions wrong in a row at the end is severe - and it's even worse if you actually run out of time and leave some blank. So I wouldn't revamp your entire pacing strategy based on the results of this one test - you could end up hurting yourself even more.

Generally, if you spend a little more time on a few questions (2-3), that won't hurt you too much. "A little more time" is defined as 2.5 to 3 min. BUT if you do it a lot and run out of time at the end, here's what can happen:

- on any questions you leave blank at the end, there's an automatic 3 percentile point deduction per question
- if you get 5 questions wrong in a row at the end (without any right answers to break up the string), the penalty ranges from about 2 to 2.5 percentile points per question

* note: for the latter, the per-question-penalty grows higher the longer that string of wrong answers is. One or two questions won't do much damage, but 5 is pretty bad and more than that is severe - your score will just tank.

Translation: you have to be VERY careful with your timing. In general, it's not a good idea to try to "game" the test by spending more time early and less time later on. Most people end up taking it too far and hitting the 5+ situation. I'm not telling you you can't do it - that's your call, of course. But I would be extremely hesitant to do that myself. It's just too hard to know exactly how much time you're spending and not go over early on.

[Raj]

Stacy,

thanks for your reply.


I have attached an image of my score details on the 2 CATs of GMAT Prep. This might also help others in analyzing the scoring pattern.

I am curious why you say my score was affected more by drop in quant (which dropped from 50 to 48) than by verbal (that dropped from 42 to 35). Can you please explain.

Also, I agree with you that I should not be spending too much time on any one question but I was mentioning that I would spend a bit more than the average 2 mins on the first 1/3 of the exam. I totally agree with your other points regarding running out of time towards the end. I typically have been ok there, and that makes me believe I did not pay enough attention to the first set of questions in this specific exam.

Thanks for taking the time to respond to this rather weird post ;-)

-Raj.

[StaceyKoprince]

Oh, I'm sorry, I read it backwards! I'm so used to seeing quant scores first that I guess I didn't even notice the V and then the Q! So reverse that comment, but the other advice about timing still does apply.

It's tough to interpret the attached data without also knowing the difficulty level of the questions you were answering. One thing I do note is that you had a long string of right answers on the first test - which allowed you to build your score up to a very high level. Most of your wrong answers came after that, but the first few were presumably on very difficult questions (because of the long string of right answers just before) and even the later ones were still presumably at higher difficulty levels than the ones you were getting wrong on your second test.

I'm guessing that that string of right answers (and all that that implies about the difficulty levels) is the big difference here.

[Raj]

Thanks Stacy. I agree with you analysis about the string of right questions.

I re-installed GMATPrep so I could take 2 more exams. Although I scored pretty close to my first time around(770), I had around 5-8 repeats in both sections, so I dont know how to interpret my score.

I have seen that a lot of posters suggest retaking the exams. Do they mean re-installing?

Is there a difference between re-installing Vs resetting the exams by answering "no" to one of the prompts when selecting the exam?

Any help will be greatly appreciated. I will be taking GMAT in 2 weeks and would like to have a shot at at least one more real GMAT without too many repeat questions.

-Raj.

Oh, I'm sorry, I read it backwards! I'm so used to seeing quant scores first that I guess I didn't even notice the V and then the Q! So reverse that comment, but the other advice about timing still does apply.

It's tough to interpret the attached data without also knowing the difficulty level of the questions you were answering. One thing I do note is that you had a long string of right answers on the first test - which allowed you to build your score up to a very high level. Most of your wrong answers came after that, but the first few were presumably on very difficult questions (because of the long string of right answers just before) and even the later ones were still presumably at higher difficulty levels than the ones you were getting wrong on your second test.

I'm guessing that that string of right answers (and all that that implies about the difficulty levels) is the big difference here.

[StaceyKoprince]

Once you've done the two tests, no matter how you do them again, you generally have a chance of getting questions you've seen before - each test is pulling from a static database.

Now, if it's been a long time, and you download the software again, then they may have swapped out some of the questions in the database, and so you might have less chance of seeing repeats - but there's still a good chance you'll see some repeats. Unfortunately, there's no way to control this.

Here's how you can minimize the potential positive impact of a repeat, though: if you see a question you recognize to the point that you either know what the answer is or have a pretty good idea without having to do all of the normal work, immediately glance at the timer. Make yourself spend the next 2 min (or 1.5 for SC) sitting on that problem, even if you can answer it in 5 seconds. This way, at least, you don't give yourself a timing advantage on other questions you haven't seen before. (And make yourself go through the work anyway on the problem - use the time to see if you can figure out a better way to do the problem or something like that.)