Tippmann Pneumatics Inc. Homepage
  FAQ FAQ  Forum Search   Register Register  Login Login

Racially Driven School Admissions

 Post Reply Post Reply
Author
ArthurBignose View Drop Down
Member
Member
Avatar

Joined: 01 August 2010
Status: Offline
Points: 137
Post Options Post Options   Thanks (0) Thanks(0)   Quote ArthurBignose Quote  Post ReplyReply Direct Link To This Post Topic: Racially Driven School Admissions
    Posted: 31 August 2011 at 5:11pm
So, I'm a senior now (engineering) and I've applied to med school(s).  While looking around, I found a breakdown on percentage acceptance based on MCAT score and GPA.  This was further broken down into ethnicity.  It shows what I think is a somewhat ridiculous trend.  The vertical axis is %chance, the x axis is MCAT score from 25-40 and the lines represent cGPA.

Overall:

http://i55.tinypic.com/v4o2om.jpg


Hispanic or Latino:

http://i53.tinypic.com/2w2lgdi.jpg


Black:

http://i52.tinypic.com/35bw9jc.jpg


Asian:

http://i53.tinypic.com/ra9mp4.jpg


White:

http://i52.tinypic.com/2efkcaf.jpg



I understand the desire to have a diverse population in a profession, and that test scores are only one part of what makes a person appealing to an admissions committee.  However, these graphs seem to show a clear difference in the likelihood of acceptance based on race with whites being about average, hispanics and black above average acceptance, and asians below average acceptance. 

For some numbers, take a 3.5 GPA and 30 MCAT which is about average.

Chances of being accepted based on race:

overall  50%
hispanic 78%
black 88%
asian 45%
white 52%

I could understand a 5 percent difference...but accepting on average 32% more black people than asians with the same test scores?  Come on...



Edited by ArthurBignose - 31 August 2011 at 5:35pm


Back to Top
oldpbnoob View Drop Down
Platinum Member
Platinum Member
Avatar
Not old, Not noob. May be Dave's grandma

Joined: 04 February 2008
Location: Yankee Stadium
Status: Offline
Points: 5676
Post Options Post Options   Thanks (0) Thanks(0)   Quote oldpbnoob Quote  Post ReplyReply Direct Link To This Post Posted: 31 August 2011 at 5:23pm
Out of curiosity, do these figures include all applicants whether foreign or U.S. citizens? This could cause part of the swings considering white Europeans and Chinese may be more apt to apply for schooling in the U.S. than say Nigerians or Chileans. Just a thought.

Edited by oldpbnoob - 31 August 2011 at 5:24pm
"When I grow up I want to marry a rich man and live in a condor next to the beach" -- My 7yr old daughter.
Back to Top
stratoaxe View Drop Down
Platinum Member
Platinum Member
Avatar
And my axe...

Joined: 21 May 2005
Location: United States
Status: Offline
Points: 6831
Post Options Post Options   Thanks (0) Thanks(0)   Quote stratoaxe Quote  Post ReplyReply Direct Link To This Post Posted: 31 August 2011 at 5:28pm
I think that college admission is tricky business. The education game is changing, and setting the bar isn't as easy as it used to be.

I know that there are alot of factors that contribute to admissions that may not be represented here. Like with law school, it's perfectly within reason to get in with a 3.5 or even lower if you have the right writing skills and community service experience.

I don't really know what motivations that colleges have for keeping a diverse population, but I'd imagine it's important.

All of that said, it is disheartening to know that you can work your butt off and still be at a disadvantage. I expect this is a trend that's only going to get worse as the percentages of higher level degrees goes up.
Back to Top
ArthurBignose View Drop Down
Member
Member
Avatar

Joined: 01 August 2010
Status: Offline
Points: 137
Post Options Post Options   Thanks (0) Thanks(0)   Quote ArthurBignose Quote  Post ReplyReply Direct Link To This Post Posted: 31 August 2011 at 5:33pm
The overall includes all applicants regardless of residency status.

