How does the AAMC Artificially "Deflate" MCAT Scores?

A 515 in 2018 is actually worth less than a 515 in 2015 and here's why.



The AAMC is a “norm"-referenced standardized exam. This means it is expected to exhibit a “normal(norm) distribution”.(Think about it in terms of the “bell-curve”). This also means that 50% of the people who take this exam should be under the 50th percentile which is roughly 501~ and 50% of test takers should score above this. 


Now students will say “cool this is highschool statistics, why would I care?”


You should care because the AAMC is changing the curve of the exam by deflating students' scores. 



Here's how this works:


1.Every year test takers get BETTER at the exam i.e the average scores go up. This is because  prep materials improve, students practice more, people post free MCAT Strategy Courses online etc. 


2. As more people start scoring highly(above the 50%ile)  the 50th percentile score moves up


3. To keep the 50th percentile relatively constant the AAMC has to move scores “down”.


The AAMC (to protect their status as a professional testing organization) needs to make sure that scores are “equivalent”. This means a 500 in 2015 should be worth the same as a 500 in 2018. Equivalency is necessary because admissions committees compare students with the same test scores from different years.



                 

        The AAMC can fix MINOR deviations by just playing with the scaled

      scoring a little bit since it's not linear anyway.



4. In recent years the curve has become very lopsided. It's “left skewed” meaning more people are scoring in the upper 25th percentile than they statistically should. 



 This is a MAJOR deviation. It’s too difficult to adjust this by playing with the curve

since you would just be “shifting” the lopsided part of the curve to the left or right.

This severely limits AAMC options and forces them to take drastic measures.



To correct major deviations the AAMC structurally modifies the exam itself to become harder. A more challenging exam will “deflate” scores across the board to affect everyone and this evens out the distribution of scores to the point where they can adjust “minor” deviations by playing with the raw score to scaled score conversion.




If this sounds like a conspiracy theory we ask students to do the following:


  1. Take a look at the AAMC Sample MCAT Exam or Full length test 1 and compare it to the newest full length. No need to complete the exam just looking through it is fine. 


Do they even look like the same test anymore?


  1. If you have friends that recently tested ask them if they felt like the real test looked even  remotely close to the practice exams from AAMC. If they are willing you can even ask them about changes in their predicted vs actual score. 



We felt it was important for students to be aware of this trend because it’s too easy for us to be so busy studying we forget that the AAMC is a professional test making company and that they gamify many components of the exam(like using psychological warfare). 


In recent testing cycles the increase in difficulty of the real test has been significant due to  increased 50/50 style questions and many other tricks such as hybrid-style CARS passages.


It can be incredibly frustrating to get an actual test score that is much lower than what we thought we would get. But knowledge of the AAMC’s tactics is the first step we need as average student’s to combat them. We hope this information helps and we have lots of other free resources available to help you gain an edge on the AAMC so feel free to check them out!