MSAR Updated

Med School Insiders
This forum made possible through the generous support of SDN members, donors, and sponsors. Thank you.
Members do not see this ad.
Does anyone know if US News has updated acceptance rate statistics for IS and OOS applicants? It would be interesting to compare that with the new MSAR changes.
 
Goro said:
And for 520, add UPenn, Northwestern, U Chicago, JHU,

I've had to spend the past 30 mins updating my recommendation file!

It's an arms race out there. I predict we'll see similar stats creep for the DO schools as well.

EDIT: still mulling over the stats. Who'd ever think that Western MI would be nipping at the heels of Keck in terms of median stats????!!!!

Jefferson is also creeping up and away from the other two members of the Philly Triplets. We live in interesting times.
Click to expand...

Goro, how much does this change the schools you recommend for reinvention cases? For people who follow the general comeback route (upward trend/postbacc/SMP) and score 513+ on the MCAT, is MD now less of a possibility?

I got a 514...is it now a waste of time to apply to schools like Duke, Pitt, Mayo, or even Tufts, Jefferson, Tulane, BU?

These schools were recommended to me but its hard to see why schools would take reinvention cases anymore when the the field is this competitive and they can presumably have their choice of shiny, unblemished 3.7+/512+ apps.
 
wildcherry said:
Goro, how much does this change the schools you recommend for reinvention cases? For people who follow the general comeback route (upward trend/postbacc/SMP) and score 513+ on the MCAT, is MD now less of a possibility?

I got a 514...is it now a waste of time to apply to schools like Duke, Pitt, Mayo, or even Tufts, Jefferson, Tulane, BU?

These schools were recommended to me but its hard to see why schools would take reinvention cases anymore when the the field is this competitive and they can presumably have their choice of shiny, unblemished 3.7+/512+ apps.
Click to expand...
You won't know unless you try, though Duke is probably out. Tufts and Tulane are still very possible.
 
wildcherry said:
Goro, how much does this change the schools you recommend for reinvention cases? For people who follow the general comeback route (upward trend/postbacc/SMP) and score 513+ on the MCAT, is MD now less of a possibility?

I got a 514...is it now a waste of time to apply to schools like Duke, Pitt, Mayo, or even Tufts, Jefferson, Tulane, BU?

These schools were recommended to me but its hard to see why schools would take reinvention cases anymore when the the field is this competitive and they can presumably have their choice of shiny, unblemished 3.7+/512+ apps.
Click to expand...
I don't believe that the numbers will change all that much. Schools that reward reinvention want you have a good MCAT score. Most of the schools in what I call Harvard/Stanford class simply have had more join the class. I believe that a 514 is still good for the following:

Case (maybe)
Mayo (maybe)
Duke (maybe)
Case (maybe)
BU (maybe)
Miami
Albany
Rosy Franklin
NYMC
EVMS
Wake Forest
Jefferson
Temple
Drexel
Tulane
Dartmouth
Loyola
Pitt
Hofstra
Tufts
Nova MD
CUSM IF you're from CA
Your state school(s).
 
Goro said:
I don't believe that the numbers will change all that much. Schools that reward reinvention want you have a good MCAT score. Most of the schools in what I call Harvard/Stanford class simply have had more join the class. I believe that a 514 is still good for the following:

Case (maybe)
Mayo (maybe)
Duke (maybe)
Case (maybe)
BU (maybe)
Miami
Albany
Rosy Franklin
NYMC
EVMS
Wake Forest
Jefferson
Temple
Drexel
Tulane
Dartmouth
Loyola
Pitt
Hofstra
Tufts
Nova MD
CUSM IF you're from CA
Your state school(s).
Click to expand...

Thanks Goro, you're the best.
 
TheVisionary said:
late to this thread, but Penn's 10th percentile C/P is a 129 😱
Click to expand...

I mean, could you sleep at night knowing that somebody with a 128 on C/P might one day treat your children (and it will be your children because they wont be able to do well enough on step to match outside of primary care).
 
