mcat numbers

[tenndoc]

Advertisement - Members don't see this ad

what do you think is a significant difference between two scores? for example, i dont really see much of a difference between an 8 and 9 in a section or a 27 and 28 total. just curious what you all thought

 

[phltz]

what do you think is a significant difference between two scores? for example, i dont really see much of a difference between an 8 and 9 in a section or a 27 and 28 total. just curious what you all thought

I'd say the difference between 8 and 9 in a section is 1.

 

[gettheleadout]

More significant in lower score ranges, as a couple points means significantly more questions incorrect than a couple points near the upper range where each point is a question's difference.

 

[tenndoc]

I'd say the difference between 8 and 9 in a section is 1.

funny... 🙄

 

[octupus]

what do you think is a significant difference between two scores? for example, i dont really see much of a difference between an 8 and 9 in a section or a 27 and 28 total. just curious what you all thought

I'd say the difference between 8 and 9 in a section is 1.

You're right, 27 isn't too different from 28 and neither is an 8 from a 9 but

1 point / section x 3 sections = 3 point difference

now is 27 very different than 30? yes

funny how math works.

 

[tenndoc]

You're right, 27 isn't too different from 28 and neither is an 8 from a 9 but

1 point / section x 3 sections = 3 point difference

now is 27 very different than 30? yes

funny how math works.

yes i know. i guess i didnt convey my original point well enough. if im an adcom, when i see a person with two mcats come in, what type of difference is enough to get my attention. a rise from a 26 to a 28 or a 27 to a 30 for example. or going from a 6 to a 9 in a section.

 

[Whiskeypunch]

yes i know. i guess i didnt convey my original point well enough. if im an adcom, when i see a person with two mcats come in, what type of difference is enough to get my attention. a rise from a 26 to a 28 or a 27 to a 30 for example. or going from a 6 to a 9 in a section.

2ish is the standard deviation. So a lot. Every point counts.

 

[music2doc]

yes i know. i guess i didnt convey my original point well enough. if im an adcom, when i see a person with two mcats come in, what type of difference is enough to get my attention. a rise from a 26 to a 28 or a 27 to a 30 for example. or going from a 6 to a 9 in a section.

I heard somewhere that it is generally assumed that your score is your score ±2 points, so scoring a 30 means you are comparable to someone who scored between a 28 and a 32. The reliability of the MCAT is, as I recall, around 0.9, so we would expect around a 19% variability in scores due to other factors. If we take a typical scoring chart from the practice exams, that 19% (assumed to be ±9.5%) can mean quite a bit of random variability in scores (about 1 point in either direction per section if we give ±9.5% of the score for each section).

 

[dmf2682]

I heard somewhere that it is generally assumed that your score is your score ±2 points, so scoring a 30 means you are comparable to someone who scored between a 28 and a 32. The reliability of the MCAT is, as I recall, around 0.9, so we would expect around a 19% variability in scores due to other factors. If we take a typical scoring chart from the practice exams, that 19% (assumed to be ±9.5%) can mean quite a bit of random variability in scores (about 1 point in either direction per section if we give ±9.5% of the score for each section).

Very nice analysis, I like it

 

[tenndoc]

I heard somewhere that it is generally assumed that your score is your score ±2 points, so scoring a 30 means you are comparable to someone who scored between a 28 and a 32. The reliability of the MCAT is, as I recall, around 0.9, so we would expect around a 19% variability in scores due to other factors. If we take a typical scoring chart from the practice exams, that 19% (assumed to be ±9.5%) can mean quite a bit of random variability in scores (about 1 point in either direction per section if we give ±9.5% of the score for each section).

noooo stop it... im about to finish my undergrad tomorrow... i dont want to think about any math for awhile

 

[zherussianbear]

30+ is all you need (34+ if you wanna go somewhere with a name), as long as you have good volunteering, shadowing, some research, good LORs, some leadership, and 3.7+ GPA.

Less than 30, and you're looking at a lot of scrutiny.

=> It's essentially a binary system.

