Yearly Applicants and Matriculants by State, Adjusted for Population

This forum made possible through the generous support of SDN members, donors, and sponsors. Thank you.

efle

not an elf
10+ Year Member
Joined
Apr 6, 2014
Messages
14,146
Reaction score
22,781
Hello everybody. Another slower day so I decided to try and kill some time looking at AAMC's tables A-3 and A-4, Applicants and Matriculants by State of Residence.

The only vaguely interesting question I could come up with for myself was about how uniform the numbers might look, after adjusting for the population sizes of regions and states.

First fun fact: In a random sample of 100,000 US residents taken during last cycle, you should get that about 16.0 applied to allopathic medical schools that year, and 6.5 matriculated!

Second fun fact: The overall regions (as classified by AAMC in their tables, minus Puerto Rico) all look fairly close to the national numbers. And, the trend for matriculation was similar to application. [Northwest should be Northeast]

WCqbS0e.png


Third fun fact: At the state level that falls apart a lot. There is a great deal of variation between states within every region - for example Massachusetts had considerably more Apps per 100k than the nation, while Maine and New Hampshire had far fewer. Often the number of matriculants is much higher or lower than you might expect from the number of applicants. Here are the top and bottom five states for Yearly Applicants and Matriculants per 100,000 Residents.

5 Highest and Lowest for Apps:
u2S2Pz6.png


5 Highest and Lowest for Matrics:
STQ3TQI.png


As you can see there are some odd ones, like Vermont falling a little short of the overall rate of application, yet coming out about 50% greater than the overall rate of matriculation.

For the full set though, the trend holds well:

WdpcGTE.png


Of course, none of this is adjusting at all for average stats, number of public school seats, percentage of the state population that attends college or graduate school, percentage that lives in cities, SES, etc.

As always you can check out the data in a google spreadsheet here or just see an image of the table of all states here (sorted by app) or here (sorted by matric).

Members don't see this ad.
 
Last edited:
  • Like
Reactions: 8 users
Members don't see this ad :)
Would be inteeresting to base it on college graduates/state and not overall population.
 
  • Like
Reactions: 1 user
pfft semantics. i just lump any and all data work including collection, processing, and analysis as data mining.
I'm not sure if I'm supposed to answer you according to your folly or not, but calling this data mining is like saying that CNA's practice medicine.

Dividing numbers != data mining
 
  • Like
Reactions: 1 user
I'm not sure if I'm supposed to answer you according to your folly or not, but calling this data mining is like saying that CNA's practice medicine.

Dividing numbers != data mining

Just to clarify, i'm not talking about actually gathering large volumes of data and processing/analyzing them using fancy software. I'm simply using the term loosely to talk about any method to gather, process and analyze data even if the data gathered and analyses done are simple and quick.

But sure if we're talking technical definitions here, you're right that this isn't data mining. Congrats!
 
I was able to find a US gov spreadsheet with the percentage in each state attaining a bachelor's degree or more, from just a few years prior to the year they got the population data I used (2009 and 2013).

The modifiers for states ranged from 38.2% with a Bachelor's or more (Massachusetts) to 17.3% (West Virginia) with the overall nation at 27.9%. D.C. being just a city was a big outlier at 48.5% so I didn't put into scatter.

I was actually expecting a lot closer relationship than this. The R^2 for rate of attending college vs rate of applying to MD school is less than 0.2! This...is really weird.

vPSGa9x.png


Anyways, if you go ahead and sort Apps per 100,000 College-Educated Residents you get this.
 
  • Like
Reactions: 1 user
See above

Interesting. I think I'll come back and take a closer look at this later when I have time. Thanks though for adding that data in!
 
So I have a theory: the people destined for medicine are a very academically capable bunch that are OK with years and years of education. These people are likely in the outlying minority that are going to seek college ed no matter what the state - say they are in the "most academic" 10% of the population of a state for example. Your state sending 20% vs 30% to college isn't going to have much of an impact on medical applicants then. The future doctors were already all going to be there before you encouraged another 10% to attend college.

There's another weak but supporting trend for this:

wJAiCG4.png


As you increase the overall percent attending college, you actually see a decrease in the density of medical applicants among college students. This would make sense if the future doctors were already all going to be attending college even if only 15-20% of the state does so. Expanding the percent attending college doesn't boost MD applicants much overall, and within the college crowd it faintly dilutes them!

Probably could see similar in stuff like graduate engineering apps, etc.
 
Last edited:
Top