ACTUAL average MCAT score nationwide

[T12]

So...according to the Kaplan Medical School Search function (citing data from MSAR 2007-2008), the average MCAT score per matriculant for any given institution in the Unites States doesn't reach well beyond 30 at all.

http://www.kaptest.com/medschoolsearch

Not sure if an account is necessary to use search function...

This would seem to disagree with some numbers I've seen from various sources including the voices being heard on this site.

I see it stated all the time that anything below a 30 spells doom for an applicant, yet in many cases, a sub-30 is the average, meaning a large percentage of accepted students don't even hit 30.

What's the deal?

---

Members don't see this ad.

 

[Raryn]

Per the AAMC:
http://www.aamc.org/data/facts/2007/2007mcatgpa.htm

Average last year was 30.8

Most likely approximately half the people who got in had below that.

Edit: Heres the US News data for all the individual schools matriculated averages in 2007. I'll bet $ its more recent than Kaplans.

 

[se2131]

dunno, but it's definitely wrong. UVA doesn't actually have the highest MCAT average in the country. WashU's MCAT average is definitely not 31...

Use this instead: http://www.aamc.org/data/facts/

 

[Decicco]

The current MSAR made it look like the average matriculant had an MCAT of 32 (11 10 11), but these numbers were rounded. WashU had an avg of 38 (the highest in the country).

 

[T12]

From what I can tell, the average per the provided source is, as stated, a 30.8 (10.3,9.9,10.6) (matriculants).

So, wouldn't this agree with the Kaplan numbers?

And if 30.8 is at the peak of the bell curve, definitely at least 20% of matriculants scored below a 30.

Judging by popular opinion around here alone, one would expect that number to be somewhere between 0-5%.

 

[harposnarpo]

Per the AAMC:
http://www.aamc.org/data/facts/2007/2007mcatgpa.htm

Average last year was 30.8

Most likely approximately half the people who got in had below that.

Edit: Heres the US News data for all the individual schools matriculated averages in 2007. I'll bet $ its more recent than Kaplans.

Super interesting. Comparing the spreadsheet you attached (averages of matriculated) with the MSAR (median of accepted):

Washu MSAR ~= usnews
Harvard MSAR > usnews
Yale MSAR < usnews
Georgetown MSAR ~= usnews
Loyola MSAR ~= usnews

Comparing this with MSAR 10% and 90% data would make for interesting analysis. Can anyone figure that out?

 

[T12]

and per Kaplan, Virginia >>> rest of universe

 

[DrMattOglesby]

now im wondering if i ABSOLUTELY NEED to retake my MCAT after seeing all these confounding data ... 😕

 

[Raryn]

now im wondering if i ABSOLUTELY NEED to retake my MCAT after seeing all these confounding data ... 😕

Your overall score would make it doable, but a 6 in any individual section is pretty fatal. Even a 7 would be iffy.

 

[T12]

Even before seeing this, I've believed that a high GPA with a 28, 29, or 30 is sufficient for your average school. If you have above a 3.9 and around a 30 (especially in some states), I think you're set with the stats. Bloated MCAT standards seem to run rampant.

This is assuming, of course, that that 28, 29, or 30 is a balanced score...9,10,10 or such

 

[Perfectionist]

so what's the average MCAT score of everyone that takes the test, not just the matriculating students?

 

[Raryn]

From what I can tell, the average per the provided source is, as stated, a 30.8 (10.3,9.9,10.6) (matriculants).

So, wouldn't this agree with the Kaplan numbers?

And if 30.8 is at the peak of the bell curve, definitely at least 20% of matriculants scored below a 30.

Judging by popular opinion around here alone, one would expect that number to be somewhere between 0-5%.

Assuming I remember how to add standard deviations together correctly, its 30.8 +/- 3.12. Of course, it cant be a bell curve around that though, it would be way too top heavy. There might be 600 people that get 39+ and get in, but I highly doubt theres that many people that score under 21 and get in.

But lets assume its fairly normally distributed around the mean. That means that 95% of the people who get in have a 25-37. Thats probably close to correct...

Edit: But only 13.6% of matriculated students would have a score between 25 and 28.

 

[Raryn]

so what's the average MCAT score of everyone that takes the test, not just the matriculating students?

25.1

http://www.aamc.org/students/mcat/examineedata/combined07.pdf

 

[T12]

Raryn, what you're saying doesn't exactly fit my less-sophisticated estimates, and I'll take yours over mine, but they still seem to invalidate the popular opinion that someone who scores sub-30 should definitely retake the MCAT.

If around 14% is 25-28, 25-30 would be around 20% or more. (Probably 10x as many accepted who scored 28-30 than 25-27, right?)

 

[Tutmos]

Raryn, what you're saying doesn't exactly fit my less-sophisticated estimates, and I'll take yours over mine, but they still seem to invalidate the popular opinion that someone who scores sub-30 should definitely retake the MCAT.

