AAMC facts and std devs

[Phenol312]

---

Members do not see this ad.

so i was messing around with some numbers ( because im ) from this page http://www.aamc.org/data/facts/2007/2007mcatgpa.htm

and i found that in 2007 the average MCAT for an accepted student was 30.8 with a standard deviation of +/-5.4. Now this would mean that 68% (roughly 2/3) of accepted students had MCATS in the range of 26-36. 33% had numbers outside that range. Now, if they had only accepted those who had scores on the positive side (31-36) then that would shift the overall average up. So that being said, if my thinking is right here then that would mean that competitiveness isn't increasing in quality its only increasing in quantity.

and for BCPM: 68% of accepted applicants were in the range of 3.26 - 3.92
and for AO: " " 3.48 - 3.98

and for overall: " " 3.39 - 3.91

am i missing something here or is there hope for those below 3.5/33 still?

 

[Textuality]

Bell curve is bell curve. So yeah, I think 16% were below that 26 then. My only issue is whether it's a bell curve, lol.

 

[Phenol312]

Unless they're using some kind of high level population statistics I haven't been exposed to ( or cant remember) then i would assume that it would have to be a bell curve.

I think this goes back to how only the most neurotic those with the highest stats post on SDN...the average and lower portion dont even know about the site. They probably make up the greater numbers that explain this data trend...if my interpretation of it is correct of course.

 

[rds726]

so i was messing around with some numbers ( because im ) from this page http://www.aamc.org/data/facts/2007/2007mcatgpa.htm

and i found that in 2007 the average MCAT for an accepted student was 30.8 with a standard deviation of +/-5.4. Now this would mean that 68% (roughly 2/3) of accepted students had MCATS in the range of 26-36. 33% had numbers outside that range. Now, if they had only accepted those who had scores on the positive side (31-36) then that would shift the overall average up. So that being said, if my thinking is right here then that would mean that competitiveness isn't increasing in quality its only increasing in quantity.

and for BCPM: 68% of accepted applicants were in the range of 3.26 - 3.92
and for AO: " " 3.48 - 3.98

and for overall: " " 3.39 - 3.91

am i missing something here or is there hope for those below 3.5/33 still?

you have a lot of free time. hahaha. im just playing. you seem to be on all the threads im at.

 

[Phenol312]

lol yeah seems that way...its a slow day and im feelign especially neurotic...I'm about to venture out to the mailbox to see if any schools have decided to say anything to me.

 

[LizzyM]

You can't sum standard deviations!

For each variable the standard deviation is less than 2. I'd wager that if you were to take the mean of the sum of the three MCAT scores you'd get an even smaller standard deviation as the above the mean in one/below in another would average out.

Furthermore, you assume that the data are a bell curve with a normal distribution but, in fact, the data may be skewed.

 

[TexasTriathlete]

Haha you said "std"

 

[Phenol312]

You can't sum standard deviations!

For each variable the standard deviation is less than 2. I'd wager that if you were to take the mean of the sum of the three MCAT scores you'd get an even smaller standard deviation as the above the mean in one/below in another would average out.

Furthermore, you assume that the data are a bell curve with a normal distribution but, in fact, the data may be skewed.

haha...thanks for catching that. It's been 2 years or so since my stat class my prof would be pissed if she caught me doing that. They didnt give the SD for the overall score though so i took some liberties. I was more interested in the GPA ranges though.

Also, instead of rifling through my closet for my stat book ill just ask here ... wouldn't having a skewed distribution alter the mean and that in turn would alter the Std Dev?

 

[Phenol312]

Haha you said "std"

lol yeah ... awesome.

 

[Textuality]

I wouldn't give up hope 🙂 First off, it's only the end of November, this marathon runs until freaking MAY. (some schools interview until April..crazy right?)

Secondly, I think if you get some EC's your second go around will be fine 🙂

 

[rds726]

lol yeah seems that way...its a slow day and im feelign especially neurotic...I'm about to venture out to the mailbox to see if any schools have decided to say anything to me.

was there anything in the mailbox?
waiting sucks...

 

[Phenol312]

just bills and a thank you card from the neighbor.

I keep having to stop myself from checking status pages again...I've been checking on a daily basis since i got the hold from Drexel.

 

[LizzyM]

haha...thanks for catching that. It's been 2 years or so since my stat class my prof would be pissed if she caught me doing that. They didnt give the SD for the overall score though so i took some liberties. I was more interested in the GPA ranges though.