The graphs based on ethnicity only include US citizens or permanent residents  (not international applicants).

The sources for the numbers come from the Association of American Medical Colleges:

https://www.aamc.org/download/157450/data/table24-mcatgpagridall2008-10.pdf.pdf

https://www.aamc.org/data/facts/applicantmatriculant/157998/mcat-gpa-grid-by-selected-race-ethnicity.html






Edited by ArthurBignose - 31 August 2011 at 5:53pm


Back to Top
oldpbnoob View Drop Down
Platinum Member
Platinum Member
Avatar
Not old, Not noob. May be Dave's grandma

Joined: 04 February 2008
Location: Yankee Stadium
Status: Offline
Points: 5676
Post Options Post Options   Thanks (0) Thanks(0)   Quote oldpbnoob Quote  Post ReplyReply Direct Link To This Post Posted: 31 August 2011 at 5:41pm
Your second link doesnt seem to work for me, but not a big deal. If true, seems pretty odd. I wonder if anything has to do with sheer numbers? Do the graphs include applicants that may have been accepted to multiple programs and declined to go? Are Asians more apt to send out more applications? I'm not Asian, but I sent out multiple applications to colleges when I was looking around and got accepted to all of the ones I sent in. I obviously only went to one of them.
"When I grow up I want to marry a rich man and live in a condor next to the beach" -- My 7yr old daughter.
Back to Top
ArthurBignose View Drop Down
Member
Member
Avatar

Joined: 01 August 2010
Status: Offline
Points: 137
Post Options Post Options   Thanks (0) Thanks(0)   Quote ArthurBignose Quote  Post ReplyReply Direct Link To This Post Posted: 31 August 2011 at 5:57pm
I fixed the links.

The data shows applicants who applied to at least one school and acceptees who were accepted to at least one school.  Obviously the acceptance rate is lower if for someone who applies to less schools than for someone who applies to more, so that will skew the plots.  However, it should skew them all similarly. 

You could make socioeconomic conjectures into how ethnicity could play into the ability to apply to schools, but there are fee waiver programs so that shouldn't be an issue.   It is super expensive though, I applied to 11 schools and have paid about $1500 so far and am expecting to have shelled out over $2k before it's all said and done.


Edited by ArthurBignose - 31 August 2011 at 5:58pm


Back to Top
agentwhale007 View Drop Down
Platinum Member
Platinum Member
Avatar
Forum's Vladimir Lenin

Joined: 20 June 2002
Location: GNV FLA
Status: Offline
Points: 11696
Post Options Post Options   Thanks (0) Thanks(0)   Quote agentwhale007 Quote  Post ReplyReply Direct Link To This Post Posted: 31 August 2011 at 6:15pm
Are the sample sizes for each ethnicity of similar range?  
"So when Romney wins in a landslide, what will the liberal media do?"
This Ma**edited**hine Kills **edited**as**edited**ists.




Back to Top
ArthurBignose View Drop Down
Member
Member
Avatar

Joined: 01 August 2010
Status: Offline
Points: 137
Post Options Post Options   Thanks (0) Thanks(0)   Quote ArthurBignose Quote  Post ReplyReply Direct Link To This Post Posted: 31 August 2011 at 6:27pm
Originally posted by agentwhale007 agentwhale007 wrote:

Are the sample sizes for each ethnicity of similar range?  


And Whale brings the reality check...

Good point, I didn't even think about it.

Sample sizes:

Overall: 124503
Hispanic: 9287
Black: 10123
Asian: 27187
White: 78024


Back to Top
agentwhale007 View Drop Down
Platinum Member
Platinum Member
Avatar
Forum's Vladimir Lenin

Joined: 20 June 2002
Location: GNV FLA
Status: Offline
Points: 11696
Post Options Post Options   Thanks (0) Thanks(0)   Quote agentwhale007 Quote  Post ReplyReply Direct Link To This Post Posted: 31 August 2011 at 6:37pm
I'd imagine that, to some extent, the fact that the sample size for black applications/acceptances is a considerable amount lower than that for white, and Hispanic is remarkably lower, is causing some of the discrepancy in the statistics. Even the Asian sample size is more than double that of the black sample size. 

