How is the Step I scored?

[Alexander99]

---

Members don't see this ad.

Just how do they determine your Step I score? I've done a search but they never seem to explain how they determine your score based on your performance on the test. I know the test has 350 questions but your score isn't your raw score, is it?

What's the maximum you can get on the Step I? What's considered a highly competitive Step I score (comparable to a 35+ on the MCAT). Thanks for any info you can share.

 

[Jalby]

Nobody knows exactly how it is scored. Judging by Q-bank scores, it you get 50-60% of the questions right, you will pass, if you get ~70% right you get about a 230.

Of the 350 questions, 50 of the questions don't count and are used for future tests, they just want to see how people do on the questions. Nobody knows if it is one block of 50 or if they are scattered around.

As for MCAT scores and USMLE Scores, studies have shown that your verbal score most correlates with the score you will recieve on the boards. I find that to be pretty interesting, but I did read the study. The maximum that you can get on the boards is about 280. It varies from year to year. Generally, the highest score someone at a school will get is 270ish.

As far as competative is, I've heard that a 230-235 will get you interviews at almost any program except the most competative. I used to know the averages of the people who go into some of the most competative programs here, but I forgot. I want to say 237 avg for Ophtho.

 

[vinoyp]

As for MCAT scores and USMLE Scores, studies have shown that your verbal score most correlates with the score you will recieve on the boards. I find that to be pretty interesting, but I did read the study. The maximum that you can get on the boards is about 280. It varies from year to year. Generally, the highest score someone at a school will get is 270ish.

Just curious... How do you know 280 is the max as opposed to 300?

--Vinoy

 

[Stinger86]

I've heard of someone making a 288 one year, but I can't provide the evidence to back it up, so you'll have to draw your own conclusions

 

[Alexander99]

So it seems to me that if there are 350 questions and 50 of them don't count, your Step I score is just your raw score out of the 300 questions that counted. So if you scored a 270, that means you got 270 out of 300 correct (90% correct). Could it really be that simple?

 

[ajl102]

My favorite theory is that 300 = 100%.

This kind of correlates to the 75% Q-bank = 230 USMLE relationship.

I don't really think those kind of statements about the Qbank % can be very meaningful on an individual basis though, makes your score seem too predetermined. Most of my friends outscored that prediction by at least ten points. My Q bank percentage was fairly crappy, not for a lack of effort either, something like 60%. Ended up getting a 238. My scores were pretty hit or miss on the practice exams - you could say I hit the USMLE at my peak performance, I guess, but I think it depends largely on how worked up you can get over a practice exam, the "shape" of your learning curve, Kaplan's spotty quality control, etc.

 

[AlexRusso]

Your USMLE score is directly related to the number of posts you have on SDN. They take the number of posts you got fit you on their curve (which in my mind looks something like a hemoglobin binding curve) and wham -- you got your score.

 

[Alexander99]

Your USMLE score is directly related to the number of posts you have on SDN. They take the number of posts you got fit you on their curve (which in my mind looks something like a hemoglobin binding curve) and wham -- you got your score.

Sounds pretty easy. So that guy with the 270+ must have been on here 24/7 to get that score.

 

[Idiopathic]

With an average of 200 and an SD of 20 (neither is right) you would have to be in the top 2% just to get a 240. I think that anyone who tells you that they 'knew someone' who got a 280+ or that they themselves got a 280+ is a liar. The most legitimate score I heard reported was a 272...someone from UCSF. Even that I find hard to believe. There are lots of scores clustered in the 240-255 region, I bet, with very few above that.

 

[Jalby]

The avg is 216 with a STV of 20. Our top score at our school for the graduating fourth years was a 272 with a guy going into nuerosurgery.

 

[Idiopathic]

The avg is 216 with a STV of 20. Our top score at our school for the graduating fourth years was a 272 with a guy going into nuerosurgery.

You get the idea...a 272 is still not +3 SD from the mean, so to imagine someone getting a 280(!) is unreal. I bet you could get every question right and not get much higher than 280.

 

[OneStrongBro]

Jalby, that is BS. How in the world can the Verbal score predict a step 1 score that is based on the basic sciences. Step 1 has a high correlation to biological science. Think about it. If someone scores a 13-14 on biological science they are more likely to get that on Step 1.