 

[gettheleadout]

30+ is all you need (34+ if you wanna go somewhere with a name), as long as you have good volunteering, shadowing, some research, good LORs, some leadership, and 3.7+ GPA.

Less than 30, and you're looking at a lot of scrutiny.

=> It's essentially a binary system.

A 35 vs a 40 would definitely make a difference in applying to competitive schools.

 

[zherussianbear]

A 35 vs a 40 would definitely make a difference in applying to competitive schools.

This seems logical but I don't believe it to be true. I only have anecdotes from maybe 15 friends in the 35-40 range. The lower people with better non-MCAT factors did better. All of the logarithmic graphs (i.e. chances of acceptance versus MCAT that AAMC publishes) corroborate this fact. Decreasing returns to scale.

I had a 34 and a lot of my friends with 32s and 31s did much better than I in terms of interviews.

The difference between a 35 and a 40 is literally a couple of questions. Adcoms know this, and you'll get an II with a 35 and good other stuff just as easily as you will with a 40. Plus, schools like Harvard or UCSF will care much more about your research (given you have a sufficiently high MCAT).

All I'm trying to say is that once you cross one of the magic lines, the battle shifts to the other stuff you're peddling. I should add "early application" to my original list; it's much more important than most of us thought.

 

[gettheleadout]

This seems logical but I don't believe it to be true. I only have anecdotes from maybe 15 friends in the 35-40 range. The lower people with better non-MCAT factors did better. All of the logarithmic graphs (i.e. chances of acceptance versus MCAT that AAMC publishes) corroborate this fact. Decreasing returns to scale.

I had a 34 and a lot of my friends with 32s and 31s did much better than I in terms of interviews.

The difference between a 35 and a 40 is literally a couple of questions. Adcoms know this, and you'll get an II with a 35 and good other stuff just as easily as you will with a 40. Plus, schools like Harvard or UCSF will care much more about your research (given you have a sufficiently high MCAT).

All I'm trying to say is that once you cross one of the magic lines, the battle shifts to the other stuff you're peddling. I should add "early application" to my original list; it's much more important than most of us thought.

Here's my reasoning:

1) Your experience is not generalizable because the sample isn't large enough to be representative of any trends sans confounding variables.

2) The graphs to which you are referring are showing percentage admission in general, not to specific schools, and we're specifically talking about the most competitive schools.

3) Many top schools have average MCAT scores above 34, and going from a 35 to a 40 can move one from near average compared to the school's typical class cohort to near the top 10th percentile.

I agree that certain schools place more emphasis on applicant uniqueness than objective qualification, but this is not true across the board (see WashU's average stats vs Mayo's.)

If data exist to show the effect of MCAT score range on admission to a specific subset of the most competitive schools I would be interested in seeing that.

I also agree that there are other equally important, possibly more important factors than having a high 30's MCAT vs a mid-30's MCAT, such as early application.

 

[Morsetlis]

what do you think is a significant difference between two scores? for example, i dont really see much of a difference between an 8 and 9 in a section or a 27 and 28 total. just curious what you all thought

The higher the scores are, the more significant the difference. That's because scores are statistical.

Subjectively, I would think that getting past the 30 mark is useful, and getting past the 40 mark makes you a genius. 40 is about 99%tile.

 

[RedSox10]

Its been said before but psychologically, a number in the 20s is very different than a number in the 30s. So, a 29 versus a 30 is much different than a 28 versus a 29...

 

[tenndoc]

Just getting opinions. Thanks

 

[octupus]

30 MCAT, accepted to medical school
27 MCAT, accepted to medical school x 10000000

 

[octupus]

27 MCAT, rejected everywhere 🙁

30 MCAT, rejected everywhere 😡 😱

lesson learned: 3 tiny points can significantly change how we feel about the outcome

you can change it to 30 vs 33 and it still holds true

 

[tenndoc]

im a little confused by your two posts octupus

 

[centraldogma]

Its been said before but psychologically, a number in the 20s is very different than a number in the 30s. So, a 29 versus a 30 is much different than a 28 versus a 29...