If around 14% is 25-28, 25-30 would be around 20% or more. (Probably 10x as many accepted who scored 28-30 than 25-27, right?)

I think you're missing one factor. URM's. That likely skews the median, including the matriculating median.

 

[Raryn]

Raryn, what you're saying doesn't exactly fit my less-sophisticated estimates, and I'll take yours over mine, but they still seem to invalidate the popular opinion that someone who scores sub-30 should definitely retake the MCAT.

If around 14% is 25-28, 25-30 would be around 20% or more. (Probably 10x as many accepted who scored 28-30 than 25-27, right?)

(Assuming a normal distribution) 13.6% have between a 25 and a 28. Another 34.1% have between a 28 and a 31, 50% have more than a 31, and only 2.5% of students have under a 25. Don't hold too much stock by those numbers though, because it is definitely a skewed distribution.

For example Tutmos is right, URMs skew the numbers. Not saying all of them have bad stats, but at the very least the 3 HBCUs take up a high portion of the low number folks. (The average for MCAT score at Howard is ~24.6).

http://www.aamc.org/data/facts/2007/mcatgparaceeth07.htm gives enough data to figure out exactly how much the numbers are skewed by URMs, but my limited knowledge of stats isn't enough to do it for sure... For white matriculated students the average MCAT score is 31.2 +/- 2.84, and its even higher for Asian students. The lower standard deviation even gives a tighter curve.

 

[radslooking]

Per the AAMC:
http://www.aamc.org/data/facts/2007/2007mcatgpa.htm

Average last year was 30.8

Most likely approximately half the people who got in had below that.

Edit: Heres the US News data for all the individual schools matriculated averages in 2007. I'll bet $ its more recent than Kaplans.

wow, wash u 37. (average!)

equally amazing...ponce in puerto rico 21!
howard 24.6! yikes. average 24.6? that's not good....

 

[LikeClockWork]

wow, wash u 37. (average!)

equally amazing...ponce in puerto rico 21!
howard 24.6! yikes. average 24.6? that's not good....

Eh. There are two ends for each bell curve.

 

[LizzyM]

Use that AMCAS chart to compare your own scores to the average for the race/ethnic group to which you belong. That is a far better measure of whether you have a shot somewhere (of course, even that's fuzzy because state of residence can have a big influence on whether you have a shot at some of the schools with modest average MCATs).

Do not assume a normal distribution. It is much more likely that the numbers are skewed to the left with a long right tail. The rare exceptions of low MCAT scores for non-URM often have a story related to some special consideration (a VIP applicant of some kind).

Imagine this picture for one school (hypothetical):
<30 = 10% of those offered admission
30 = 25%
31 = 25%
32 = 20%
33 = 10%
34 = 5%
35 = 5%
>35 = 5%

The median is 31 (half are above and half are below)

Now if most of those high MCAT applicants go elsewhere and the number who matriculate is half of those who are offered admission:

<30 = 20% of those who matriculate
30 = 40%
31 = 30%
32 = 5%
>32 = 5%

The median is 30 but the distribution is quite different.

 

[T12]

So while the numbers continue to be tinkered with, the trend still seems to suggest, by all estimates, that a MCAT of 3-31 will account for at least half of students at an average school (as much as 90% by the estimate directly above)

So, once again, why do you see it seen so often that student with a score around 30 is directed to take the test again?

 

[LizzyM]

So while the numbers continue to be tinkered with, the trend still seems to suggest, by all estimates, that a MCAT of 3-31 will account for at least half of students at an average school (as much as 90% by the estimate directly above)

So, once again, why do you see it seen so often that student with a score around 30 is directed to take the test again?

It depends on the breakdown of the scores (V15, P7, B8 would warrant a retake in my book, but 10/10/10 might be ok at mid-tier private schools).

It also depends on the student's goal. If the student has a very high gpa from a top school, research experience, & a desire to do a thesis in med school then retaking so as to have a shot at Harvard/Yale/Hopkins/Case Western/ Mayo might make sense. If the applicant has a state school that gives a strong preference to in-state applicants, and a decent gpa, then the 30 might be adequate. On the other hand, if the applicant is an ORM from California, then retaking might be a good idea.

 

[TheRealMD]

The really important thing most people forget is that the figures AAMC has is after classes are all set and done. This is after applicants have been shuffled left and right and the last minute acceptances.

That being said, no one should plan on being a last minute acceptance (if it happens, it happens). So with all the other stuff said in this thread, take the figures you see as a floor and not just a goal to reach but to do much better than.

 

[nontrdgsbuiucmd]

What I see as confusing for those with average statistics backgrounds (most of us I'd bet) is that AAMC seems to publish averages per section, rather than averages overall. MSAR seems to be taking the same tact. Given its impossible to be average (how does one earn a 10.8?) for any section, and likewise impossible to be average on all sections, I'd think it would be MUCH more helpful to see what the overall MCAT averages and overall standard deviations are. I'd doubt that any individual scored, say roughly 5% below average on all sections, or roughly 10% above average on all sections.