Lucca said:
I mean, could you sleep at night knowing that somebody with a 128 on C/P might one day treat your children (and it will be your children because they wont be able to do well enough on step to match outside of primary care).
Click to expand...
I dont undserstand how someone could be a good pediatrician with only a well above average understanding of physics and gen chem #notmychild
 
I took a closer look at UCLA, and their overall accepted student stat distributions are far broader than its peer schools. their 10/90 is low 500s all the way to mid 520s and their 25/75 spans 10 points. Geffen's peers have 25/75 distributions that fall entirely within +/- 3-4 points of the median. So, while their median is significantly higher this year, Geffen probably remains one of the few truly holistic top medical schools.

Another example that comes to mind is UCSF. Similarly high median, but a really broad distribution for accepted students. A little tighter overall than UCLA but far, far more "generous" than its peers Hopkins, Harvard, Penn, and Stanford.

The distributions are so tight at places like Hopkins that I wouldn't be surprised if applicants were literally divided into 3 piles: Shred; 130+ in 3/4 sections no section lower than 128; and sub 520 MCAT but exceptional in some way (military, urm, d1 athlete, first-author publications, rhodes scholar, prestigious undergrad, donor baby etc.)
 
Last edited:
Lucca said:
I took a closer look at UCLA, and their overall accepted student stat distributions are far broader than its peer schools. their 10/90 is low 500s all the way to mid 520s and their 25/75 spans 10 points. Geffen's peers have 25/75 distributions that fall entirely within +/- 3-4 points of the median. So, while their median is significantly higher this year, Geffen probably remains one of the few truly holistic top medical schools.

Another example that comes to mind is UCSF. Similarly high median, but a really broad distribution for accepted students. A little tighter overall than UCLA but far, far more "generous" than its peers Hopkins, Harvard, Penn, and Stanford.

The distributions are so tight at places like Hopkins that I wouldn't be surprised if applicants were literally divided into 3 piles: Shred; 130+ in 3/4 sections no section lower than 128; and sub 520 MCAT but exceptional in some way (military, urm, d1 athlete, first-author publications, rhodes scholar, prestigious undergrad, donor baby etc.)
Click to expand...
I don’t buy this notion that UCLA is somehow more holistic than other top schools because their score distribution is wider. Holistic doesn’t mean stats are less important. It just means they aren’t the only important factor. I can guarantee you that top schools with sky high average stats have well rounded and overall impressive student bodies.

It’s just that the top schools like Yale, Hopkins, and Stanford have the luxury of attracting students both with high stats and impressive experiences. These schools routinely reject high stats applicants every year.
 
Lucca said:
I took a closer look at UCLA, and their overall accepted student stat distributions are far broader than its peer schools. their 10/90 is low 500s all the way to mid 520s and their 25/75 spans 10 points. Geffen's peers have 25/75 distributions that fall entirely within +/- 3-4 points of the median. So, while their median is significantly higher this year, Geffen probably remains one of the few truly holistic top medical schools.

Another example that comes to mind is UCSF. Similarly high median, but a really broad distribution for accepted students. A little tighter overall than UCLA but far, far more "generous" than its peers Hopkins, Harvard, Penn, and Stanford.

The distributions are so tight at places like Hopkins that I wouldn't be surprised if applicants were literally divided into 3 piles: Shred; 130+ in 3/4 sections no section lower than 128; and sub 520 MCAT but exceptional in some way (military, urm, d1 athlete, first-author publications, rhodes scholar, prestigious undergrad, donor baby etc.)
Click to expand...

Are first author publications really that exceptional? Would it be worth pushing an app cycle back a year for if you were working on one but it wasn't it published yet? Or could you inform the schools of it when it happens?
 
Noobie Question - why is there a discrepancy b/w class profile data on a school's website and MSAR?

Side note... I feel like an absolute turd w/ a 510 MCAT.
 
scienceguy95 said:
I don’t buy this notion that UCLA is somehow more holistic than other top schools because their score distribution is wider. Holistic doesn’t mean stats are less important. It just means they aren’t the only important factor. I can guarantee you that top schools with sky high average stats have well rounded and overall impressive student bodies.

