Medschool GPA and USMLE correlation

This forum made possible through the generous support of SDN members, donors, and sponsors. Thank you.
Brigade4Radiant said:
From the University of Tennessee Medschool there seems to be a good correlation between grades in Medschool and USMLE scores.

http://www.uthsc.edu/Medicine/Acad_Affairs/UME/acad_perf/index.php


I think this is one of the positives of having the ABCDF grading scale compared to Pass/Fail.
Click to expand...

2 points:

1) That data would be a bit more interesting if we could see the range of USMLE scores for each GPA, rather than the mean.

2) You're confusing correlation and causality. Students who have a 4.0 GPA at UT would probably do just as well in a P/F school and get equally impressive Step I scores, and students with a 2.0 GPA would probably do just as poorly.
 
KeyzerSoze said:
2 points:

1) That data would be a bit more interesting if we could see the range of USMLE scores for each GPA, rather than the mean.

2) You're confusing correlation and causality. Students who have a 4.0 GPA at UT would probably do just as well in a P/F school and get equally impressive Step I scores, and students with a 2.0 GPA would probably do just as poorly.
Click to expand...

This. A 4.0 student is a 4.0 student. They will do just fine with or without grades.
ABCD results in unneeded stress.
 
lord_jeebus said:
I don't think you can say it doesn't change usmle scores - they had a 7 point drop! The study is really underpowered to draw a conclusion either way.
Click to expand...

That's true, just thought it was implied I meant there was no significant change since there's always some inherent variability in any measurement.

In the two other studies I found on pubmed about pass/fail on USMLE, there was a nonsignificant increase in USMLE scores.

http://www.ncbi.nlm.nih.gov/pubmed/20012686

http://www.ncbi.nlm.nih.gov/pubmed/19704204

I think that's a good point about not reaching a definitive conclusion from the first study, but every published report I've seen has concluded there's been no change in USMLE scores (and an increase in psychological well-being).

It's only a matter of time before most people stop clinging to the idea of a tiered grading system to motivate learning in the preclinical years
 
The second article from pubmed above is from the professors at my school. Our USMLE scores have been on the uptrend with averages in the mid 230s. These would've been the classes where the P/F first went into effect.

As per the op, there may be a rough correlation between grades and Step 1 score where as if you were struggling to pass before, you may be equally struggling to pass the step 1. But people who repeatedly excel in classes may or may not excel on the exam. There are many cases at my school.
 
I go to a pass/fail school, and school administrators have often shown us correllations between average percentage earned and USMLE score. The correllations are not extremely tight, but they do show definite trends. Even if you go to a pass/fail school, generally you do get your percentage on tests and for the class as a whole. Someone averaging ~90 in their classes is likely to do better on the USMLE than someone barely passing. No correllation is perfect, of course, but the trend is inarguable.

One does not need to go to an ABDC graded school to know how they stand within their class and how this correllates to USMLE success.
 
So let me get this straight. You are implying that students who do well on a series of multiple choice tests about medically related basic sciences are more likely to do well on an extensive multiple choice test about medically related basic sciences. This is groundbreaking research.
 
BigRedder said:
So let me get this straight. You are implying that students who do well on a series of multiple choice tests about medically related basic sciences are more likely to do well on an extensive multiple choice test about medically related basic sciences. This is groundbreaking research.
Click to expand...

No, he's implying that students who do well on a series of multiple choice tests and are assigned grades for their performance are more likely to do well on an extensive multiple choice test than students who do well on a series of multiple choice tests but are not assigned grades for their performance. That would be groundbreaking research, if it were true.
 
Last edited:
BigRedder said:
So let me get this straight. You are implying that students who do well on a series of multiple choice tests about medically related basic sciences are more likely to do well on an extensive multiple choice test about medically related basic sciences. This is groundbreaking research.
Click to expand...

No that overall it might help you on the boards since you know how you are performing.
 
Brigade4Radiant said:
No that overall it might help you on the boards since you know how you are performing.
Click to expand...

Why wouldn't you know how well you are performing at a pass/fail school? I find it hard to beleive that there are pass/fail schools where students don't even know how many questions they got wrong on their tests. One can calculate their own percentages from there. My school supplies averages on the exams for the class as well. I get grades, I know what they are and how I stand in the class. The only difference is that these percentages/grades are not listed on my record.
 