This is it right here. 27 vs 28 doesn't show much. To an adcom they'd be the same (given that the distribution is even like 10/9/9).

Theres a huge difference between any score above and below 30. From my experience the schools I interviewed at set their bar at numbers at or above their average rather than at 30.

 

[RedSox10]

im a little confused by your two posts octupus

when you get in to med school with a 27, this is unexpected and you are ecstatic. when you get in with a 30, you are excited. When you dont get in with a 27, you're obviously disappointed, but not pissed because its somewhat expected, but not getting in with a 30 is aggravating as you had almost assumed you'd get in....

 

[tenndoc]

gotcha. lol

 

[octupus]

when you get in to med school with a 27, this is unexpected and you are ecstatic. when you get in with a 30, you are excited. When you dont get in with a 27, you're obviously disappointed, but not pissed because its somewhat expected, but not getting in with a 30 is aggravating as you had almost assumed you'd get in....

yep. exactly.

I do see what you mean by a 27 vs 28 though. At the end of the day, it doesn't matter if it's a 27 or a 32, what matters most is just one acceptance. While on the topic, I don't feel like admissions is as much of a crapshoot as everyone says. Sure, it might be if you're a borderline or average applicant, it might be. With a average or subpar MCAT/GPA, it's hard to gauge how far your ECs and PS will take you and how things are going to end up. But for those with good stats and applied wisely, they're sure to get in somewhere good. A 34/3.7 applicant who doesn't get in anywhere certainly isn't doing something right.

 

[MedBound1]

There's a much more significant difference between an 8 and a 9 then there is a 13 and a 15. When you get up in the 12-15 range on any section, each question could cost you close to an entire point.

 

[Knocked Up]

OP, would you like my opinion on the matter. Let me know before I post my opinion...

 

[tenndoc]

OP, would you like my opinion on the matter. Let me know before I post my opinion...

Sure go ahead? I'm not worried about it. Just pondering a question

 

[tenndoc]

The reason I ask is that I bumped my score up 3 points to a 27. And brought up a section by 3 pts.

 

[EnergyDrink]

More significant in lower score ranges, as a couple points means significantly more questions incorrect than a couple points near the upper range where each point is a question's difference.

You can argue the same for the higher point ranges.

I view a score of 8-9 as understanding the material that you know will be on the test. You know there will be something about the digestive tract, or you know there will be electrochemistry on it. This is about 35 questions.

The next 1-2 points require you to make inferences based on what you know, this is about 7 questions.

The last 10 questions that decide a score from 11-15 can be drawn from pretty much anywhere. That means you need to know a larger "bank" of information. Yes, you can luck out and get some right, but law of averages works to balance this out.

So the difference from a 6 to an 8 requires more understanding of standard material, a score from 10 to a 12 requires a lot more knowledge. From 12-15 is pretty much variable, which is why Adcoms don't really care about differences in scores >35

AAMC does use generalizations. My scores may vary 2-4 points with each exam. I scored 3 points lower on my real test than the AAMC practice tests simply because I had an organic chem heavy Bio. They have no way of seeing that, and are pleased because their numbers were balanced overall.

 

[yellowfish]

The difference between a 35 and a 40 is literally a couple of questions. Adcoms know this, and you'll get an II with a 35 and good other stuff just as easily as you will with a 40. Plus, schools like Harvard or UCSF will care much more about your research (given you have a sufficiently high MCAT).

The difference is not a couple of questions. A 5 point difference is significant anywhere on the scale... even from a 40 to a 45. Of course you'll get a II with a 35. But an interview? An acceptance? You bet it's easier with a 40+. We can all agree that the difference gets less and less significant as your score gets higher and higher and that a high MCAT won't make up for other deficiencies in your application, but to say that it just doesn't matter anymore isn't true.

 

[yellowfish]

The reason I ask is that I bumped my score up 3 points to a 27. And brought up a section by 3 pts.

To answer your question, I think that bump makes a huge difference. Definitely has the potential to make the difference between a successful/unsuccessful application season.

 

[BHaus9]

Anyone else think it's strange that it's theoretically possibly for person A to get a lower composite score than person B even if person A gets a greater % of the questions correct overall?