Students that get 13-15 on the MCAT verbal tend to be students that have a nonscience major. If you are arguing that nonscience majors score higher on STEP 1 than science majors you are clueless. I say show the article on here, and stop BSing.

As for the Verbal, it has the highest correlation to M3 and M4 where you are doing your clerkships, and have to integrate stuff on your feet rather than on an exam that is heavily based on the basic sciences.

Biological Section subscore predictive of Step 1.
Verbal Score more predictive of Clerkship grades and performance.

Nobody knows exactly how it is scored. Judging by Q-bank scores, it you get 50-60% of the questions right, you will pass, if you get ~70% right you get about a 230.

Of the 350 questions, 50 of the questions don't count and are used for future tests, they just want to see how people do on the questions. Nobody knows if it is one block of 50 or if they are scattered around.

As for MCAT scores and USMLE Scores, studies have shown that your verbal score most correlates with the score you will recieve on the boards. I find that to be pretty interesting, but I did read the study. The maximum that you can get on the boards is about 280. It varies from year to year. Generally, the highest score someone at a school will get is 270ish.

As far as competative is, I've heard that a 230-235 will get you interviews at almost any program except the most competative. I used to know the averages of the people who go into some of the most competative programs here, but I forgot. I want to say 237 avg for Ophtho.

 

[Jalby]

Wait a second. You are back OneStrongBro??? I thought we had proved that you are a liar and not the many different people you said you are. Oh, wait, we did. Check out good monkey and Nuclearrabbit at the bottom of this thread:

http://forums.studentdoctor.net/showthread.php?t=103368

yeah i was wondering what was up with onestrongbro also. i know that umich curriculum isn't subject based. on top of that he has alot of pre-allo posts that describe him on the interview trail during the 2002-2003 year. that would make him an M1 right now.

post with him describing interview in late 2002.
http://forums.studentdoctor.net/showthread.php?s=&threadid=48142&highlight=onestrongbro

wait, i thought you went to michigan?
http://forums.studentdoctor.net/showthread.php?s=&threadid=97104&highlight=umich
and by this post, it looks like you're a 4th year? you linked to the class of '04 and you acted as though you knew some of the people in their facebook and on that site. 😕

in any case, umich doesn't have discreet classes like biochemistry, pharmacology or gross anatomy - especially 2nd year. gross (and maybe biochem, but i think that may have been lumped into MCB 1st year, if you are indeed a 2nd year right now) was a class, but is no longer, as it is integrated into the systems-based new curriculum and runs concurrently all year...

uh, right now, if you are a 2nd year and if you are at michigan, you are in your neurology sequence, not taking pharm as a discreet course, and would have just finished up, what, GI, with a big batch of standardized patient exams in there? and if you're at umich, why does your location say chi-town?

i'm confused.

anyway, umich is pass/fail - NO honors, NO grades, NO rank first year. 2nd year moves to honors/pass/fail, and rank is established. 3rd and 4th year are h/hp/p/f.

 

[Alexander99]

Wait a second. You are back OneStrongBro??? I thought we had proved that you are a liar and not the many different people you said you are. Oh, wait, we did. Check out good monkey and Nuclearrabbit at the bottom of this thread:

http://forums.studentdoctor.net/showthread.php?t=103368

Human encyclopedia and now troll slayer. This man does it all folks! Jalby for 2003-2004 board MVP. 👍

By the way, good job sniffing out the troll. I'm sure he/she/it won't be back for, oh, about 3-4 months. I actually find someone that's not in med school but pretending to be a med student on an anonymous internet message board to be somewhat sad, and I don't find too many things in life to be sad.

Where's SnS when you need him? He knows how to troll in style.

 

[Jalby]

lol. Thanks Alex. But I didn't sniff him out, I just unburied the sniffing. On a side note, I have on person who is a troll with two fake names and I'm just waiting for he/she to slip up to get proof.