All this is great and good except true Pass/Fail is an illusion in all but a few select schools. The vast majority of P/F schools still rank their students on their collective test scores and bins them (by quartiles or whatever) accordingly.
 
ThaliaNox said:
Why wouldn't you know how well you are performing at a pass/fail school? I find it hard to beleive that there are pass/fail schools where students don't even know how many questions they got wrong on their tests. One can calculate their own percentages from there. My school supplies averages on the exams for the class as well. I get grades, I know what they are and how I stand in the class. The only difference is that these percentages/grades are not listed on my record.
Click to expand...

qft, we get that info too. it's not hard to see roughly where you stand and you're crazy if you think that you can extrapolate anything else from any more detail. you being in the 2nd quartile and the next guy being in the 3rd says zippo about your potential board scores.

Orbo said:
All this is great and good except true Pass/Fail is an illusion in all but a few select schools. The vast majority of P/F schools still rank their students on their collective test scores and bins them (by quartiles or whatever) accordingly.
Click to expand...

you're right, but the vast majority of P/F schools retain ABCF for 3rd year because those grades are what matter to residency directors. pre-clinical grades are so far down the list in terms of what PDs look at, so P/F is a huge benefit. Getting rid of grades only rids you of something that can hurt you but has precious little power to help.
 
The whole idea of P/F over standard grading has more to do with decreasing the amount of competition against fellow students. Get them to work together and teach each other. Whether intentional or not, I think a lot of people at my school are passive aggressive about things. We may not give you wrong information or intentionally sabotouge you, but we will maybe not commit as much time to teaching another student a concept or working together to create chapter summaries. Random crap that does increase overall learning. I feel both systems are still kind of "me first", but P/F is a little less than straight up grades. It seems nice to just focus on doing the best you can rather than worrying what percentage you need on the last exam to get an A.

I suspect the vast majority of us in med school (even the "lazy" ones) are still not the type that would decrease overall effort because we don't get gold stars. Most of us have a weird internal drive that goes beyond that. Personally, regardless of how well I did on an exam, I always feel there was room for improvement. I've gotten 100s on quizzes and still left feeling I could've known more.
 
MossPoh said:
The whole idea of P/F over standard grading has more to do with decreasing the amount of competition against fellow students. Get them to work together and teach each other. Whether intentional or not, I think a lot of people at my school are passive aggressive about things. We may not give you wrong information or intentionally sabotouge you, but we will maybe not commit as much time to teaching another student a concept or working together to create chapter summaries. Random crap that does increase overall learning. I feel both systems are still kind of "me first", but P/F is a little less than straight up grades. It seems nice to just focus on doing the best you can rather than worrying what percentage you need on the last exam to get an A.

I suspect the vast majority of us in med school (even the "lazy" ones) are still not the type that would decrease overall effort because we don't get gold stars. Most of us have a weird internal drive that goes beyond that. Personally, regardless of how well I did on an exam, I always feel there was room for improvement. I've gotten 100s on quizzes and still left feeling I could've known more.
Click to expand...

This is generally correct. The schools that have gone to Pass/Fail have done so to increase student cooperation. If you attend a school that uses grades instead of Pass/Fail and believe for one second that because you have an A in a course, you are a lock for USMLE Step I, you are sadly mistaken. While coursework provides the educational base work for Step I, you have to be able to apply that coursework. This is why review books for coursework are usually a waste of time and why textbooks for Step I (or any of the other USMLE steps) are usually a waste of time. It's a process that you don't appreciated until you are on the side where I am, at the attending level. You do learn that you have one shot at not "screwing the pooch" so don't try to "bulls--" your way through coursework. It comes back at some point to bite.
 
Our school has grades, but I don't know that it decreases cooperation, per se. I'd say in general we're a pretty cohesive group.

What it DOES do is make the school go to RIDICULOUS lengths to make sure that the average on every single exam is exactly an 89 with one of the tightest bell curves you've ever seen. Seriously, you can go up 25%ile by getting a 91 or 92 (equivalent to missing 1 or 2 different questions on a given exam). The school doesn't change the exams from year to year, or change questions because they are so fixated on giving certain questions that a certain percentage of students will get correct. Your grade is reported as the number percentage that you get - if you get an 89% your transcript will show an 89% (not A-F).