 

[gettheleadout]

You can argue the same for the higher point ranges.

I view a score of 8-9 as understanding the material that you know will be on the test. You know there will be something about the digestive tract, or you know there will be electrochemistry on it. This is about 35 questions.

The next 1-2 points require you to make inferences based on what you know, this is about 7 questions.

The last 10 questions that decide a score from 11-15 can be drawn from pretty much anywhere. That means you need to know a larger "bank" of information. Yes, you can luck out and get some right, but law of averages works to balance this out.

So the difference from a 6 to an 8 requires more understanding of standard material, a score from 10 to a 12 requires a lot more knowledge. From 12-15 is pretty much variable, which is why Adcoms don't really care about differences in scores >35

AAMC does use generalizations. My scores may vary 2-4 points with each exam. I scored 3 points lower on my real test than the AAMC practice tests simply because I had an organic chem heavy Bio. They have no way of seeing that, and are pleased because their numbers were balanced overall.

This is all conjecture, and I haven't seen any proof from anyone that the bolded is true.

 

[Knocked Up]

Sure go ahead? I'm not worried about it. Just pondering a question

Ok, here's the truth then. Some people who haven't applied yet will call me crazy and whatnot, but it's not crazy.

The MCAT is everything. Yes, those three points will help you (if you are a URM they will help a lot). Assuming you got a 24 before, you went from having no chance at all for MD to a very slight chance. What are you ECs and your GPA? They are far second to your MCAT score but they need to be fantastic for you to stand any chance.

You could still use another two to three points. I feel the safe zone for allopathic is around 29-30, anything else will wreck your application entirely. Adcoms don't care, they aren't looking for sympathetic, good natured, well-rounded, or grade dedicated students. They are looking for good standardized test takes (that's it, that's all). Good test takers = high Step scores = good matches = good residencies = successful specialists = alumni donations $$$. Also high MCAT scores factor into rankings.

 

[yellowfish]

This is all conjecture, and I haven't seen any proof from anyone that the bolded is true.

+1. Although a mentor of mine who has been on Harvard's adcom told me that the difference between a 38 and a 42 on the MCAT wasn't a huge deal.

You could still use another two to three points. I feel the safe zone for allopathic is around 29-30, anything else will wreck your application entirely. Adcoms don't care, they aren't looking for sympathetic, good natured, well-rounded, or grade dedicated students. They are looking for good standardized test takes (that's it, that's all). Good test takers = high Step scores = good matches = good residencies = successful specialists = alumni donations $$$. Also high MCAT scores factor into rankings.

Unfortunate, but true.

 

[tenndoc]

Ok, here's the truth then. Some people who haven't applied yet will call me crazy and whatnot, but it's not crazy.

The MCAT is everything. Yes, those three points will help you (if you are a URM they will help a lot). Assuming you got a 24 before, you went from having no chance at all for MD to a very slight chance. What are you ECs and your GPA? They are far second to your MCAT score but they need to be fantastic for you to stand any chance.

You could still use another two to three points. I feel the safe zone for allopathic is around 29-30, anything else will wreck your application entirely. Adcoms don't care, they aren't looking for sympathetic, good natured, well-rounded, or grade dedicated students. They are looking for good standardized test takes (that's it, that's all). Good test takers = high Step scores = good matches = good residencies = successful specialists = alumni donations $$$. Also high MCAT scores factor into rankings.

yeah... not asking for a critique of my numbers. just posing the question to SDN and used my #s as an example

 

[zherussianbear]

The difference is not a couple of questions. A 5 point difference is significant anywhere on the scale... even from a 40 to a 45. Of course you'll get a II with a 35. But an interview? An acceptance? You bet it's easier with a 40+. We can all agree that the difference gets less and less significant as your score gets higher and higher and that a high MCAT won't make up for other deficiencies in your application, but to say that it just doesn't matter anymore isn't true.

A 35 is ~20 questions wrong total. A 40 is ~5-10 questions wrong total. 10 questions difference split among three sections is not a big difference.