Also, instead of rifling through my closet for my stat book ill just ask here ... wouldn't having a skewed distribution alter the mean and that in turn would alter the Std Dev?

If there is a long right tail, or outliers at the upper range, then the mean is going to be higher than the median (the midpoint of the range, with half of the matriculants being at or above that score and half being at or below that score) and the standard deviation is going to be larger than it would be if the scores were normally distributed.

Now, we know that about 5% of the applicants score at or above 37. Let's presume that 90% of them get admitted somewhere. So, at least 4.5% of the matriculants are more than 2 standard deviations above the mean although in a normal distrbution ("bell curve") we'd expect only 2.5% to be more than 2 standard deviations above the mean.

So, we might assume that there is a long right tail, that the mean is higher than the median, and that the standard deviation is greater than it would be if the scores were normally distributed.

Although we are all guilty of summing MCAT scores, they should be examined individually and are by adcoms: a 5 14 14 is not the equivalent of a 9 12 12 nor a 13 10 10.

 

[batman1983]

Man, I thought this thread was about something else......

 

[Phenol312]

If there is a long right tail, or outliers at the upper range, then the mean is going to be higher than the median (the midpoint of the range, with half of the matriculants being at or above that score and half being at or below that score) and the standard deviation is going to be larger than it would be if the scores were normally distributed.

Now, we know that about 5% of the applicants score at or above 37. Let's presume that 90% of them get admitted somewhere. So, at least 4.5% of the matriculants are more than 2 standard deviations above the mean although in a normal distrbution ("bell curve") we'd expect only 2.5% to be more than 2 standard deviations above the mean.

So, we might assume that there is a long right tail, that the mean is higher than the median, and that the standard deviation is greater than it would be if the scores were normally distributed.

Although we are all guilty of summing MCAT scores, they should be examined individually and are by adcoms: a 5 14 14 is not the equivalent of a 9 12 12 nor a 13 10 10.

My stat book agrees with you 😛. Thank you for working through this with me. I guess I'm guilty of just trying to put a positive spin on my numbers...I'm not really concerned about the MCAT more so my GPA 🙄 and it being an obstacle despite the absurd performance jump (overall 3.0 fresh/soph year to 3.7 junior/senior year). And junior senior year is when i took the high level science stuff too!

 

[Phenol312]

Man, I thought this thread was about something else......

??? such as?

 

[The Nigal]

The MSAR has graphs of all the GPAs and MCAT scores for accepted students. Just from glancing at the charts; it looks like MCAT scores are more or less on a bell curve, while GPAs definitely show a steadily rising slope.

I think mean GPAs are typically somewhat higher than average GPAs. While individual MCAT score are fairly consistent between the mean and the average (maybe skewed a little to the right?).

BTW the writing scores are all over the place, but I don't think anyone really cares.

 

[ChubbyChaser]

Yeah the GPA was definitely skewed rightward, but I also think the overall MCAT was skewed to the right just not as drastic as the GPA. By the looks of it, the number of matriculants dropped off much faster from 28-26 than from 32-35. I also wonder if a good number of those with low MCATs, like sub 24, are in BS/MD programs where you arent required to make a certain score,and you just have to take it??

 

[Phenol312]

How much of a rightward skew are we seeing on these BCPM/AO/overall GPA numbers in the past 2 cycles? Anyone?

 

[Textuality]

I'm also amused that the writing score hasn't really shown in increase compared to 1996, but everything else has, hehe.

Yeah, I waiver between feeling confident about my MCAT score to feeling worried about it due to a lot of people I know having like 35+ scores, and I'm one point below the median scores of most schools I applied to. I know that doesn't really mean anything, but bah....when you have no acceptances in hand, you pretty much worry over anything and everything, or at least I do 😛

 

[Phenol312]

....when you have no acceptances in hand, you pretty much worry over anything and everything, or at least I do 😛

Join the club. I'm getting less confident day by day lol

My MCAT and AO GPA are right on the mean...the only thing hurting me is my science gpa being like .07 outside 1 SD which brings my overall GPA .02 outside 1 SD

 

[Textuality]

Haha, yeah...when I started this process I was feeling pretty good. And now I'm trying to remind myself why I'm not hopeless before I go to my next interview! So I think in the end, I'm neither cocky about my chances, nor despairing. I'm just confused.

