COMLEX to USMLE conversion formula from Journal of Grad Med Ed.

[niko327]

I found this recently. Its from Journal of Graduate Medical education. The authors caution the formula slightly underestimates the USMLE score. Interesting though.

Here's the formula: (0.2392 x COMLEX I score) + 82.563 = USMLE I score .

Here's the Journal link:
http://www.jgme.org/doi/abs/10.4300/JGME-D-13-00302.1

Authors are from College of Osteopathic Medicine of the Pacific at Western University of Health Sciences. If this formula is already widely known about, I apologize in advance.

---

Members don't see this ad.

 

[Drrrrrr. Celty]

It's been discussed multiple times before. It's neither useful nor does comparing it provide any useful information. Someone who studied for USMLE & COMLEX may obtain discorded results as major effort was towards USMLE and minor effort towards OMM. As opposed to only COMLEX where effort was in specifically targeting its high yield information and majorly going over omm.

In the end this formula really just continues to push the notion that having two separate tests has legitimate issues, especially in the coming years.

 

[acapnial]

I found this recently. Its from Journal of Graduate Medical education. The authors caution the formula slightly underestimates the USMLE score. Interesting though.

Here's the formula: (0.2392 x COMLEX I score) + 82.563 = USMLE I score .

Here's the Journal link:
http://www.jgme.org/doi/abs/10.4300/JGME-D-13-00302.1

Authors are from College of Osteopathic Medicine of the Pacific at Western University of Health Sciences. If this formula is already widely known about, I apologize in advance.

Not even close to correct for me

 

[Drrrrrr. Celty]

Not even close to correct for me

Not close for my class either. I.e average person who took the COMLEX in our class and usmle was at around a 560-565, average usmle = 224. That's a big margin of error.

 

[Geraltofrivia]

I got 225 on USMLE and COMLEX was 500. Not very close IMO?

 

[Mad Jack]

I found this recently. Its from Journal of Graduate Medical education. The authors caution the formula slightly underestimates the USMLE score. Interesting though.

Here's the formula: (0.2392 x COMLEX I score) + 82.563 = USMLE I score .

Here's the Journal link:
http://www.jgme.org/doi/abs/10.4300/JGME-D-13-00302.1

Authors are from College of Osteopathic Medicine of the Pacific at Western University of Health Sciences. If this formula is already widely known about, I apologize in advance.

It's within 4 points for me, which is close to the margin of error you'd expect. My USMLE was higher than expected by 4 points, which fits with the finding of the calculation being lower than real-world scores.

 

[MLT2MT2DO]

This is pretty god awful if you're outside of a small SD of the average.
617 comlex
253 USMLE

At some point you can't compare a trash test someone is trying to make a buck on vs something that goes through multiple QA sessions

Sent from my SM-G900V using SDN mobile

 

[SLC]

558 COMLEX
230 USMLE
USMLE score prediction based on that formula: 216

Nope, not close for me.

 

[chizledfrmstone]

Not close for my class either. I.e average person who took the COMLEX in our class and usmle was at around a 560-565, average usmle = 224. That's a big margin of error.

Not everyone who took the COMLEX took the USMLE.

 

[Drrrrrr. Celty]

Not everyone who took the COMLEX took the USMLE.

Right, hence why I referenced closer to the original average of the people who did which coincided with the majority of the people who also took the usmle and not our class average i.e 552.



Sent from my iPad using SDN mobile

 

[cheng994]

I completely understand the uselessness of this conversion. However, this formula is absolutely spot on for me. 237/649

 

[ortnakas]

Actually works for my scores within one point. Crazy.

 

[yankees527]

Mine was EXACT

Mind blown

 

[hallowmann]

Completely off, like underpredicted my Step by >15 pts.

It might work for some, but it doesn't work (and mainly underpredicts) enough for others to make it useless or even harmful.

 

[Postictal Raiden]

Completely off, like underpredicted my Step by >15 pts.

It might work for some, but it doesn't work (and mainly underpredicts) enough for others to make it useless or even harmful.

Yup, under predicts my step1 by 18 points.

 

[Drrrrrr. Celty]

Completely off, like underpredicted my Step by >15 pts.

It might work for some, but it doesn't work (and mainly underpredicts) enough for others to make it useless or even harmful.

Yup, under predicts my step1 by 18 points.

were your scores significantly different in percentiles?

 

[Postictal Raiden]

were your scores significantly different in percentiles?

No. that's the odd thing. Comlex was 637 (91%) and USMLE was 252 (88%).

 

[dfib slim]

Mine was only off by 2 points.

 

[jw3600]

No. that's the odd thing. Comlex was 637 (91%) and USMLE was 252 (88%).

I would say you under performed (relatively) on comlex.

 

[Postictal Raiden]

I would say you under performed (relatively) on comlex.

Thanks to OMM