Real Test Curve vs. AAMC practice curves

[youcouldnever2]

Everyone says that not to worry about a real MCAT that is harder than the practice tests because there will be a greater curve. But realisticly what do you think is the number wrong that you can get and still get the same score with a curve.

What i'm trying to say is if you need to get 44 questions correct for an 11 in
BS, for most of the AAMC practice test, even if the real test is harder, i don't think the curve would pull the number of correct say to 39 or 37 for an 11. I would think that even if the test is difficult, it will at most lower the number correct by 1 or 2 questions at most. So you would still need like a 42 or 43 correct answers to get an 11.

Does anyone have an idea if i'm right or wrong in assuming this? Or any more insight on how the entire process works?

---

Members don't see this ad.

 

[Zendoc]

Everyone says that not to worry about a real MCAT that is harder than the practice tests because there will be a greater curve. But realisticly what do you think is the number wrong that you can get and still get the same score with a curve.

What i'm trying to say is if you need to get 44 questions correct for an 11 in
BS, for most of the AAMC practice test, even if the real test is harder, i don't think the curve would pull the number of correct say to 39 or 37 for an 11. I would think that even if the test is difficult, it will at most lower the number correct by 1 or 2 questions at most. So you would still need like a 42 or 43 correct answers to get an 11.

Does anyone have an idea if i'm right or wrong in assuming this? Or any more insight on how the entire process works?

Okay, when I looked at the posts of people who took the exam on April 7th and already received their scores, one person posted their scores and percentile ranges. From that, I calculated the least number of questions right for her score ranges. They are as follows:

In physical sciences, (on that specific test keep in mind) you could get as low as 42 out of 52 correct for an 11. In the verbal reasoning, the percentile ranges were surprisingly low, for a 10 you could get as low as 28 out of 40 correct. For biological sciences to earn an 11 you had to get at least 41 out of 52.

Please keep in mind these were computed from one persons scores and ranges they posted and that they are also only specific to that particular test. Nonetheless, it gave me some comfort last night...good luck waiting for your scores, I have bitten my nails so short, I am now chewing on my arms. 🙂

 

[DiverDoc]

Its all based on the testing population and percentiles. One form might have been "easier" in a certain section and therefore a greater percentage of people got more answers correct. In that case it would be a tighter curve, so 42 out of 52 might be a 9. While on a "tougher" exam, 42 out of 52 may be at a higher percentile range in that testing population and could represent a score of 10. Hope this helps

 

[badasshairday]

I honestly think the AAMC practice tests are misleading. At first I thought they were the real deal and expected that the MCAT was only hard because it was easy and everyone scores high so the curve is tight. This really isn't true. I don't think the AAMC practice tests are ACTUAL past MCATs. I beleive they just pick some passages out of a pool and make a practice test out of that. Towards the end of my MCAT study I was tired of doing Kaplan tests because I though they were not representative of the real MCAT because the PS especially was much harder than that of the AAMC practice tests. Boy did those AAMC practice tests mislead me, the PS had a ****load of calculations, more similar to Kaplan practice tests.

However, with that being said, my friend took the test on April 16 and said that PS was just like the AAMC practice tests but the bio was weird.

It really just depends on your test form as you can see from the poster above who calculated the amount wrong you can get from the April 7 test. 28 out of 40 on the VR to get a 10 is very low when compared to the AAMC practice CBT scaling. Most of those had 31-33 ranges for a 10.

Screw AAMC and the MCAT. The MCAT should go to hell. 😡

 

[jochi1543]

In the verbal reasoning, the percentile ranges were surprisingly low, for a 10 you could get as low as 28 out of 40 correct.

WOW. *flicker of hope in the heart of darkness*

 

[qtrlifecrisis]

Okay, when I looked at the posts of people who took the exam on April 7th and already received their scores, one person posted their scores and percentile ranges. From that, I calculated the least number of questions right for her score ranges. They are as follows:

In physical sciences, (on that specific test keep in mind) you could get as low as 42 out of 52 correct for an 11. In the verbal reasoning, the percentile ranges were surprisingly low, for a 10 you could get as low as 28 out of 40 correct. For biological sciences to earn an 11 you had to get at least 41 out of 52.

Please keep in mind these were computed from one persons scores and ranges they posted and that they are also only specific to that particular test. Nonetheless, it gave me some comfort last night...good luck waiting for your scores, I have bitten my nails so short, I am now chewing on my arms. 🙂

I don't mean to discourage you, but I wouldn't get my hopes up based on this information.

As a previous poster said, the stats above are based on percentiles, not raw score percentages. So if someone who gets a 10 has a lower bound percentile of 70, that means she has done better than 70% of the people who took it, not that she has scored 28/40 (=70%).

 

[CATallergy]

yeah, the curve probably varies from test to test.

in theory, there could even be a really hard test without a curve

for example:
15/52 right = 15,
14/52 right = 14, etc

 

[youcouldnever2]

yeah, the curve probably varies from test to test.

in theory, there could even be a really hard test without a curve

for example:
15/52 right = 15,
14/52 right = 14, etc

Well in theory i suppose you are right, i'm just saying the probability that a harder test shifts the number of correct dramatically for a corresponding scores is low. I base this on the fact that AAMC is been doing this a while, and they are not going to make a test that much harder than any other test. So i think the curve is not a big factor from test to test. Of course, i hope i'm wrong.

 

[WurangBlang]

In physical sciences, (on that specific test keep in mind) you could get as low as 42 out of 52 correct for an 11. In the verbal reasoning, the percentile ranges were surprisingly low, for a 10 you could get as low as 28 out of 40 correct. For biological sciences to earn an 11 you had to get at least 41 out of 52.

mm, that sounds encouraging... My score comes out tomorrow and if the verbal reasoning curve is that generous, I would be very happy (Freaking verbal was hard on apr 12.)

 

[ILPsychDoc]

Well in theory i suppose you are right, i'm just saying the probability that a harder test shifts the number of correct dramatically for a corresponding scores is low. I base this on the fact that AAMC is been doing this a while, and they are not going to make a test that much harder than any other test. So i think the curve is not a big factor from test to test. Of course, i hope i'm wrong.

I disagree... it seems that the April tests all had their specific killer sections, April 7th was bio, April 12th was verbal... they can vary the difficulty of the test as much as they want as long as the questions are good questions, so if they make a harder PS section so be it, everyone scores lower and the curve goes up...

Maybe not by a HUGE amount but how big a difference do you think there is between a "easy" Bio passage where the average correct is 80% and a "hard" bio where the average is 40%, now imagine several "easy" passages vs several "hard" passages... you can see how the curve would be effected.