To what extent are scores tied together/correlated? I've read on SDN anecdotally about the science person who's above 11 on BS/PS but below 9 on VR. To me, it seems quite likely that the science major who is very strong at BS is weaker at VR than the person who worked as a journalist for the past 5 years, the capabilities required to do well in these areas seem to be too disparate.

So here's what I think would be really cool to see posted: from AAMC, averages for the full MCAT. From schools, averages for the MCAT for accepted applicants, as well as minimum section scores required to consider an applicant. If a school requires a 32 and 3.65 GPA, or a min 10 section score, to consider an OOS app, why isn't this stated when the secondary invite app is sent out? (in the case of the student who does not meet the threshold).

 

[LizzyM]

What I see as confusing for those with average statistics backgrounds (most of us I'd bet) is that AAMC seems to publish averages per section, rather than averages overall. MSAR seems to be taking the same tact. Given its impossible to be average (how does one earn a 10.8?) for any section, and likewise impossible to be average on all sections, I'd think it would be MUCH more helpful to see what the overall MCAT averages and overall standard deviations are. I'd doubt that any individual scored, say roughly 5% below average on all sections, or roughly 10% above average on all sections.

To what extent are scores tied together/correlated? I've read on SDN anecdotally about the science person who's above 11 on BS/PS but below 9 on VR. To me, it seems quite likely that the science major who is very strong at BS is weaker at VR than the person who worked as a journalist for the past 5 years, the capabilities required to do well in these areas seem to be too disparate.

So here's what I think would be really cool to see posted: from AAMC, averages for the full MCAT. From schools, averages for the MCAT for accepted applicants, as well as minimum section scores required to consider an applicant. If a school requires a 32 and 3.65 GPA, or a min 10 section score, to consider an OOS app, why isn't this stated when the secondary invite app is sent out? (in the case of the student who does not meet the threshold).

As with the GRE, it is not good science to add the three sections together and treat them as one. Furthermore, some schools like a high verbal, othes will accept a lower verbal but look for a slightly higher physical science score as it seems to be a better predictor of success at that school. An average of 10.8 could mean that most of the applicants scored an 11 or higher and the rest scored 10 or less bringing the average above 10 but less than 11. So, aim for above average for that school in each section and you'll be safe. Aim for above average in 2 sections and less than 0.9 below average in the third section and you should be relatively OK.

Why don't schools publicize thresholds? Because for every 99 applicants who they will toss for being below the threshold there is 1 applicant whom they are willing to consider because of something else the applicant brings to the table. (This can be a donor, VIP, political connections, or other special consideration). Isn't fair, can't say I like it, but it happens.

 

[Raryn]

What I see as confusing for those with average statistics backgrounds (most of us I'd bet) is that AAMC seems to publish averages per section, rather than averages overall. MSAR seems to be taking the same tact. Given its impossible to be average (how does one earn a 10.8?) for any section, and likewise impossible to be average on all sections, I'd think it would be MUCH more helpful to see what the overall MCAT averages and overall standard deviations are. I'd doubt that any individual scored, say roughly 5% below average on all sections, or roughly 10% above average on all sections.

To what extent are scores tied together/correlated? I've read on SDN anecdotally about the science person who's above 11 on BS/PS but below 9 on VR. To me, it seems quite likely that the science major who is very strong at BS is weaker at VR than the person who worked as a journalist for the past 5 years, the capabilities required to do well in these areas seem to be too disparate.

So here's what I think would be really cool to see posted: from AAMC, averages for the full MCAT. From schools, averages for the MCAT for accepted applicants, as well as minimum section scores required to consider an applicant. If a school requires a 32 and 3.65 GPA, or a min 10 section score, to consider an OOS app, why isn't this stated when the secondary invite app is sent out? (in the case of the student who does not meet the threshold).

I see no reason why adding the individual section averages wouldn't give you the overall average, as long as you add the standard deviations properly (which I did so above).

 

[LovelyBRass]

I see it stated all the time that anything below a 30 spells doom for an applicant, yet in many cases, a sub-30 is the average, meaning a large percentage of accepted students don't even hit 30.

What's the deal?

The numbers are warped because state schools (except maybe NY - I'm a little bitter) and schools that generally have a higher disadvantaged student population will have lower numbers. If you don't live in a state with a publicly funded medical school, or if they have higher standards, then you're going to need more competitive numbers. So for example, schools like Howard or Morehouse (correct me if I'm wrong!) will forgive lower stats in the case of disadvantaged students (more so, anyway, then most schools) and Marshall's JCE Med school in West Virginia will interview candidates with lower stats if they are from WV.

Basically, the problem is that statistics for state schools and schools that generally accept disadvantaged students are included in these overall averages. I am from NY, where I don't really see much bias for residents, and I am not disadvantaged - therefore, these nationwide statistics mean very little to me. I know that whatever they say, I should be +1 or +2, and that's minimum.