Most predictive practice test compared to USMLE step 2CK

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

Most predictive practice test


  • Total voters
    43

Dr. Anonymouss

Full Member
5+ Year Member
Joined
Nov 12, 2018
Messages
1,124
Reaction score
1,636
I have done some searching around and based on what I found, people on reddit seem to say that UWORLD SA2 is the most similar and predictive to the real test. I have also seen that the new free 120 is also a good predictor. For those who have take step 2, what practice tests did you feel like were the best predictor and most similar to the real thing?

Members don't see this ad.
 
Took NBME's every week (9, 10, 11, UWSA1, UWSA2) leading up to my exam.
NBME9 underpredicted by 10.
NBME10 underpredicted by 10.
NBME11 underpredicted by 4.
UWSA1 I think underpredicted by 9; don't have the score report.
UWSA2 was dead on.
Free120 I just kind of buzzed through on my last day without thinking too much/taking it seriously.

Disclaimer is that I was scoring highly, so there was a very narrow difference in scores across exams.
Wouldn't hang your hat too heavily on any single exam, though.
 
  • Like
Reactions: 1 user
Are UWSA questions different style and quality than those on the QBANK?
 
Members don't see this ad :)
I scored 20 points higher than my last NBME, and 4 points lower than UWAS2. All taken in the last week before my exam
 
  • Like
Reactions: 1 user
UW2 overpredicted me by 8 points (was kind of ticked about this haha). NBME 10 and 11 were pretty close.

The Free120 felt most similar, but of course it doesn’t give you an official score.
 
UWORLD SA1 = 240 (5 weeks out)
NBME 10 = 242 (3 weeks out)
UWORLD SA2 = 256 (6 days out)
Free 120 = 77% 3 days out

Really hoping that SA2 is a strong predictor. Will follow up after I get results.

**EDIT** - actual score - 260
 
Last edited:
  • Like
Reactions: 1 users
Free 120 matches the style of questions most. Took it yesterday and it was a little harder than free 120, but easier than any of the NBMEs. It’s been about a month since I took UWSA2 so I don’t remember it too much. I’m not super thrilled that the next score release isn’t until end of July
 
  • Like
Reactions: 1 user
Free 120 matches the style of questions most. Took it yesterday and it was a little harder than free 120, but easier than any of the NBMEs. It’s been about a month since I took UWSA2 so I don’t remember it too much. I’m not super thrilled that the next score release isn’t until end of July
I know, it's crazy. It will be a 6 week wait time for me :/ even longer for you... absolutely brutal.
 
NBME 9: 265 (baseline)
NBME 10: 266(3 weeks out)
UWSA1: 266 (2 weeks out)
UWSA2: 270 (1 week out)
Actual= UWSA2 on the money.
I think most will score between UWSA1 and UWSA2.
Is there that much backlog that it is taking >4weeks to get your result?
I took it in March on a Friday and got the result 12 days later (2nd Wednesday after the test).
 
  • Like
Reactions: 1 users
NBME 9: 265 (baseline)
NBME 10: 266(3 weeks out)
UWSA1: 266 (2 weeks out)
UWSA2: 270 (1 week out)
Actual= UWSA2 on the money.
I think most will score between UWSA1 and UWSA2.
Is there that much backlog that it is taking >4weeks to get your result?
I took it in March on a Friday and got the result 12 days later (2nd Wednesday after the test).

They are doing something, maybe administering a new test version and need enough people to take the test, but honestly not 100% positive. Either way they said anyone who takes after June 1st won't receive their score until July 26th lol.
 
They modify the question pool every year to try and keep up with the groups of cheaters that compose question banks from everything they remember from the test.
 
  • Like
Reactions: 1 users
Whoever has this sort of memory deserves a reward.
That's the thing, they typically don't. What they report can be completely erroneous and misleading, especially if the reporter got the question wrong as they may not even recall the correct answer choice.
 
  • Like
Reactions: 1 user
Whoever has this sort of memory deserves a reward.

That's the thing, they typically don't. What they report can be completely erroneous and misleading, especially if the reporter got the question wrong as they may not even recall the correct answer choice.