For example, I got interviews to UPenn and Hopkins, but no love from GW, Drexel, or Jefferson 🙁 And I have stopped trying to reason out why, I'm just going to be gratefully happy that I'm somehow lucky enough to get the interviews I have.

 

[Phenol312]

Wow congrats on those interviews...I'm still waiting to have my interview cherry popped lol

 

[batman1983]

Haha, yeah...when I started this process I was feeling pretty good. And now I'm trying to remind myself why I'm not hopeless before I go to my next interview! So I think in the end, I'm neither cocky about my chances, nor despairing. I'm just confused.

For example, I got interviews to UPenn and Hopkins, but no love from GW, Drexel, or Jefferson 🙁 And I have stopped trying to reason out why, I'm just going to be gratefully happy that I'm somehow lucky enough to get the interviews I have.

The fact that you got interviews at UPenn and Hopkins implies you will get in. As for GW, I have heard that they dont give interviews to over-qualifed applicants.

 

[LizzyM]

For example, I got interviews to UPenn and Hopkins, but no love from GW, Drexel, or Jefferson 🙁 And I have stopped trying to reason out why, I'm just going to be gratefully happy that I'm somehow lucky enough to get the interviews I have.

I think that some schools will not waste an interview spot on someone who appears to have chosen them as a "back-up". Really, given a choice between UPenn and Drexel which ill get the nod? Ditto a choice between Hopkins and GW? (thus removing any geographic preference) Schools may not know where you've applied but they know that they shouldn't aim too high (if they are a lower ranked or unranked school) or they'll be left with no one accepting their offers of admission.

 

[Textuality]

This is probably true, thanks for the insight. Hmm..I guess it's really hard to try to play things safe then huh? You just have to apply to a lot of schools that honestly fit your numbers, and hope for a little luck here and there 🙂

Hehe, sorry to have jacked this thread a little! It's just a little worrisome that my advisers in college kept stressing the importance of "back up " schools (as unsavory as I find that term), and now I find that I don't really have any back-up schools...I just have a lot of schools, lol.

But good luck to Phenol!! I really hope you make it this round, your heart really seems to be into it 🙂

 

[Phenol312]

thanks ... i dont mind the jack lol

I'm trying to stay positive but like i posted in another thread each day is like water torture to my confidence...drip...drip...drip.

I would love to be in the position to choose safety schools...hell I'd go to school in the middle of the arctic if they'd accept me.

 

[Kikaku21]

You can't sum standard deviations!

For each variable the standard deviation is less than 2. I'd wager that if you were to take the mean of the sum of the three MCAT scores you'd get an even smaller standard deviation as the above the mean in one/below in another would average out.

Furthermore, you assume that the data are a bell curve with a normal distribution but, in fact, the data may be skewed.

They are absolutely skewed. To assume normality would imply that there is no self selection. We see self selection on SDN everytime someone holds off on applying to retake b/c they have a low MCAT.

 

[Textuality]

thanks ... i dont mind the jack lol

I'm trying to stay positive but like i posted in another thread each day is like water torture to my confidence...drip...drip...drip.

I would love to be in the position to choose safety schools...hell I'd go to school in the middle of the arctic if they'd accept me.

"Just keep swimming....just keep swimming" 🙂

 

[Phenol312]

"Just keep swimming....just keep swimming" 🙂

damn im gonna have to go watch that movie now...at least itll get my mind off things 🙂

 

[Textuality]

Oh, if you've never seen Lilo and Stitch, you should see it. One of Disney's BEST.

 

[ChubbyChaser]

Oh, if you've never seen Lilo and Stitch, you should see it. One of Disney's BEST.

Nope, Fox and the Hound is Disney's best, with Fern Gulley a close second.😀

 

[Textuality]

Nope, Fox and the Hound is Disney's best, with Fern Gulley a close second.😀

I just said ONE of disney's best 😛 Man, I don't think I've seen Fox and the Hound since I was 5...haha. Now I want to get some people over to have a Disney movie marathon! And let's not forget the Incredibles!

 

[ChubbyChaser]

Haha, someone should make a poll with the different disney movies. I mean there was a peanut butter poll, so come on!!

 

[Phenol312]

Oh, if you've never seen Lilo and Stitch, you should see it. One of Disney's BEST.

downloading off mininova as i type lol

 

[Phenol312]

Haha, someone should make a poll with the different disney movies. I mean there was a peanut butter poll, so come on!!