I think the utter ridiculousness of the system increases cohesion if anything:meanie::meanie:.
 
pianola said:
Our school has grades, but I don't know that it decreases cooperation, per se. I'd say in general we're a pretty cohesive group.

What it DOES do is make the school go to RIDICULOUS lengths to make sure that the average on every single exam is exactly an 89 with one of the tightest bell curves you've ever seen. Seriously, you can go up 25%ile by getting a 91 or 92 (equivalent to missing 1 or 2 different questions on a given exam). The school doesn't change the exams from year to year, or change questions because they are so fixated on giving certain questions that a certain percentage of students will get correct. Your grade is reported as the number percentage that you get - if you get an 89% your transcript will show an 89% (not A-F).

I think the utter ridiculousness of the system increases cohesion if anything:meanie::meanie:.
Click to expand...

That's an absolutely ridiculous system. Then again there are few systems which aren't ridiculous.

As sort of an example, some classes in my school will take the class average and use that as the mean (and score for high pass) if it's >80%, then have mean + 1 SD = honors and mean - 1.5 SD = pass.

Then on the other hand some classes will just set a blanket limit for honors, high pass, and pass and that's that. No curve no nothing (unless of course people do tremendously poorly on an exam).
 
MossPoh said:
The whole idea of P/F over standard grading has more to do with decreasing the amount of competition against fellow students. Get them to work together and teach each other. Whether intentional or not, I think a lot of people at my school are passive aggressive about things. We may not give you wrong information or intentionally sabotouge you, but we will maybe not commit as much time to teaching another student a concept or working together to create chapter summaries. Random crap that does increase overall learning. I feel both systems are still kind of "me first", but P/F is a little less than straight up grades. It seems nice to just focus on doing the best you can rather than worrying what percentage you need on the last exam to get an A.

I suspect the vast majority of us in med school (even the "lazy" ones) are still not the type that would decrease overall effort because we don't get gold stars. Most of us have a weird internal drive that goes beyond that. Personally, regardless of how well I did on an exam, I always feel there was room for improvement. I've gotten 100s on quizzes and still left feeling I could've known more.
Click to expand...

I go to an ABCfail school and competition barely exists because there is no official curve. Professors may try to manipulate tests so there is a certain mean (as someone mentioned in a diff thread) but there is no detriment to students helping out other students. We help each other a lot and it's a great atmosphere.

Personally ABC is a good system to push through those tough nights when motivation is lacking. But it is a bad system for when you're just cramming minute details to ensure you get that A and not learning the best way.

I know in P/F you still have motivation b/c of class rank but ABC just gives more tangible goals like a 4.0. I know without grades I would not have taken Micro and Pharm seriously the 1st half of 2nd yr and I would have paid for it later
 
Ok i could not make anything out of this discussion
So I'm doing my mbbs and I'm a good student but not getting a good gpa because of my university poor exam evaluation which is not my fault.
But i think i can do much better in usmle
So i want to ask how much does the gpa matter in usmle scores?
 
Sushil kumar said:
Ok i could not make anything out of this discussion
So I'm doing my mbbs and I'm a good student but not getting a good gpa because of my university poor exam evaluation which is not my fault.
But i think i can do much better in usmle
So i want to ask how much does the gpa matter in usmle scores?
Click to expand...

Define matter
 
Sushil kumar said:
Ok i could not make anything out of this discussion
So I'm doing my mbbs and I'm a good student but not getting a good gpa because of my university poor exam evaluation which is not my fault.
But i think i can do much better in usmle
So i want to ask how much does the gpa matter in usmle scores?
Click to expand...

There's a correlation. If you're getting mostly As, those students tend to achieve higher USMLE scores than those who are getting mostly Cs. Although, this is not a hard and fast rule obviously. People who get As can easily get a low USMLE and visa versa. It just shows a correlation. It's most likely due to the students getting As, would continue their study habits for their board studying, thus resulting in a correlation of higher scores, of course, a higher knowledge base of the A students obtained during pre-clinicals may also play a role.
 
