What do y’all make of the high 70’s bump on the LizzyM score for new data?

  • Thread starter Thread starter 907914
  • Start date Start date
Discover Your Odds of Getting into Medical School
This forum made possible through the generous support of SDN members, donors, and sponsors. Thank you.
9

907914

So, the old MCAT/GPA data follows a solid bell curve. The new data has a prominent upper range bump. Does this show that there really is a higher number of upper-stats applicants?
Members don't see this ad.
 

Attachments

  • C14202EB-A4AE-41AB-A95C-42FB2B856ACE.png
    C14202EB-A4AE-41AB-A95C-42FB2B856ACE.png
    135 KB · Views: 432
  • CF96C33D-C8A3-424B-8D49-5B66E8BFA8FD.png
    CF96C33D-C8A3-424B-8D49-5B66E8BFA8FD.png
    144.3 KB · Views: 434
Members don't see this ad :)
seanm028 said:
I mean, these are self-reported LizzyM scores, which are themselves based on more, self-reported data, so I'd take it all with a grain of salt. It could also very well be some sampling bias, since we can be reasonably sure that SDN is not a random sample of all applicants.
Click to expand...
I thought the LizzyM data was based on the MCAT/GPA grids?
 
seanm028 said:
I thought those were self reported by SDN users... Are they pulled from elsewhere?
Click to expand...
Those are official AAMC datasheets that include all applicants to medical school. I'm not sure what year the latest data comes from, though.
 
LizzyM said:
More people applying with inflated GPAs (grade inflation, people with high GPAs more likely to apply than others)??

I believe that MCAT is normally distributed and that hasn't changed, AFAIK.
Click to expand...
Any idea why there was not that bump in years past? Could undergraduate institutions somehow have coordinated to yield only a specific set of inflated GPA candidates over the last three years?
 
Wahkoon said:
Most likely grade inflation, but why would it just now be showing up? This isn’t exactly a new concept...
Click to expand...
I know it is not a new concept, which is why it is odd that it appears in the 2016-2019 but not the 2013-2016
 
Members don't see this ad :)
I asked @pizzadog , who created the Application Assistant (LizzyM application), to double check. They will pull the number of applicants with LizzyM > 70 to share on the forums for the new and old data.
 
Last edited by a moderator:
There's an error in here for sure. There are absolutely NOT more people with a LizzyM of 80 than 75. In fact that entire area of the curve is wildly inaccurate because a LizzyM of 80+ can happen, at maximum, less than 0.5% of the time (even with a 4.0 you'd need a 40+/525+)

Wouldnt trust AT ALL
 
Last edited:
Well, it could be that the total number of Tiger Parents is the same, though more of their children apply to medical school than did in the past.
 
Hi everyone, as @WildWing mentioned I helped create this calculator. I pulled some data and got some interesting results!

In 2013-2016 there were 146,690 applicants of which 62,132 matriculated. Of the applicants 34,920 (or ~24%) had LizzyM >= 70 and 27,445 matriculants (~44%) had LizzyM >=70

From 2016-2019 there were 123,445 applicants that we have data on of which 24,512 (or ~ 20%) had LizzyM >=70 and 19,638 matriculants (~40%) had LizzyM > 70.

It seems actually that the overall proportion of LizzyM scores has gone down above 70! The calculator's distribution curves actually show this. If you put in a LizzyM of 70 you will see it predicts you would have been in the 79th percentile of all applicants prior to 2016 and in the 83rd percentile of all applicants after.

If we drill down further though we can see that those who get a LizzyM score above 70 tend to get higher scores than previously and that is where we see the bump. Before 2016, 20% of applicants who were at or above 70 also scored at or above 75. After 2016 that number jumps to 34%. Let me know if you guys have any more questions.

Edit:
Forgot to reply to @efle comment
In fact that entire area of the curve is wildly inaccurate because a LizzyM of 80+ can happen, at maximum, less than 0.5% of the time
Click to expand...
This is actually a very good observation. If you look at the percentile of the 2016+ distribution curve for a LizzyM score of 80 you will see it places you at the 97th percentile. This means we estimate that ~3% of individuals have a LizzyM >= 80. However, as you point out, to have a LizzyM >= 80 you need an MCAT of at least 523 which places you in the 99th percentile of the MCAT distribution. So how could 3% of applicants have LizzyM >= 80 if only 1% of MCAT takers score at least a 523? This seems paradoxical but actually has a simple explanation - a lot of people who take the MCAT do not submit an application to any medical school, or may not submit to an allopathic medical school, or may retake it! Therefore, the percent of allopathic medical school applicants with an MCAT >= 523 will be greater than the score's corresponding MCAT percentile

It is also very important to keep in mind that curves are an estimation based on the limited data we have access to.
 
Last edited:
efle said:
There's an error in here for sure. There are absolutely NOT more people with a LizzyM of 80 than 75. In fact that entire area of the curve is wildly inaccurate because a LizzyM of 80+ can happen, at maximum, less than 0.5% of the time (even with a 4.0 you'd need a 40+/525+)

Wouldnt trust AT ALL
Click to expand...

How does one calculate LizzyM using the new MCAT?
 
efle said:
You'd just line up the percentiles between new and old scores

Can't really be accurate above 38/524 any more. But by then you're talking about near 4.0 and top 1% so it doesn't matter anyways
Click to expand...

I'm surprised at the number of posters here who have LizzyM >= 80.
 
