COMLEX - USMLE calculation

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

builttorock

Full Member
10+ Year Member
15+ Year Member
Joined
Jan 16, 2007
Messages
101
Reaction score
0
Members don't see this ad.
Is it me or does the equation for the COMLEX-USMLE score conversion (JAOA, Sept. 2006) not really add up, per se?

USMLE Step 1 = 67.97 + 0.24 x COMLEX-USA Level 1 (R^2=0.68)

Do people who get a 225-230 USMLE really rip almost 700 on their COMLEX?

For those who have taken the exam already, how'd this correlate?
 
builttorock said:
Is it me or does the equation for the COMLEX-USMLE score conversion (JAOA, Sept. 2006) not really add up, per se?

USMLE Step 1 = 67.97 + 0.24 x COMLEX-USA Level 1 (R^2=0.68)

Do people who get a 225-230 USMLE really rip almost 700 on their COMLEX?

For those who have taken the exam already, how'd this correlate?
Click to expand...

it undervalues your comlex score compared to your USMLE. It was off by 20 points for my step 1 USMLE. That would have hurt me if i didnt try to do both.
 
That formula underpredicted my actual USMLE 1 score by 25. It makes me glad I have a real score to show and not an unaccurate predicted score.
 
builttorock said:
Is it me or does the equation for the COMLEX-USMLE score conversion (JAOA, Sept. 2006) not really add up, per se?

USMLE Step 1 = 67.97 + 0.24 x COMLEX-USA Level 1 (R^2=0.68)

Do people who get a 225-230 USMLE really rip almost 700 on their COMLEX?

For those who have taken the exam already, how'd this correlate?
Click to expand...

I wouldn't put much into this "study". Scores from 155 students at KCOM who took both COMLEX and USMLE step 1 exams were examined once to come up with this equation. I'm not impressed by the sample selection, size, single site, the laughable R^2, etc... there are numerous flaws.

Using this, my converted scores (240 and 612) would be:
USMLE->COMLEX: 240 = 67.97 + 0.24x = 716
COMLEX->USMLE: x = 67.97 + 0.24(612) = 214

Bottom line: just take both tests so your COMLEX score isn't subject to being converted by using a poorly designed equation derived from a ****ty "study", if this equation is in fact used by PDs to "convert" your COMLEX score for purposes of comparison to USMLE scores.
 
tideleonheart said:
I use this - see the attachment.

@cyclohexanol - using this chart, seems your scores were almost dead on.
Click to expand...

Even if my n=1 scores happen to fit this chart, it doesn't matter. The point is that the COMLEX->USMLE "conversion equation" published in JAOA for allo PDs to use is based on a totally bogus study (surprise? it's JAOA...).
 
builttorock said:
For those who have taken the exam already, how'd this correlate?
Click to expand...

A significant percentage of a COMLEX score is based on knowledge of manipulation. The USMLE is strictly a test of medical knowledge. Furthermore, the questions on the USMLE are lengthier and more complex than their counterparts on COMLEX.

There's really no valid way of equating the two scores.
 
Old_Mil said:
There's really no valid way of equating the two scores.
Click to expand...

You could definitely standardize the scores if you had a big enough sample size and do a simple student's t-test. It's basic biostats if you have the data. You just have to compare the number of SD your exam scores are compared to the mean for the sample.

Generating an equation on the other hand, is not going to be as simple.
 
Old_Mil said:
A significant percentage of a COMLEX score is based on knowledge of manipulation.
Click to expand...

Hmmm I did not realize OMT was THAT involved on the COMLEX.
 
Cp22kjer said:
Hmmm I did not realize OMT was THAT involved on the COMLEX.
Click to expand...

It's almost 20% of the test. For every 2 or 3 step question on the comlex, there's an OMM question that an 8th grader with Savarese could answer. That's why in general the USMLE is a much harder exam.
 
my scores are all over the board. USMLE step 1 213, COMLEX step 1 589, USMLE step 2 225, COMLEX step 2 569.... go figure. I still think the COMLEX is a crappier, more convoluted test. 🙄
 
Yup, it was a bad study when it came out in 2006, and it's still a bad study today.
 
The thing to consider:

Does anyone honestly believe that any ACGME program director really gives a hoot about the existence of a COMLEX-USMLE score converter?

If they don't accept the COMLEX, then the formula is irrelevant.
If they do accept the COMLEX, they will just look at the percentile and guess your relative performance on the USMLE. Or, they will only care that you passed (noncompetitive programs).
 
wanna_be_do said:
The thing to consider:

Does anyone honestly believe that any ACGME program director really gives a hoot about the existence of a COMLEX-USMLE score converter?

If they don't accept the COMLEX, then the formula is irrelevant.
If they do accept the COMLEX, they will just look at the percentile and guess your relative performance on the USMLE. Or, they will only care that you passed (noncompetitive programs).
Click to expand...

Exactly. Do urselves a huge favor and take USMLE if u looking into a competitive ACGME programs.
 
funshi baba said:
Exactly. Do urselves a huge favor and take USMLE if u looking into a competitive ACGME programs.
Click to expand...

X2 It underestimated my real USMLE score by 30 points. I definitely still passed using the converted score, but come on. 30 points is a huge margin, it can be the difference between a non-competitive and a decent or even competitive score.
 
You must log in or register to reply here.

Similar threads

F
comlex/ usmle in 8 weeks
Replies
7
Views
2K
frightenedmedicalstudent
F
I
DR comlex level 2 competitiveness?
Replies
2
Views
1K
DOVinciRobot
DOVinciRobot
O
Applying to IM Residency with only COMLEX scores
Replies
6
Views
4K
Dr. Robb Bassett
Dr. Robb Bassett
A
Comlex 1 Preparation Advice
Replies
2
Views
3K
yougotquestions
Y
T
Residency Explorer COMLEX cutoff scores (applying FM)
Replies
1
Views
1K
DrStephenStrange
DrStephenStrange
Top