DO IT!!!

make one for the CGI films and another for the classics

I was always a big peter pan fan myself

 

[ChubbyChaser]

done, and since I dont know what the difference bw a CGI film and a classic is, I just stuck to the cartoons lol

 

[Textuality]

Haha, you totally beat me to it by like 2 minutes since I had to go drop by my boss' office.. Aw...I was looking forward to making that thread too! hehe, no problem, I'll just vote in it!!

And yeah, tell me what you think of Lilo and Stitch after you watch it.

PS..sorry I jacked the thread again 😛

 

[LikeClockWork]

...I'm still waiting to have my interview cherry popped lol

Don't worry, it's not as bad as everyone says. As long as you ease into it and it's with a school you love, you'll be fine. 😉👍

 

[batman1983]

Don't worry, it's not as bad as everyone says. As long as you ease into it and it's with a school you love, you'll be fine. 😉👍

I love sexual in-ur-endo

 

[drizzt3117]

The other issue, which no one has alluded to in this particular thread, but others have mentioned in other threads of this type, is that the statistics don't really show which scores are tied to which applicants.

I'm also not sure how carefully adcoms will compare applicant statistics to the means. Is there median data for matriculants from any source?

 

[Tropicana100]

I love sexual in-ur-endo

laughing my ass off

 

[LizzyM]

The other issue, which no one has alluded to in this particular thread, but others have mentioned in other threads of this type, is that the statistics don't really show which scores are tied to which applicants.

I'm also not sure how carefully adcoms will compare applicant statistics to the means. Is there median data for matriculants from any source?

I'm not sure I understand the question. Every school knows its own median and/or mean for each variable. Applicants are compared to the current student body... is this applicant a good fit? might this one be at risk for academic failure at our school given a gpa of x and a MCAT of y (putting this applicant in the bottom 1% of all matriculants to our school).

An adcom gets a feel for what it tends to see... if the average for the students at a med school is 10 11 11 then it has a feeling that a 9 12 11 is good and a 9 10 10 is below average and 12 12 15 is wonderful and we just hope that the applicant loves us. I think that adcoms seldom want to go more than 2 points below the average on any one measure and they usually like to see an above average score in one of the other components.

With the gpa, one really has to look at the 30 or more courses that make up that gpa... is a 3.33 the result of a student who had B+ in every subject with a single B and a single A- or is it a case of someone who had a horrible year due to a catastrophic event followed (and sometimes preceeded) by very good performance, or is it someone who had B- and B in the pre-reqs but boosted the gpa with many 4.0 credits in a soft subject (credits for chorus & orchestra, and phys ed. come to mind).

 

[Phenol312]

Thanks Lizzy

I feel a little more confident after reading your post. I can only hope the schools will look favorably on my MCAT and strong trend of improvement.

 

[drizzt3117]

I'm not sure I understand the question. Every school knows its own median and/or mean for each variable. Applicants are compared to the current student body... is this applicant a good fit? might this one be at risk for academic failure at our school given a gpa of x and a MCAT of y (putting this applicant in the bottom 1% of all matriculants to our school).

I guess my question has several different components now that I think about it. First of all, I didn't mention this but some statements in this thread raise a significant question. It seems likely that there's some sort of quantitative screening done prior to qualitative assessment of applications. Is this strictly a distance from the mean/median evaluation for each individual component, with point values assigned accordingly or is it somewhat qualitative?

After that is done, from the discussions I've had with adcom members etc different schools tend to have different methods of such evaluations. One simple (and probably common) method I've heard of assigns different point values to quantitative and qualitative components, others are purely qualitative evaluations followed by voting.

As far as qualitative assessment of grades, is the entirety of this done pre-interview with future discussions looking at the applicant as a whole?

Your final point brings up another interesting question. I am curious about the evaluation of non-science grades, especially in the context of non-science majors, as my academic performance during my actual undergrad was pretty much in line with the second scenario you outlined, B+/ performance in most classes, with the occasional A/A- simply because in my field grades were far less important than significant work experience and standardized test scores. I haven't had a single question in any of my open-file interviews about my undergrad grades, I felt this was probably because I have strong postbac/upper division grades and MCAT as well as b/c these grades are a number of years removed from the present but was curious if this particular factor was considered and dismissed prior to an interview invite or if it's something that may be considered at a later date.

