Keeping you sane through statistics

[FA22raptero]

Hi All, 

Im a law student accepted to UBC, but I keep an eye on these forums as I have a friend who is desperately wanting to get into UBC med. 

I figured I'd share a perspective from a law applicant where it seems like unlike medschool, a lot more weight is put on our standardised testing (LSAT) than for you folks. Trying to stick to numbers kept me sane through the admissions process, and I figured I'd share jusy in case it helped someone else. For law school, second to LSAT is GPA, and then we also submit a personal statement, and while the PS might have some effect on your application, for most schools something like 80+% of the weighting is some combination of GPA and LSAT. 

It sounds like for Medschool your MCAT and GPA are very important, along with your NAQ's for getting an interview, and then post-interview a huge factor is then in your interview score, something like 50%int./50%rest. 

About 100,000 people each year write the MCAT in North America, the most recent number I could find was 86,000 from a 2012 U.S. news. AAMC percentile distribution for 2018/2019 is based on scores from 2015 2016 and 2017, and is listed as n=~240,000, or about 80,000 per year. So lets round up to 100,000.

The United States is about 10x larger than Canada, and there is no substantial reason in my mind to believe that one country is going to have a greater proportion of people write the MCAT than another. Lets also assume people only apply within their own country (this is obviously untrue, but it sounds like not too many americans apply to Canadian schools). So in Canada, roughly 10,000 people write the MCAT. Lets round up to 15,000. That means than in Canada, only roughly 3000 students score higher than a 510 (80th percentile) on the MCAT. I think this is high, based on the AAMC numbers. If i dont do any rounding, only 1600 students score higher than a 510 each year. It is generally accepted that if you are going to score lower than 80th percentile on the MCAT, you better have a killer GPA and NAQ's. With that said, there must be a fairly strong degree of corrolotaion based on the scientific nature of the MCAT between GPA and MCAT scores. 

What I'm getting at is, if you look at UBC's average accepted MCAT, it is about 513/514, which is the 90th percentile. On the high end of my estimate, there are only 1500 people in the country who scored that high. On the low end, about 800 did. Because this is an average, obviously some people score higher and some people score lower. But if you were to take the average MCAT admission statistics for all Canadian schools and reference them against the number of canadian MCAT test takers and the associatated percentile scores, I am confident that you would find that it would be almost impossible for all canadian schools to have such high average MCAT scores without admitting almost all of the people who scored above a 514. It seems like UofT and a couple others might be exceptions to the pervasiveness of the MCAT. What I'm saying really is that there isnt much reason to believe that admission to medical schools is bimodal (Aka people with really high MCAT get in, and people with really low MCAT get on, and people with medium MCAT generally dont get in). Maybe someone has information to suggest otherwise.

But do you see the beauty in this? It doesn't matter if UBC says that your MCAT score only makes up 10% (i know they dont say exactly that) of their admission decision, because the truth is in the numbers. Lets take a hypothetical school that has an average entrance MCAT of 514, and the school claims that they do NOT use the MCAT AT ALL in their evaluation of your application. The beauty is it doesnt matter, because if you scored a 514, there are only about 800-1500 other students "better or equal to you". If a school publishes the average MCAT score of who got into their program, even if they dont use it, what this tells you then is the MCAT score is still a nearly perfect predictor of your medschool chances. Most of these students probably all applied to the same schools, too, which makes it so that theres this inflated idea that your odds are extraordinarily low, but that with roughly 2000 spots in medschool in Canada, if you score a 515/516 on your MCAT and apply to all of them, your odds of acceptance to at least one is getting really close to 100% (to find out what it really is, you could do the analysis I mentioned previously)

So then, wouldnt GPA also be a great predictor? It, too, is a great predictor! But how? Well, it turns out the the correlation between GPA and MCAT scores, based on MCAT percentile data as well as admission statistics, must be very close. 

So what if you are a splitter? In my law school apps, i had a relatively abysmal gpa and a 98th percentile lsat. So heres the other cool thing, statistically, the chances of being a splitter are quite low, but the probability that you post on a forum to inquire about you splitter status is substantially greater than for someone who is flat average. So it seems like more of us than there should be, but statistically it is unlikely that there are. 

So what am I trying to say? Im a little bit baffled, because in another post someone had a table suggesting that there were 13,000 individual canadian medschool applicants. Based on my MCAT percentile analysis, this is very odd. Most people acknowledge, I feel, that if you re going to score lower than say, the 70th percentile on the MCAT, you better be a Nature published olympian. So why are there so many applicants who are not at all competitive applying? I dont know. Maybe my numbers are totally wrong, but even if they are, I'm not sure they are off by a factor of 10, unless since 2011 the number of MCAT writers increased by a factor of 10 to almost 1 million students. The other option is people get so emotionally attached to medschool that they succumb to a sunk cost fallacy; even though they score poorly on the MCAT, because they spent so much time studying for it they feel the need to apply to every school for which they met the minimums, no matter what they scored. 

Anyways, this is the sort of analysis that I did when applying to law school. Allard takes 180 students, with like 1600-1800 applicants which felt insane to me. But the truth is that a good chunck of those 1600-1800 applicants were a bunch of the same people all appying to the same schools, and the rest were people applying out of a desperate hope that maybe somehow they could get in even though they were well below average.

This is a bit of a cynical analysis, but I always like thinking analytically. Is this something thay makes sense? Or is there a major flaw in my reasoning? This type or reasoning is probably reasuring if you have a strong MCAT and GPA, but probably not so reassuring if you dont. 

In the end, I'm trying to say that if you are a hardworking student who did well on the MCAT and has a strong GPA and who volunteered once in a while, I think your admission chances are actually higher than it would at first appear.