It’s just that the top schools like Yale, Hopkins, and Stanford have the luxury of attracting students both with high stats and impressive experiences. These schools routinely reject high stats applicants every year.
Click to expand...

UCLA has the same luxury and rejects plenty of the same types of students as the institutions listed in your second paragraph. From the looks of it, UCLA is also doing something different than the rest in their admission process.
 
OchemOficionado said:
UCLA has the same luxury and rejects plenty of the same types of students as the institutions listed in your second paragraph. From the looks of it, UCLA is also doing something different than the rest in their admission process.
Click to expand...
I don’t dispute UCLA is able to pick from a pool of highly qualified candidates, but there are certainly degrees of separation. Although UCLA’s average accepted MCAT is 517, their matriculating average is 514 according to MSAR. Yale, on the other hand, has an accepted average of 521 and matriculating average of 520.

I just find issue with this idea that UCLA is somehow more holistic than other top schools because they accept a wider range (read: lower) of scores.
 
So I’m guessing I’m screwed with a 502 for UW as an idaho resident?
 
DBC03 said:
Before this discussion continues, can everyone read up about the differences between medians and means? It’s actually a fairly important difference that is leading to misunderstandings of a lot of these statistics.


Sent from my iPad using SDN mobile
Click to expand...
what are u referring to specifically
 
DBC03 said:
Before this discussion continues, can everyone read up about the differences between medians and means? It’s actually a fairly important difference that is leading to misunderstandings of a lot of these statistics.


Sent from my iPad using SDN mobile
Click to expand...
What misunderstandings? I've been using "average" in the colloquial sense, but obviously we are dealing with medians here, which are less sensitive to outliers and don't rely on a normal distribution to provide an accurate measure of central tendency. Not sure why this is an issue in our discussion.
 
People are throwing around the term average, but a school website may list average MCATs while the MSAR is a median - it has a pretty important impact on listed data.


Sent from my iPhone using SDN mobile
 
DBC03 said:
People are throwing around the term average, but a school website may list average MCATs while the MSAR is a median - it has a pretty important impact on listed data.


Sent from my iPhone using SDN mobile
Click to expand...
It's just a colloquial way to talk. I would sincerely hope everyone participating in this discussion knows the difference between means and medians--it's literally stats 101. I am curious how you think it has led to misunderstandings here though.
 
DBC03 said:
People are throwing around the term average, but a school website may list average MCATs while the MSAR is a median - it has a pretty important impact on listed data.


Sent from my iPhone using SDN mobile
Click to expand...
Pretty sure just about everyone here is talking about medians, and knows that the MSAR lists medians...
 
georgia_boy1 said:
What’s the difference?
Click to expand...
Accepted gives you the median stats for all students accepted into the program, regardless of whether they choose to attend. matriculated gives you the stats of only the entering class.
 
DubbiDoctor said:
Pretty sure just about everyone here is talking about medians, and knows that the MSAR lists medians...
Click to expand...

scienceguy95 said:
It's just a colloquial way to talk. I would sincerely hope everyone participating in this discussion knows the difference between means and medians--it's literally stats 101. I am curious how you think it has led to misunderstandings here though.
Click to expand...

I get that - especially with the MSAR, but schools’ websites list the data however they want, and often list the average, so it adds a layer of confusion. Let’s say you compare MSAR data to what a school states on its website - those often differ because one is a median, other could be an average. Or one is from 2016 cycle, the other is from a more recent one. This is part of why it can be difficult to compare step scores from schools (among other issues with score reporting).

Also, median is so different from mean that colloquial usage of “average” for either is confusing to those of us following the conversation. My assumption is if you say average, you mean the average or mean. And then I assume you don’t get stats.

I also think we need to consider the fact that not all students were applying with the new MCAT, and we don’t have data on the old scores accepted. It’s frustrating - and just one more reason schools’ self-reported data may not match the MSAR.


