What's Wrong Here: 97.5% get into Med School

[tryingagain]

I am pasting this from another post because it has me very confused and I was wanting some help clearing it up. It is posted from the "Raw Numbers" post.

*
*
*
*

If the number of applicants was 34,859 and 17,000 were accepted I have some questions:

1. How many people of the 34,859 applicants receive interviews?

2. If roughly fifty percent of the applicants applied receive interviews (I am just using a "guesstimate" here) that would total 17429.5 or about 17430 people. That means that if this were the case then just about everyone interviewed would get accepted somewhere (97.5%). However, I am sure this cannot be the case and I am positive there are problems with my logic. Can you provide insight?

 

[MD2b06]

Originally posted by tryingagain
2. If roughly fifty percent of the applicants applied receive interviews (I am just using a "guesstimate" here) that would total 17429.5 or about 17430 people. That means that if this were the case then just about everyone interviewed would get accepted somewhere (97.5%). However, I am sure this cannot be the case and I am positive there are problems with my logic. Can you provide insight?

Your premise that only 50% of all applicants receive interviews is false. Therefore, your conclusion that the acceptance rate is 97.5% is erroneous as well.

 

[tryingagain]

So how many of the 34000 do receive interviews?

 

[SMW]

Varies widely from school to school. For example, Columbia interviews almost 50%, whereas GWU interviews only about 9%.

Of course, there's the question of how many applicants get at least one interview somewhere. I really don't know how we're going to come up with the answer to that.

 

[moo]

If 50% of applicants receive interviews, it is entirely conceivable that the post interview acceptance rate is much less than 97.5% for EACH INDIVIDUAL SCHOOL. Let's say 17000 people receive at least one interview. However, most of these people receive MORE than one interview. So they can get accepted at several schools, withdraw, and then have other people interview at one or two schools, get waitlisted and never get off. Some get accepted at one or two schools and get waitlisted at four or five, so your post-int acceptance rates are not going to be that high for each particular school, but people end up matriculating anyway to A medical school because it just takes one acceptance.

 

[MD2b06]

Originally posted by tryingagain
So how many of the 34000 do receive interviews?

There are no published statistics that reveal exactly how many applicants receive interviews. The problem with looking at the number of applicants each school interviews is that there is a tremendous overlap between schools. It's not uncommon for many highly qualified applicants to interview at many of the same places.

 

[tryingagain]

I just wonder if we did know that answer what the percentage would be. I find this very interesting. I wish we had more data.

 

[tBw]

I guess I just don't care. At the end of the day I am not a statistic. It doesn't matter if 97.5% or 3.1% of interviewees are accepted. I am only accepted or not. I cannot be 97.5% accepted. So while I understand the drive to put some 'meaning' or order into this craziness, my perspective is that we each get discrete results, and the 'averages' therefore show us nothing about our own futures.

...are you a physicist??

 

[tryingagain]

What in the hell are you talking about? You ARE a statistic.

The averages show us absolutely everything about our futures.
Sorry if I offended you but I am very interested in the numbers. This whole thing IS a numbers game and a crapshoot in a lot of ways. Numbers are important whether we like it or not.

 

[tBw]

Originally posted by tryingagain
What in the hell are you talking about? You ARE a statistic. ...Sorry if I offended you

why the "hell" would I be offended?

Originally posted by tryingagain

The averages show us absolutely everything about our futures.
[/B]

hmm, so you are a physicist....

If looking at the numbers makes you feel better, go for it. My point was that this is really just a form of obsessing and at the end of the day it's not going to tell you whether you will get into medical school or not as statistics only tell you about trends, not individual stories or futures.

Take a room of 10 people. One person in the room is the stereotypical ideal candidate. 4.0 GPA 43T MCAT, first author published research, good clinical exposure, outstanding personal essay, good interviewing skills, father is dean of the medical school. The other 9 people in our room have a GPA of 2.50 and an MCAT of 22, no clinical exposure, no research, and their parents are not in medicine. We can predict that one of these people will be accepted to medical school. Does that mean each person in this room has an equal 10% chance? No. If 10 interviews are offered, does it mean each one will get 1 interview? No - more than likely they will all go to that one person. Thus statistics can tell us many things, but it certainly can't replace common sense. In this analysis we make the error of applying statistics to a non-homogeneous dataset. I believe your analysis does the same.

 

[relatively prime]

Another error is that it's not about getting ~one~ interview... it's about how many interviews you get. If a student only gets one interview, there's still a pretty good chance he/she won't get in to medical school (unless it's their low-tier state school or soemthing). If she/he gets 5-6 interviews, it's almost certain that he/she will get into at least one of those schools.

 

[SMW]

Originally posted by relatively prime
Another error is that it's not about getting ~one~ interview... it's about how many interviews you get. If a student only gets one interview, there's still a pretty good chance he/she won't get in to medical school (unless it's their low-tier state school or soemthing). If she/he gets 5-6 interviews, it's almost certain that he/she will get into at least one of those schools.