I know that everyone here loves to argue that higher is better, but after a cut-off it does not matter. This is a fact, and it's likely because getting a 40 versus a 36 is a stochastic process; you can't study up to a consistent 40+ on multiple MCATs.

Get your score high enough and you're golden; that's all I'm saying. This is true, even for the super competitive schools because they care much more about your publications; they're research institutions.

 

[zherussianbear]

Ok, here's the truth then. Some people who haven't applied yet will call me crazy and whatnot, but it's not crazy.

The MCAT is everything. Yes, those three points will help you (if you are a URM they will help a lot). Assuming you got a 24 before, you went from having no chance at all for MD to a very slight chance. What are you ECs and your GPA? They are far second to your MCAT score but they need to be fantastic for you to stand any chance.

You could still use another two to three points. I feel the safe zone for allopathic is around 29-30, anything else will wreck your application entirely. Adcoms don't care, they aren't looking for sympathetic, good natured, well-rounded, or grade dedicated students. They are looking for good standardized test takes (that's it, that's all). Good test takers = high Step scores = good matches = good residencies = successful specialists = alumni donations $$$. Also high MCAT scores factor into rankings.

Not at all. Sure, they won't even consider you if you're below a threshold. Above it though, you're in the realm of average MD acceptances, and you have to stand out.


Pre-meds should focus on:

-Getting 30+ if they don't care which med school they go to, or 35+ for competitive ones.
-3.7+ GPA (both cum. and sci.)
-Sounding human and organized in the PS, and not being a dick at interviews
-Lots of Shadowing and Volunteering
-Some research to say you did it
-one or two dedicated extracurricular activities
-Good relationships with professors for LORs
-Try to network; if you're at conferences that attract adcom people, and you get them to know you, you're golden and can relax some of the above assumptions.

I should also say: Schools like Harvard don't generally take people out of undergrad directly. Most of those students will be very research oriented, and generally older.

It is by no means, ever, in no shape or form, "just about the MCAT."

 

[phltz]

I know that everyone here loves to argue that higher is better, but after a cut-off it does not matter. This is a fact, and it's likely because getting a 40 versus a 36 is a stochastic process; you can't study up to a consistent 40+ on multiple MCATs.

Are you basing this assertion on anything? Some students may score in the mid 30s on practice tests and land a 40 on the real thing, but some students also score around 40 on their practice tests and then land a 40 on the real thing. The signal to noise ratio gets worse at the higher end of the scale, but it is not 0.

 

[gettheleadout]

Are you basing this assertion on anything? Some students may score in the mid 30s on practice tests and land a 40 on the real thing, but some students also score around 40 on their practice tests and then land a 40 on the real thing. The signal to noise ratio gets worse at the higher end of the scale, but it is not 0.

Yeah, I'm going to agree that zherussian's assertions that A: it is a fact that after some cutoff it doesn't matter and B: that it is impossible to consistently score 40+ are complete and total BS.

 

[zherussianbear]

The signal to noise ratio gets worse at the higher end of the scale, but it is not 0.

Agreed. Not zero, but converging to it; fatter tails = more random. More random = adcoms don't care as much. That's been my argument the whole time.

 

[yellowfish]

A 35 is ~20 questions wrong total. A 40 is ~5-10 questions wrong total. 10 questions difference split among three sections is not a big difference.

I know that everyone here loves to argue that higher is better, but after a cut-off it does not matter. This is a fact, and it's likely because getting a 40 versus a 36 is a stochastic process; you can't study up to a consistent 40+ on multiple MCATs.

How is 10 questions wrong not a big difference? 10 questions is significant any day for the vast majority of individuals, and every single time for a population.

You can't study up to a consistent 40? I averaged slightly above a 40 over all the AAMCs, and have another friend I studied with who did so as well.

What are you basing all of this off of?

Seriously dude, the foundation of all of your statements is incredibly weak. Do you have experience on an adcom? You act as if you designed adcoms themselves.

Agreed. Not zero, but converging to it; fatter tails = more random. More random = adcoms don't care as much. That's been my argument the whole time.