Sent from my iPhone using SDN mobile
 
Last edited:
DBC03 said:
I get that - especially with the MSAR, but schools list the data however they want, and often list the average, so it adds a layer of confusion. Let’s say you compare MSAR data to what a school states on its website - those often differ because one is a median, other could be an average. Or one is from 2016 cycle, the other is from a more recent one. This is part of why it can be difficult to compare step scores from schools (among other issues with score reporting).

Also, median is so different from mean that colloquial usage of “average” for either is confusing to those of us following the conversation. My assumption is if you say average, you mean the average or mean. And then I assume you don’t get stats.

I also think we need to consider the fact that not all students were applying with the new MCAT, and we don’t have data on the old scores accepted. It’s frustrating - and just one more reason schools’ self-reported data may not match the MSAR.


Sent from my iPhone using SDN mobile
Click to expand...

I think starting this cycle the vast majority probably applied with new mcat. 2 cycles before that were probably more affected by the change
 
Lucca said:
I think starting this cycle the vast majority probably applied with new mcat. 2 cycles before that were probably more affected by the change
Click to expand...
In 2016, 45.9% of the applicants applied with scores from the 1991 version (the old version) of the MCAT® exam only, 20.0% applied with scores from the old exam and the new exam, and 32.4% applied with scores from the new exam only. In 2017, 11.5% of the applicants applied with scores from the old version of the MCAT® exam only, 15.9% applied with scores from the old exam and the new exam, and 71.0% applied with scores from the new exam only.
 
Pooker_ said:
In 2016, 45.9% of the applicants applied with scores from the 1991 version (the old version) of the MCAT exam only, 20.0% applied with scores from the old exam and the new exam, and 32.4% applied with scores from the new exam only. In 2017, 11.5% of the applicants applied with scores from the old version of the MCAT exam only, 15.9% applied with scores from the old exam and the new exam, and 71.0% applied with scores from the new exam only.
Click to expand...

So the new data from the MSAR is from the 2016 application cycle? I always wonder if the year refers to when the application cycle started vs when the applicants matriculated. I know the data isn’t from the current cycle.


Sent from my iPhone using SDN mobile
 
Does anyone have stats on total number of applications or the average number of schools each applicant is applying too? I would think if there was a national increase in the number of schools everyone is applying too it could drive the increased MCAT competitiveness among acceptances, as schools would have access to a greater number of higher stats candidates. Obviously this would then not be reflected in the matriculant data as applicants may only attend one school.
 
theKingLT said:
Does anyone have stats on total number of applications or the average number of schools each applicant is applying too? I would think if there was a national increase in the number of schools everyone is applying too it could drive the increased MCAT competitiveness among acceptances, as schools would have access to a greater number of higher stats candidates. Obviously this would then not be reflected in the matriculant data as applicants may only attend one school.
Click to expand...
16 schools per applicant for the latest cycle: https://www.aamc.org/download/321442/data/factstablea1.pdf
Some applicants apply to only a couple schools - usually state schools - some applicants somehow apply to 50+ schools. Location was important but not essential to me, so I applied to 18 MD and 3 DO schools.
 
So this is the data on Mcat percentiles from may 2016 to april 2017. Does anyone know if the N=64,504 that they cite is the total number of test takers for the year? They can't possibly produce enough 511+ scores to sustain this new national median if that's the case

Edit: it won't let me post the link but you can google it
 
How does the MSAR deal with multiple MCAT attempts? If someone is accepted to a school, are both MCAT attempts included in the total number of MCAT scores considered for each school? I imagine this also might differ from how schools consider multiple attempts when self-reporting scores.



Sent from my iPad using SDN mobile
 
theKingLT said:
So this is the data on Mcat percentiles from may 2016 to april 2017. Does anyone know if the N=64,504 that they cite is the total number of test takers for the year? They can't possibly produce enough 511+ scores to sustain this new national median if that's the case

Edit: it won't let me post the link but you can google it
Click to expand...
People that took the MCAT in the past can apply with a score that is a year old, not every applicant has a score from the previous year.
 