Granted, the statistics are still rather interesting, I just think that might be causing some skewing of results. 

Then again, I'm not the best at judging this kind of stuff by eye. You can use some test to determine the fit of compatibility of sample size variations. Or a few tests. I know that the output of a Tukey's HSD in an ANOVA can tell you, but I don't think that would work here. 

Also, so I have this straight, those sample sizes represent those accepted to medical colleges within the polling/sampling data as a whole, yeah? So, 124,503 would be 100% of the acceptances within the period? 



Edited by agentwhale007 - 31 August 2011 at 6:40pm
"So when Romney wins in a landslide, what will the liberal media do?"
This Ma**edited**hine Kills **edited**as**edited**ists.




Back to Top
ArthurBignose View Drop Down
Member
Member
Avatar

Joined: 01 August 2010
Status: Offline
Points: 137
Post Options Post Options   Thanks (0) Thanks(0)   Quote ArthurBignose Quote  Post ReplyReply Direct Link To This Post Posted: 31 August 2011 at 6:53pm
Originally posted by agentwhale007 agentwhale007 wrote:

I'd imagine that, to some extent, the fact that the sample size for black applications/acceptances is a considerable amount lower than that for white, and Hispanic is remarkably lower, is causing some of the discrepancy in the statistics. Even the Asian sample size is more than double that of the black sample size. 

Granted, the statistics are still rather interesting, I just think that might be causing some skewing of results. 

Then again, I'm not the best at judging this kind of stuff by eye. You can use some test to determine the fit of compatibility of sample size variations. Or a few tests. I know that the output of a Tukey's HSD in an ANOVA can tell you, but I don't think that would work here. 

Also, so I have this straight, those sample sizes represent those accepted to medical colleges within the polling/sampling data as a whole, yeah? So, 124,503 would be 100% of the acceptances within the period? 



Very good points.  It could make an interesting study for someone so motivated that is good at statistics.

The numbers given were the number of applicants.

Here is a break down of applicants, acceptees, and % acceptance.



Also a note on the validity of the numbers, this is just for MD, not DO.  The first application is sent to AMCAS, a part of the AAMC, and only 8 schools (out of 133) do not use this service.  The numbers are reported to the AAMC by all of the participating schools.


Edited by ArthurBignose - 31 August 2011 at 6:59pm


Back to Top
agentwhale007 View Drop Down
Platinum Member
Platinum Member
Avatar
Forum's Vladimir Lenin

Joined: 20 June 2002
Location: GNV FLA
Status: Offline
Points: 11696
Post Options Post Options   Thanks (0) Thanks(0)   Quote agentwhale007 Quote  Post ReplyReply Direct Link To This Post Posted: 31 August 2011 at 7:06pm
Originally posted by ArthurBignose ArthurBignose wrote:

 It could make an interesting study for someone
 

That it would. I'd be interested in seeing how these ethnicity pieces of data break down when looking at first round acceptance, second round, late acceptance, etc. 

Quote that is good at statistics.
 

Well, I'm out. 

Quote this is just for MD, not DO.  

You know, I worked for a osteopathic medical school for almost a year and I still don't really know the difference between an MD and a DO. 

I know that when drafting releases I used the phrase "a more hands-on approach," quite often, but I have no idea what that actually means. 
"So when Romney wins in a landslide, what will the liberal media do?"
This Ma**edited**hine Kills **edited**as**edited**ists.




Back to Top
 Post Reply Post Reply
  Share Topic   

Forum Jump Forum Permissions View Drop Down

Forum Software by Web Wiz Forums® version 10.03

This page was generated in 0.156 seconds.