most f-ed retake...

[drizzt3117]

Advertisement - Members don't see this ad

I talked to some guy today who took the MCAT in June and received a 33Q: 13BS 13PS 7VR.

He retook the test in August and just got his score back, 32R:

10BS 8PS 14VR

So... some schools take the best score in each section (Tufts) and that would give him a 40R, while schools that averaged the scores would give him a 32.5, and schools that took the newest score would give him a 32.

Is this guy really better off having retaken, given the circumstances? Clearly he didn't show overall improvement and dropped substantially in the sciences, but has shown the potential for high scores in each section. I had no idea what to tell him. Clearly he's much better off for the schools that supposedly take the highest score in each section, but are they really going to consider his score better than someone who took the test once and got 37+?

 

[SN2ed]

It depends on why he took the test. If his verbal score was preventing him from getting in, then it would be a good choice. I can see a 7 in verbal holding someone back more than an 8 in PS. I don't think they'd take him over someone who scored a 37 in the first go, unless if they just take the highest from each section. Even then I'm not sure. I wouldn't suggest a retake with that score unless it was holding you back somehow. I really doubt that would be the case with most people. I think better ECs would do more than a slightly higher MCAT.

 

[koopa_troopa]

I talked to some guy today who took the MCAT in June and received a 33Q: 13BS 13PS 7VR.

He retook the test in August and just got his score back, 32R:

10BS 8PS 14VR

So... some schools take the best score in each section (Tufts) and that would give him a 40R, while schools that averaged the scores would give him a 32.5, and schools that took the newest score would give him a 32.

Is this guy really better off having retaken, given the circumstances? Clearly he didn't show overall improvement and dropped substantially in the sciences, but has shown the potential for high scores in each section. I had no idea what to tell him. Clearly he's much better off for the schools that supposedly take the highest score in each section, but are they really going to consider his score better than someone who took the test once and got 37+?

Thats an interesting story. The guy went from below average verbal to top 1% verbal. I would imagine the schools that take the best section score would use some common sense. Your friend seems very fluky and inconsistent. I'm sure ADCOMs will notice this. If I were an ADCOM I would put the guy who actually got a 37+ score over someone who technically never got over 33.

I had almost the same score (13/12/7) and retook the test, with much better results.

 

[drizzt3117]

I think the guy had practice test results in the mid-high 30s which was the reason he retook, but I guess I just don't understand how you drop 5 points in PS. I don't remember the exact MCAT retester data but certainly there has to be a singularly low percentage of that happening.

 

[BrokenGlass]

I think the guy had practice test results in the mid-high 30s which was the reason he retook, but I guess I just don't understand how you drop 5 points in PS. I don't remember the exact MCAT retester data but certainly there has to be a singularly low percentage of that happening.

Computer-based MCATs vary greatly in terms of difficulty. In addition, if a test-taker is lucky and gets science topics s/he is familiar with, s/he will probably score really high, else s/he will probably score very low. The same is true of verbal. If a given testaker understands the passages well, s/he can score really high on verbal, but if a few passages seem incomprehensible to this person, s/he will score really low. It really is that simple.

MCAT doesn't measure intelligence all that well. Any given test taker gets a very small subset of questions that MCAT could potentially test and his/her MCAT score will depend heavily on what passages s/he encounters. MCAT should be used the way a drunk uses a lamppost: for support, not for illumination.

Also, lucky guesses play a huge role and could probably swing the score in either direction by a few points. I would not at all be surprised to see someone score a 38 and then a 28 or vice versa within the same weak.

 

[koopa_troopa]

Computer-based MCATs vary greatly in terms of difficulty. In addition, if a test-taker is lucky and gets science topics s/he is familiar with, s/he will probably score really high, else s/he will probably score very low. The same is true of verbal. If a given testaker understands the passages well, s/he can score really high on verbal, but if a few passages seem incomprehensible to this person, s/he will score really low. It really is that simple.

MCAT doesn't measure intelligence all that well. Any given test taker gets a very small subset of questions that MCAT could potentially test and his/her MCAT score will depend heavily on what passages s/he encounters. MCAT should be used the way a drunk uses a lamppost: for support, not for illumination.

Also, lucky guesses play a huge role and could probably swing the score in either direction by a few points. I would not at all be surprised to see someone score a 38 and then a 28 or vice versa within the same weak.

Just from my own experience, my first test I got all the topics I like for BS and got a 13. On my retake I got a lot of topics I don't like and still got a 13. I think the main reason for the fluctuations is due to the number of administrations and test takers. Before everyone took the test on april or august, so you are effectively competing with everyone. Now that there are 22 tests, you are only competing with a fraction of the people. Whether that group is smart or not can affect your MCAT scores.

I agree to some extent with the lucky guesses part. It has become especially important since the number of questions have been reduced for the CBT. However, going from a 38 to a 28 is a stretch.

 

[axp107]

This thread actually gives me hope.. for the sciences atleast...

If the guy can go from 13s to 10s and 8s on sciences... I can go from the 10s I got on Aug 20th to 13s on the real thing (since I've done it on practice tests). It really seems so erratic to me... lol... too many test administrations.

I noticed that a LOT of people on SDN did really well on Aug 9th... whereas not so many did well on the 20th. Perhaps, Aug20th people were really well prepared since they had the whole summer to study. Hmmm, I have to strategically pick a good test date 😀

I don't think going from 28 -> 38 is a stretch.. a guy posted last week going from a 32 -> 38 in 2 weeks... and I know people who've gone from 30s to 34s/35s in 2 weeks as well.