DBC03 said:
How does the MSAR deal with multiple MCAT attempts? If someone is accepted to a school, are both MCAT attempts included in the total number of MCAT scores considered for each school? I imagine this also might differ from how schools consider multiple attempts when self-reporting scores.
Click to expand...
I'd (assume) it would be based on the highest score because that is the score that you got accepted with.
 
Some schools accept based on most recent scores. So it would most likely be that score.
 
Pooker_ said:
People that took the MCAT in the past can apply with a score that is a year old, not every applicant has a score from the previous year.
Click to expand...
This is an important factor for considering any individual year, but for it to be sustainable over time you would need however many applicants you take from the previous year to also not get accepted from the current year's test takers, so that they can go on to supply the next year. Assuming a steady 21,000 matriculants per year, the mcat would need to produce an average of 10,500 people scoring 511+ every year (which according to last years data is pretty accurate if not a little high), but then every single one of these people would have to end up with an acceptance either that year or in future years. I'm not sure what the stats are for 511+ scorers eventually getting an acceptance upon reapplication, but I doubt it would be near 100%. Unless I'm missing something big, this would indicate that schools have overdrawn from the pool of current and past applicants with 511+ scores to cause the spike this year. Therefore they will not have enough to maintain this median in future years. Then again, all of this is assuming that Mcat percentiles, the total number of test takers, and the number of yearly matriculants all remain constant.
 
theKingLT said:
This is an important factor for considering any individual year, but for it to be sustainable over time you would need however many applicants you take from the previous year to also not get accepted from the current year's test takers, so that they can go on to supply the next year. Assuming a steady 21,000 matriculants per year, the mcat would need to produce an average of 10,500 people scoring 511+ every year (which according to last years data is pretty accurate if not a little high), but then every single one of these people would have to end up with an acceptance either that year or in future years. I'm not sure what the stats are for 511+ scorers eventually getting an acceptance upon reapplication, but I doubt it would be near 100%. Unless I'm missing something big, this would indicate that schools have overdrawn from the pool of current and past applicants with 511+ scores to cause the spike this year. Therefore they will not have enough to maintain this median in future years. Then again, all of this is assuming that Mcat percentiles, the total number of test takers, and the number of yearly matriculants all remain constant.
Click to expand...
I could be wrong, but I was under the impression roughly 100k people take the MCAT each year, combining that with past years, I don't think it would be that hard to find enough 511+ people.
 
Pooker_ said:
I could be wrong, but I was under the impression roughly 100k people take the MCAT each year, combining that with past years, I don't think it would be that hard to find enough 511+ people.
Click to expand...
That would be the something big I'm missing then :laugh:. I was running under the impression that only 64,500 people took it last year which is the sample size from the percentile data AAMC released for May 2016 through April 2017. Although now that I think about it, that may have just been the sample size for assessing the percentiles, not the total tests or testers. If total test takers is 100,000 than your right there would definitely be enough 511+
 
theKingLT said:
That would be the something big I'm missing then :laugh:. I was running under the impression that only 64,500 people took it last year which is the sample size from the percentile data AAMC released for May 2016 through April 2017. Although now that I think about it, that may have just been the sample size for assessing the percentiles, not the total tests or testers. If total test takers is 100,000 than your right there would definitely be enough 511+
Click to expand...
Regardless of if its 65k or 100k, it doesn't just have to be 511+ because there will be a 10th, 25th, etc. of people that get in, which may drop down to 505 or whatever (we don't know). So if 511 if the 50th then only 50% of acceptances have to be over 511. Pair that with people from past years that didn't get in, took a gap year, etc. It clearly isn't an issue and I would not be surprised if that stats just went up more.
 