Good point. There was a thread from last year's application cycle that tried to correlate numbers of secondaries and interviews with acceptances. Unless you were a very stellar applicant, or a very poor one, the more secondaries you returned and interviews invites you received, the more likely you were to get at least one acceptance.

 

[tryingagain]

not according to wonder boy because 5-6 interviews does not tell the whole story or provide common sense.

i would agree with you that common sense would tell you that after 5-6 interviews the statistical odds are in your favor and thus you would probably get into one of those schools

Its "the whole story" vs. "the statistics"

 

[relatively prime]

Originally posted by tryingagain
not according to wonder boy because 5-6 interviews does not tell the whole story or provide common sense.

i would agree with you that common sense would tell you that after 5-6 interviews the statistical odds are in your favor and thus you would probably get into one of those schools

Its "the whole story" vs. "the statistics"

Yes, but we're talking about statistics here. When you trying to see trends, you can't focus on these "special cases." If you have 6 interviews, each school accepting %50 of those interviews... statistically speaking there's a 95+% you'll get into at least one of those schools. Now that still leaves ~5% chance that you still might not get in for whatever reason.

If you want to talk about statistics... you can't get hung up on that one guy who wore a lime green, pink-poka-dotted suit to all his interviews. You have to focus on the general case.

 

[Jeff698]

As much as I hate to break the time-honored SDN rule of not using real data, here are some interesting numbers from this weeks JAMA. It doesn't directly answer your question, but I suspect you'll be interested anyway (I was).

This is from the 2001-2002 entering class (mine...yeah!):

No. of applicants: 34,859
Total No. of applications 403,609
Avg No. of applications/person: 11.6
No. accepted applicants: 17,456
Applicant/acceptance ratio: 2.0
First-year enrollment: 16,933

The thing that might be encouraging for y'all is the applicant/acceptance ratio. For the class of 2005, there was a 2:1 ratio. In other words, a 50-50 chance, on average.

Here is some other interesting information, especially for those of us who are getting sick of hearing how today's medical students just aren't as bright as those geniuses from yesteryear:

The ratio from 20 years ago was 2.1. This value appears to be cyclic. Over the period reported (1981 - 2001), there was a high of 2.7 (1995-96 & 1996-97) and a low of 1.7 (1990).

While there was a 6.0% decrease in applications from 2000 and a 9.5% decrease from 1999 (something that has been made WAY too much of, IMHO), this also appears to be cyclic. Again, from the same time period, there was a low number of applicants in 1990-91 (29,243) and a high in 1996-97 (46,968).

Reference:
Barzansky, B, Etzel, SI. Educational Programs in US Medical Schools, 2001-2002JAMA. 2002;288:1067-1072

Unfortunately, this article is only available online for subscribers. There is, however, a nice version of msJAMA, the version for medical students (available at http://www.ama-assn.org/sci-pubs/msjama/index.htm) that has absolutely nothing to do with this subject.

Take care,
Jeff

 

[relatively prime]

Originally posted by Jeff698
As much as I hate to break the time-honored SDN rule of not using real data, here are some interesting numbers from this weeks JAMA. It doesn't directly answer your question, but I suspect you'll be interested anyway (I was).

This is from the 2001-2002 entering class (mine...yeah!):

No. of applicants: 34,859
Total No. of applications 403,609
Avg No. of applications/person: 11.6
No. accepted applicants: 17,456
Applicant/acceptance ratio: 2.0
First-year enrollment: 16,933

The thing that might be encouraging for y'all is the applicant/acceptance ratio. For the class of 2005, there was a 2:1 ratio. In other words, a 50-50 chance, on average.

As much as I hate to break the time-honored SDN rule of not using logic.... a 2:1 applicant/aceptance ratio does not translate into a 50-50 chance of getting in. As we all know a significant enough number of people get more than one acceptance.

 

[owen_osh]

Originally posted by relatively prime
As much as I hate to break the time-honored SDN rule of not using logic.... a 2:1 applicant/aceptance ratio does not translate into a 50-50 chance of getting in. As we all know a significant enough number of people get more than one acceptance.

he should have written (from the article): "the ratio of applicants to accepted applicants was 2.1:1.0 "

 

[relatively prime]

Originally posted by owen_osh
he should have written (from the article): "the ratio of applicants to accepted applicants was 2.1:1.0 "

Oh... if that's what he meant... yeah, then obviously there is a 50-50 chance (statistically).

 

[Random Access]

Originally posted by tryingagain
What in the hell are you talking about? You ARE a statistic.

The averages show us absolutely everything about our futures.
Sorry if I offended you but I am very interested in the numbers. This whole thing IS a numbers game and a crapshoot in a lot of ways. Numbers are important whether we like it or not.

Hey, did you guys know that 82.5% of all statistics are made up?


-RA

 

[Jeff698]

Originally posted by relatively prime
Oh... if that's what he meant... yeah, then obviously there is a 50-50 chance (statistically).

Yep, that's what I meant. Owen appropriately quoted the article where I got in a hurry and retyped data from a table.

Take care,
Jeff