I don't think people recognize the complexity of some of these operations. Groups are arranged to discuss the exams immediately post taking it and are able to corroborate questions that two people independently recalled. These are arranged into banks and the more people that have recalled an individual question, the higher it is rated in the bank. There are also people who have no intention to pursue US residency who go in with the sole intention of memorizing everything they can while also failing so they can maximize lifetime attempts. They are often paid for this. The question banks are either sold or kept internal, which is particularly common in certain international regions/schools. It's not hard to find out how to buy them through telegram, though I strongly recommend everyone NOT to do this as tying your name to this would make you dead in the water should there ever be an investigation into it.

I am very close with an individual involved in the residency app process, and they said they become skeptical of super high step scores (280+) from IMG because they know this practice is rampant in certain areas. Whether that's fair or not is up to you to decide but that's what they told me. They also didn't say they don't rank people because of this, just that they become skeptical, so interpret that as you will.

I am in no way implying every IMG cheats and it actually sucks that the honest ones (most) get screwed over by the cheaters. I'm also not saying this never happens in the US, I have just never heard of such an operation happening, especially anything sanctioned by a US school or admin.
 
Last edited:
  • Wow
  • Like
Reactions: 3 users
There are also people who have no intention to pursue US residency who go in with the sole intention of memorizing everything they can while also failing so they can maximize lifetime attempts. They are often paid for this.

Only medical students are allowed to take the exam though. Why would someone give up their entire medical career for this? Is it because they are planning to do residency in their own country?

EDIT: In 2021, NBME decreased the number of attempts from 6 to 4
 
I don't think people recognize the complexity of some of these operations. Groups are arranged to discuss the exams immediately post taking it and are able to corroborate questions that two people independently recalled. These are arranged into banks and the more people that have recalled an individual question, the higher it is rated in the bank. There are also people who have no intention to pursue US residency who go in with the sole intention of memorizing everything they can while also failing so they can maximize lifetime attempts. They are often paid for this. The question banks are either sold or kept internal, which is particularly common in certain international regions/schools. It's not hard to find out how to buy them through telegram, though I strongly recommend everyone NOT to do this as tying your name to this would make you dead in the water should there ever be an investigation into it.

I am very close with an individual involved in the residency app process, and they said they become skeptical of super high step scores (280+) from IMG because they know this practice is rampant in certain areas. Whether that's fair or not is up to you to decide but that's what they told me. They also didn't say they don't rank people because of this, just that they become skeptical, so interpret that as you will.

I am in no way implying every IMG cheats and it actually sucks that the honest ones (most) get screwed over by the cheaters. I'm also not saying this never happens in the US, I have just never heard of such an operation happening, especially anything sanctioned by a US school or admin.
Damn. Your prediction came true.
 
  • Like
Reactions: 2 users
I'll go against the grain and say I was scoring in the 250s two weeks prior to my exam and got a 80% on the new Free 120 and ended up with a 247. Idk what happened, I didnt even feel nervous during the exam because I was trusting my practice exams.

Comes to show that some can get 80 on the Free 120 and drop to the 240s while others can get a 260.
 
I'll go against the grain and say I was scoring in the 250s two weeks prior to my exam and got a 80% on the new Free 120 and ended up with a 247. Idk what happened, I didnt even feel nervous during the exam because I was trusting my practice exams.

Comes to show that some can get 80 on the Free 120 and drop to the 240s while others can get a 260.
Well the standard deviation is around 15 so that’s reasonable
 
Well the standard deviation is around 15 so that’s reasonable
It’s reasonable, it’s just unfortunate given a lot of stories have upward trends or small score drops. Of course it’s because of sampling bias but nonetheless, it sucks no longer being able to seriously consider a specialty because of this one score
 
  • Care
Reactions: 1 user
It’s reasonable, it’s just unfortunate given a lot of stories have upward trends or small score drops. Of course it’s because of sampling bias but nonetheless, it sucks no longer being able to seriously consider a specialty because of this one score
yeah it’s not a good metric to stratify applicants
 
Top