MemeLord said:
schools.studentdoctor.net

LizzyM Score Calculator - What are my medical school chances?

The LizzyM Score Application Assistant Calculator helps answer What are my chances for medical school based on AAMC admissions data on admitted med students.
schools.studentdoctor.net schools.studentdoctor.net

First result when you google LizzyM...

This is what the entire thread has been about...
Click to expand...

I thought LizzyM was originally based on the old MCAT and there was a dispute about converting it to the new MCAT. Lots of LizzyM >= 80 in these parts.
 
MyOdyssey said:
I'm surprised at the number of posters here who have LizzyM >= 80.
Click to expand...
MyOdyssey said:
I thought LizzyM was originally based on the old MCAT and there was a dispute about converting it to the new MCAT. Lots of LizzyM >= 80 in these parts.
Click to expand...
People are inaccurate when converting at the upper end. Like they treat a 523 as a 40, instead of a 38. That's why it's 10x as commonly listed in people's signatures now.
 
Perfect scores are more common these days than they once were; a SDNer once posted a chart indicating there were ~13 perfect scorers per year. In the old MCAT, perfect scores were something that happened a few times a decade.
 
efle said:
People are inaccurate when converting at the upper end. Like they treat a 523 as a 40, instead of a 38. That's why it's 10x as commonly listed in people's signatures now.
Click to expand...

So what's the correct way to compute LizzyM scores by hand?
 
MyOdyssey said:
So what's the correct way to compute LizzyM scores by hand?
Click to expand...
Find the percentiles for old and new, convert, then add your GPAx10.
Way up at the high end there's no real way to do it accurately. On the old scale a 40 was 99.8th percentile though, so really "80+" is already a small fraction of a percentage of people.
 
LizzyM said:
Someone came up with something clever but I can't remember what it was.
Click to expand...
While this does not directly correlate with LizzyM scores from years gone by, Maybe we can have a new LizzyM scale that is GPA*10+(MCAT-472)....? That would give a max score of like 96. Hell, make it GPA*10+(MCAT-468) that way it has a scale from 4 to 100 lol
 
Pretty sure there’s an error in the graph on the LizzyM website. If you put in 520 and 4.0 That’s a LizzyM score of 77, before the 2nd peak. The MCAT is normally distributed so unless there are a lot of people with 520 on the MCAT not applying to medical school (much greater than the number of people with a 523 not applying to med school) then the graph makes no sense. The GPA can’t explain the bump bc we put a 4.0 above and couldn’t hit that bump.
 
detritus said:
Pretty sure there’s an error in the graph on the LizzyM website. If you put in 520 and 4.0 That’s a LizzyM score of 77, before the 2nd peak. The MCAT is normally distributed so unless there are a lot of people with 520 on the MCAT not applying to medical school (much greater than the number of people with a 523 not applying to med school) then the graph makes no sense. The GPA can’t explain the bump bc we put a 4.0 above and couldn’t hit that bump.
Click to expand...
That is the whole point of this thread
 
detritus said:
Pretty sure there’s an error in the graph on the LizzyM website. If you put in 520 and 4.0 That’s a LizzyM score of 77, before the 2nd peak. The MCAT is normally distributed so unless there are a lot of people with 520 on the MCAT not applying to medical school (much greater than the number of people with a 523 not applying to med school) then the graph makes no sense. The GPA can’t explain the bump bc we put a 4.0 above and couldn’t hit that bump.
Click to expand...

The point of the score is to compare your LizzyM score to the LizzyM score of the schools you are interested in. You should be within 1 point of the school's score. If you are applying with a 4.0/520, you really don't need to compute a LIzzyM score to know that you can apply anywhere you damn well please and you won't be out of your league (you might be yield protected but that's another story).
 
LizzyM said:
The point of the score is to compare your LizzyM score to the LizzyM score of the schools you are interested in. You should be within 1 point of the school's score. If you are applying with a 4.0/520, you really don't need to compute a LIzzyM score to know that you can apply anywhere you damn well please and you won't be out of your league (you might be yield protected but that's another story).
Click to expand...

I concur - the bump is probably because it starts to blend significantly around the upper tiers due to significant correlation between MCAT and GPA. A perfect bell curve makes sense only if GPA and MCAT score are independent (as the formula uses both values), but GPA is highly correlated with MCAT score, especially at the 520+ regime, so most people with a high MCAT also have a high GPA and vice-versa. Essentially, GPA and MCAT score could independently be normally distributed, but when pairing the data, the two situations are not independent at all and so the LizzyM score isn't normally distributed.

I think that this may also (not necessarily consciously) be recognized by schools, and so at that region GPA and MCAT mean much less (assuming you make the 75+LizzyM "cutoff", you're going to be more heavily weighed on every other part of your application).
 
You must log in or register to reply here.

Similar threads

LindaAccepted
New AAMC Data on Medical School Applicants and Enrollment in 2024
Replies
15
Views
4K
deleted1199414
D
MonsterGod
  • Locked
What are the secondary for University of Minnesota 2024-2025?
Replies
1
Views
610
wysdoc
wysdoc
S
What's the benefit of attending a highly prestigious medical school over a local one?
Replies
18
Views
3K
mistafab
mistafab
P
What is the limit of appropriate topics for failure essays
Replies
6
Views
873
pleasehelpme12
P
amateurbater
Are y’all sure you want to do this?
2
Replies
51
Views
8K
@#Frogthejam#@
@
Top