There's a reason grades, curves, and other comparative metrics have been around for a long time. The fear of being below average is excellent motivation for most students. 90% of humans are extrinsically motivated with an external locus of control. They respond to reward. The idea that people will try equally hard without reward is so obviously *****ic that it does not even merit a response. Competition brings out the true best in people. In pure P/F systems, a student may think that he or she is giving it 100%, but I doubt that he/she would stay up past midnight to get the same P as everyone else.
 
FiveOClock said:
There's a reason grades, curves, and other comparative metrics have been around for a long time. The fear of being below average is excellent motivation for most students. 90% of humans are extrinsically motivated with an external locus of control. They respond to reward. The idea that people will try equally hard without reward is so obviously *****ic that it does not even merit a response. Competition brings out the true best in people. In pure P/F systems, a student may think that he or she is giving it 100%, but I doubt that he/she would stay up past midnight to get the same P as everyone else.
Click to expand...

Yeah, but no medical student is going to voluntarily tow the line. No reason to bust, fail out of medical school and all you're sitting on is a mountain of debt and depression.
 
100% agree. My school has a decade's worth of data showing that the best predictor of COMLEX performance is preclinical GPA. Nothing else comes close, except MCAT, but much weaker.

Two other observations are worth noting: our worst students who fail Boards usually explain that they were unprepared, while better students who fail (and surprise us) get hit with some life event just before taking the exam.

We all understand the rationale for the P/F scheme. But a perverse mission creep always sneaks in. You see this scenario: P/F becomes H/P/F which then becomes H/HP/P/F, which ends up the same as A/B/C/F.

A number of schools use a percentile grading system because it at least removes the stress of your having an 89 and wanting that B to be a a 90 and an A.


FiveOClock said:
There's a reason grades, curves, and other comparative metrics have been around for a long time. The fear of being below average is excellent motivation for most students. 90% of humans are extrinsically motivated with an external locus of control. They respond to reward. The idea that people will try equally hard without reward is so obviously *****ic that it does not even merit a response. Competition brings out the true best in people. In pure P/F systems, a student may think that he or she is giving it 100%, but I doubt that he/she would stay up past midnight to get the same P as everyone else.
Click to expand...
 
Rekt said:
Yeah, but no medical student is going to voluntarily tow the line. No reason to bust, fail out of medical school and all you're sitting on is a mountain of debt and depression.
Click to expand...


I could pass med school without sweating. So could most people at reasonable med schools. But I would not get nearly as much out of it as I do when extra motivated. It takes exponentially more work to score 90 on an exam when the class average is 80. You have to know and understand so much more.
 
FiveOClock said:
There's a reason grades, curves, and other comparative metrics have been around for a long time. The fear of being below average is excellent motivation for most students. 90% of humans are extrinsically motivated with an external locus of control. They respond to reward. The idea that people will try equally hard without reward is so obviously *****ic that it does not even merit a response. Competition brings out the true best in people. In pure P/F systems, a student may think that he or she is giving it 100%, but I doubt that he/she would stay up past midnight to get the same P as everyone else.
Click to expand...
My preclinicals were true P/F, but we still saw the class average and knew our own scores. Plenty of motivation to do well, not a big deal if you couldn't kill a particular sequence because you were ill, had to attend a wedding, whatever. Having that cushion was very helpful. Don't think it negatively affected my motivation or learning at all in the long term.
 
Sushil kumar said:
Ok i could not make anything out of this discussion
So I'm doing my mbbs and I'm a good student but not getting a good gpa because of my university poor exam evaluation which is not my fault.
But i think i can do much better in usmle
So i want to ask how much does the gpa matter in usmle scores?
Click to expand...

Says every single medical student ever.
 
You must log in or register to reply here.

Similar threads

R
USMLE Step2
Replies
3
Views
2K
Redpancreas
Redpancreas
Krakatoa83
ERAS USMLE Transcript Sent One Day Late
Replies
2
Views
774
JuicedToTheSocks
J
U
Is late august ~28-30th too late for USMLE 2CK this cycle?
Replies
1
Views
422
ontiquicuo
O
H
Practice questions for medschool
Replies
5
Views
2K
amath03
A
SSerenity
Worlds Highest USMLE Score by an A.I
Replies
27
Views
3K
Cornfed101
Cornfed101
Top