I suppose I diverged a bit from my original point but those were some thoughts that came to my mind.

 

[LizzyM]

I guess my question has several different components now that I think about it. First of all, I didn't mention this but some statements in this thread raise a significant question. It seems likely that there's some sort of quantitative screening done prior to qualitative assessment of applications. Is this strictly a distance from the mean/median evaluation for each individual component, with point values assigned accordingly or is it somewhat qualitative?

After that is done, from the discussions I've had with adcom members etc different schools tend to have different methods of such evaluations. One simple (and probably common) method I've heard of assigns different point values to quantitative and qualitative components, others are purely qualitative evaluations followed by voting.

As far as qualitative assessment of grades, is the entirety of this done pre-interview with future discussions looking at the applicant as a whole?

Your final point brings up another interesting question. I am curious about the evaluation of non-science grades, especially in the context of non-science majors, as my academic performance during my actual undergrad was pretty much in line with the second scenario you outlined, B+/ performance in most classes, with the occasional A/A- simply because in my field grades were far less important than significant work experience and standardized test scores. I haven't had a single question in any of my open-file interviews about my undergrad grades, I felt this was probably because I have strong postbac/upper division grades and MCAT as well as b/c these grades are a number of years removed from the present but was curious if this particular factor was considered and dismissed prior to an interview invite or if it's something that may be considered at a later date.

I suppose I diverged a bit from my original point but those were some thoughts that came to my mind.

Yes, there is usually some quantitative assessment. Do you have a 2.56 and a 21? Most likely, you are in the bottom 1 or 2% at the allopathic med school applicant pool at that school. That might just preclude an further time being spent on the application. If you have a 3.5 and a 30 then you might have a good shot of a deeper look into your application. The question comes when you have someone with a 2.56 and a 30 or 3.50 and 21. some schools may have a formula for looking at those, others may leave it up to indivdual readers who may look for a reason to recommend an intrview or to set the application aside.

Generally, an applicant should have the "paper" credentials to be suitable for admission before the interview is granted. So, the post-interview assessment becomes a question of "is he as good in person as he is on paper". The difficulty comes in deciding if the person with the academic credential that were outstanding but who was not very impressive in person is more deserving of admission than the "average on paper" applicant who impressed the interviewers in a major way.

 

[0Complications]

I think that some schools will not waste an interview spot on someone who appears to have chosen them as a "back-up". Really, given a choice between UPenn and Drexel which ill get the nod? Ditto a choice between Hopkins and GW? (thus removing any geographic preference) Schools may not know where you've applied but they know that they shouldn't aim too high (if they are a lower ranked or unranked school) or they'll be left with no one accepting their offers of admission.

Wow! This is a great point. I had never thought about it, but it makes perfect sense.

It seems that an applicant has to base a fall-back school not only on the schools requirements, but also on how much they exceed said schools requirements. Maybe they should start calling them "lean-back" schools or "fall-just-a-smidge-back" schools...

I'll be sure to consider this next year.

 

[drizzt3117]

I guess I am more curious about whether the decision process pre interview is more commonly done by a point-based system where candidates are evaluated on their individual criteria and assigned points and passing a cestain threshold gets you an invite, or a system that is more subjective/holistic. I know both systems are used, but I am curious what ppl think.

 

[LizzyM]

I guess I am more curious about whether the decision process pre interview is more commonly done by a point-based system where candidates are evaluated on their individual criteria and assigned points and passing a cestain threshold gets you an invite, or a system that is more subjective/holistic. I know both systems are used, but I am curious what ppl think.

consider the ratio of interview slots to applicants. I imagine that it varies by school. At the private schools that don't have an in-state/out of state issue to deal with, their may be 5, 6 or 7 applicants for each interview slot. Obviously the process of choosing the lucky few is going to be different than a state school that has a relatively small pool of in-state applicants plus the OOS applicants and the method for choosing applicants for interview at those schools may be different (threshold for in-state and much tougher criteria for OOS).

 

[drizzt3117]

That makes sense. I have kinda given up on trying to decipher what individual schools are looking for. That said, I do feel for the most part, the adcoms of the schools Ive interviewed at did a good job in selecting ppl, as more often than not it seemed that me and the other interviewees were good fits. With a ton of qualified applicants, the decisions must be tough.