And no, your original statement was that the difference between a 35 and a 40 is "literally a couple of questions" and that you'll get in "just as easily with a 35 as with a 40." Both turned out not to be true. It would appear, however, that you're giving ground. Your modified argument is something that every single person on SDN already admits to.

 

[zherussianbear]

I understand that everyone will have their opinions, and many will disagree with mine. None of you have presented any verifiable, rigorous evidence contrary to my points, other than an exploration of the weakness inherent to my evidence - sure there's plenty. I started off saying this is anecdotal; I have completed no large-scale studies. That said, there is some experience (again, anecdotes, many of them), talking to a lot of adcom members where I got in and where I didn't. My thoughts are the impressions I have formed.

If I were to design a study to explore whether higher != better scores (the null), I would want to collect info on accepted scores from, say, Harvard, WashU, etc. I would then mark the dependent variable binary (1 = accepted, 0 o.w.), and run a logit or probit regression (use STATA software), the latter if it could be shown that the Law of Large Numbers applied (which I think it likely could, though for a better, more cynical, discussion of the Gaussian Distribution, read up on Taleb's The Black Swan) . The independent variable would be the applicant's MCAT score. The regression coefficient's value and p-value would give insight to the problem at hand. I would then run additional regressions, adding in independent variables for hours of community service, GPA, shadowing hours, presence of research, presence of strong LORs, ECs, URM status, and leadership. Likely, the effects of MCAT scores would decrease. In the discussion section, I would note studies that have failed to demonstrate correlation between MCAT and USMLE performance, studies that have demonstrated lack of the MCAT's predictive power on curricular performance, and interviews with adcom members. All of these being reasons for adcoms to focus on factors besides the MCAT.

Luckily, much of this data has been collected and presented on the MSAR, where
Harvard's distribution of MCAT Scores is10th percentile: 32, Median: 37, 90th: 41, and
WashU's is 34, 38, 41 (respectively). If I were to construct an appropriate acceptance distribution for each of these (fatter right tails, and approximately symmetric between 10th and 90th percentiles), and generate applicants from a normal distribution, simulating their decisions based on their probability of success from the acceptance distribution, then run the above described regressions, I expect to see:

A sharp cutoff, followed by a diminishing rate of increase in the resulting binary sigmoid. I would also expect to see a good measure of fit.

Now, all I'm saying is: high MCATs will ensure a thorough inspection of your application, but make sure that when that happens, you've got a lot of other good stuff to sell yourself on. So, yes, study for the damned test, but don't think for a second if you don't get an interview at Harvard it was only because your score wasn't high enough. Also, don't expect to get in, just because it was. Build yourselves as strong, developed, human applicants, and I wish you all the greatest success!

 

[zherussianbear]

Also:

A 5 point difference is significant anywhere on the scale... even from a 40 to a 45.

...

+1. Although a mentor of mine who has been on Harvard's adcom told me that the difference between a 38 and a 42 on the MCAT wasn't a huge deal.



Unfortunate, but true.

 

[yellowfish]

Also:



...

Yes... and yes?

Strong point being made here, keep it up 👍

EDIT: Just going to go ahead and respond to this since you will inevitably bring these points up and I'll feel compelled to respond to them later-

A) Difference between a 40 and a 45 is significant. Less significant than any other 5 point margin. We already talked about this, please stop beating the dead horse.

B) Not "a huge deal." Your entire premise was based on the radical statement that could be applied to this situation as "38 == 42"

C) "Unfortunate, but true." Entirely taken out of context.

 

[MCAT Guru]

Do you all think that there's certain percentile "cut off" where getting higher doesn't make much of a difference? Say, 95% and higher is pretty much all the same?

 

[EverStriving]

This thread is from 2012 btw...

 

[Lawpy]

Do you all think that there's certain percentile "cut off" where getting higher doesn't make much of a difference? Say, 95% and higher is pretty much all the same?

@gyngyn and others mentioned that scores above a 37 are essentially identical. So we're talking 98th percentile and higher (or roughly 519+ on the new exam).