 

[drizzt3117]

My thoughts are the following:

1) Statistically, there isn't a big difference between a 11 and a 13 on a section, probably just a few questions, so certainly luck will play a part here in terms of guessing. The difference between a 33 and 39 on the MCAT could be relatively small in terms of the amount of questions answered correctly, which is why medical schools don't consider the difference between a 33 and 39 to be anywhere near as large as the difference between a 27 and 33.

2) However, there is a huge difference between a 8 and 13 on a section and of course an even larger difference between a 7 and 14. Part of this comes from the fact that the MCAT is a test with a broad breadth of subjects but relatively little depth (I mean, how in depth can you get with only ~175 questions?) If you get subjects that you are familiar with, you might do extremely well, but if you get unfamiliar subjects you might do poorly.

3) My thoughts would be that the person I described is probably someone with a good deal of knowledge in some subject areas of the MCAT but not all, so he excelled when he got topics he knew but bombed when he got passages in unfamiliar topics. I feel that if I took the test a few more times (which I wouldn't do, of course) I'd get a similar score because I have a good understanding of all of the subject matter (and my practice tests show that) but without a solid understanding of absolutely everything that might be covered, luck is going to play some role in your score, while practice and mastery of the subject material might minimize subject matter related fluctuations.

4) All that said, as an adcom I would find a drop from 13 to 8 on PS especially troubling, even given the meteoric 7 to 14 increase on VR.

 

[BrokenGlass]

My thoughts are the following:

1) Statistically, there isn't a big difference between a 11 and a 13 on a section, probably just a few questions, so certainly luck will play a part here in terms of guessing. The difference between a 33 and 39 on the MCAT could be relatively small in terms of the amount of questions answered correctly, which is why medical schools don't consider the difference between a 33 and 39 to be anywhere near as large as the difference between a 27 and 33.

2) However, there is a huge difference between a 8 and 13 on a section and of course an even larger difference between a 7 and 14. Part of this comes from the fact that the MCAT is a test with a broad breadth of subjects but relatively little depth (I mean, how in depth can you get with only ~175 questions?) If you get subjects that you are familiar with, you might do extremely well, but if you get unfamiliar subjects you might do poorly.

3) My thoughts would be that the person I described is probably someone with a good deal of knowledge in some subject areas of the MCAT but not all, so he excelled when he got topics he knew but bombed when he got passages in unfamiliar topics. I feel that if I took the test a few more times (which I wouldn't do, of course) I'd get a similar score because I have a good understanding of all of the subject matter (and my practice tests show that) but without a solid understanding of absolutely everything that might be covered, luck is going to play some role in your score, while practice and mastery of the subject material might minimize subject matter related fluctuations.

4) All that said, as an adcom I would find a drop from 13 to 8 on PS especially troubling, even given the meteoric 7 to 14 increase on VR.

You make good points. Also, the same person can see a given problem and know the answer (almost) immediately at a given point in time, but a little while later this same person may take much longer to figure out the exact same problem (it has happened to me many times). Since MCAT is a timed test, this could hurt your score if you are unwilling to let go of these questions and this could also account for variance. Everybody knows that it's not good to spend too much time on any one question because the price paid will probably be taking too many blind guesses at the end of the section, but sometimes it's easier said than done...

 

[PreMD86]

Just from my own experience, my first test I got all the topics I like for BS and got a 13. On my retake I got a lot of topics I don't like and still got a 13. I think the main reason for the fluctuations is due to the number of administrations and test takers. Before everyone took the test on april or august, so you are effectively competing with everyone. Now that there are 22 tests, you are only competing with a fraction of the people. Whether that group is smart or not can affect your MCAT scores.

I agree to some extent with the lucky guesses part. It has become especially important since the number of questions have been reduced for the CBT. However, going from a 38 to a 28 is a stretch.

Great point. I remember I tried very hard to register for aug 20th exam, because I knew the kids taking it are likely the kaplan students that only started studying this summer and had to take the test when their course ended in mid-august. I think next year will be full of speculation on which dates will have the more forgiving curve.

 

[drizzt3117]

Yeah, the conventional wisdom with the August paper/pencil would be that people would have been studying the entire summer and it would be a difficult curve. I'm not sure that was the case with my exam. I felt that Bio was just about right, Verbal was easy, Phys was hard, and got the same score on all three.

 

[Dukhtar]

Thats an interesting story. The guy went from below average verbal to top 1% verbal. I would imagine the schools that take the best section score would use some common sense. Your friend seems very fluky and inconsistent. I'm sure ADCOMs will notice this. If I were an ADCOM I would put the guy who actually got a 37+ score over someone who technically never got over 33.

I had almost the same score (13/12/7) and retook the test, with much better results.

Actually, I believe they rank you against all previous test takers who've had similar passages. So it doesn't matter as much how smart the people you took it with are.

 

[BrokenGlass]

Actually, I believe they rank you against all previous test takers who've had similar passages. So it doesn't matter as much how smart the people you took it with are.

Yeah. Someone posted a reply from AAMC regarding this. The curve is pre-set. It doesn't matter if the people taking the same test as you are a stronger or a weaker pool or test-takers. So it's really strange that we have to wait for a month to get our scores.

 

[MamadP]

Something very similar actually happened to me on my retake. I went from 6 to a 11 in VR but dropped in PS from a 13 to a 9! Overall, I had a 3 point increase in my score though, so I'm not that angry.

The main reason I went down in PS was b/c of bad testing conditions. My testing center's A/C was Fu*ked up and I was dying during PS. I soaked my hair and sat for the remaining section, which clearly helped a lot. I would suspect that something similar happened to this guy...