Pooker_ said:
Regardless of if its 65k or 100k, it doesn't just have to be 511+ because there will be a 10th, 25th, etc. of people that get in, which may drop down to 505 or whatever (we don't know). So if 511 if the 50th then only 50% of acceptances have to be over 511. Pair that with people from past years that didn't get in, took a gap year, etc. It clearly isn't an issue and I would not be surprised if that stats just went up more.
Click to expand...
I agree 100k would not be an issue but I still think 65k would be an issue. a 511 median means that out of the 21,000 people who will be matriculating come fall, 10,500 scored a 511+. Let's say 3500 511+ applicants were left over from the previous cycle and all reapplied and all got accepted. The MCAT then produces another 10,500 511+ scores (which would be a high estimate if only 65k people took it a year). 7000 of these would matriculate with the previous year's 3500 to meet the quota for a 511 median. This would leave 3500 511+ applicants left for next year's cycle. The problem is, if the numbers continue like this and only 10,500 511+ scores are produced a year, every single one of these applicants would have to be accepted to keep the median where it is. This obviously isn't going to happen, so either way more than 65K take the exam a year, or the median will have to drop as the 511+ pool is exhausted. I'm more inclined to believe that your 100k a year estimate is more accurate, or even 80k, but I either failed google class or the true numbers are hidden somewhere because I can't seem to find an estimate on total test takers per year.
 
theKingLT said:
I agree 100k would not be an issue but I still think 65k would be an issue. a 511 median means that out of the 21,000 people who will be matriculating come fall, 10,500 scored a 511+. Let's say 3500 511+ applicants were left over from the previous cycle and all reapplied and all got accepted. The MCAT then produces another 10,500 511+ scores (which would be a high estimate if only 65k people took it a year). 7000 of these would matriculate with the previous year's 3500 to meet the quota for a 511 median. This would leave 3500 511+ applicants left for next year's cycle. The problem is, if the numbers continue like this and only 10,500 511+ scores are produced a year, every single one of these applicants would have to be accepted to keep the median where it is. This obviously isn't going to happen, so either way more than 65K take the exam a year, or the median will have to drop as the 511+ pool is exhausted. I'm more inclined to believe that your 100k a year estimate is more accurate, or even 80k, but I either failed google class or the true numbers are hidden somewhere because I can't seem to find an estimate on total test takers per year.
Click to expand...
The numbers for how many people took the MCAT in recent years are difficult to find, but I believe it's close to 100k per year - or at least 100k test administrations per year. In 2013, around 95k MCATs were administered (https://aamc-orange.global.ssl.fast...dcc3-49d4-bbde-2ded03bbadd4/combined13pdf.pdf). In 2015, around 60k new MCATs were administered (https://www.aamc.org/download/454206/data/mcatatglance2015.pdf), but between 2015-16 150k new MCATs were administered (https://aamc-orange.global.ssl.fast...ction_score_percentile_ranks_2017_for_web.pdf). So in 2016 it seems that 150k - 60k = 90k exams were administered. I suspect around 100k MCATs were administered in 2017, though I am not sure. AAMC will release more data in May.
 
Wow. I got to make sure I have to apply to many osteopathic schools as well this cycle. My home state (New York), all the in state MD schools have a 514 median.
 
The Ilusive Man said:
Wow. I got to make sure I have to apply to many osteopathic schools as well this cycle. My home state (New York), all the in state MD schools have a 514 median.
Click to expand...
For accepted students, Upstate has a 513 median, buffalo has a 512 median
 
You must log in or register to reply here.

Similar threads

Mr.Smile12
2025-2026 AMCAS MSAR Reports / Discussions
Replies
0
Views
1K
Mr.Smile12
Mr.Smile12
W
MSAR Letter of Intent/Interest Description
Replies
2
Views
941
Waitlistno2
W
Mayajae
  • Locked
Why are schools not accepting urm's in the 2024 msar data?
Replies
24
Views
2K
wysdoc
wysdoc
S
MSAR Waitlist Data
Replies
6
Views
3K
gyngyn
gyngyn
0
How Accurate is MSAR for Coursework Requirements?
Replies
9
Views
1K
Med Ed
Med Ed
Top