 

[OneStrongBro]

You can deflect the question all you want. BUT THE BOTTOMLINE IS that you haven't proved that VERBAL REASONING is the most accurate gauge of Step 1 performance. That is counterintuitive. Just answer the question, and post your "source". Then we can move on. If you can't answer this QUESTION everyone that reads your post will think you are a BS'er. Because all you are doing is deflecting the question.

THE BALL is in your court. Just list the article, or give us a link. That is all I ask. If you don't your credibility is shot.

 

[Jalby]

You are calling out the credibility of me?? Wow. Simple enough, I have been very credible on issues like this in the past, but right now I don't have time to go search for the article that I read about the subject. Unlike one of us who always lies about who they are, I am actually an MS II and do have a ton of board studying to do thansearch through hundreds of articles.

 

[Jalby]

Here is a link to the thread where I had the link of the article posted up, but apparently it didn't even work back then. I'll see if I can find a different link.

http://forums.studentdoctor.net/showthread.php?t=120027

 

[Smurfette]

A couple comments:

I've NEVER heard of any VALID correlation between any part of the MCAT and Step I performance. It's pretty much comparing apples to oranges. Non-science majors may have a steeper learning curve when first starting med school, but by the time you take the boards the playing field btw science and nonscience majors has been levelled assuming everyone's passed all coursework. If my bio (or verbal) MCAT score correlated to the USMLE, my score would be very different; keep in mind that MCATs are taken by many people who ultimately are not admitted into med school, so the USMLE-ers are a specific sub-population of MCAT-ers with an overall higher average score. I have heard that scoring at least 8 on MCAT sub-sections IS correlated with higher likelihood of PASSING the USMLE and that is why med schools generally won't accept students who get less than an 8 on a given subsection even if they performed very well on the other subsections. Plus the amount and quality of time spent studying for each test is significantly different, IMO. There are rumors that scoring X on [insert rumored test here--Q-bank, MCAT, SAT, etc.] correlates to USMLE score=Y and so forth, but they never seem to actually hold true in real life, everyone just has "heard it somewhere".

I know someone in my class who got a 269. Rumor has it, someone else scored even higher. According to my test results, it says on the bottom of the score report that scores ranged from 142ish to 288---I remember commenting on how unbelievable a 288 is and wondering what kind of super-genius got that kind of score. A 288 is so high I would think residency directors would be double-checking the validity of that score (and frankly, someone with that kind of score is a walking textbook, which may make them difficult to work with and somewhat less desirable. But that's my opinion. Scoring 20 points above the other truly brilliant med students who scored in the 260s is a whole other level.).

Let's flip this question around---what is the LOWEST step I score you've heard of? I've heard that the vast majority of failers fail by only a handful of points, but that if you score below 165 or so, your likelihood of passing it the second time is very low, and you only get three chances total.

 

[Jalby]

bl2seq

Well said. I hate irrational people. They make it sooo much harder for the rational ones.

 

[bl2seq]

A couple comments:

I've NEVER heard of any VALID correlation between any part of the MCAT and Step I performance. It's pretty much comparing apples to oranges. Non-science majors may have a steeper learning curve when first starting med school, but by the time you take the boards the playing field btw science and nonscience majors has been levelled assuming everyone's passed all coursework. If my bio (or verbal) MCAT score correlated to the USMLE, my score would be very different; keep in mind that MCATs are taken by many people who ultimately are not admitted into med school, so the USMLE-ers are a specific sub-population of MCAT-ers with an overall higher average score. I have heard that scoring at least 8 on MCAT sub-sections IS correlated with higher likelihood of PASSING the USMLE and that is why med schools generally won't accept students who get less than an 8 on a given subsection even if they performed very well on the other subsections. Plus the amount and quality of time spent studying for each test is significantly different, IMO. There are rumors that scoring X on [insert rumored test here--Q-bank, MCAT, SAT, etc.] correlates to USMLE score=Y and so forth, but they never seem to actually hold true in real life, everyone just has "heard it somewhere".

I'm sorry, but you are 100% wrong. The MCAT has a very high correlation with your Step I score. In fact, it is quite a bit more predictive than your undergraduate gpa. The correlation is actually .69 with your Step I score. Here is the actual study, which used the conglomeration of the 1993 and 1994 entering classes from 14 US med schools:

http://www.aamc.org/students/mcat/research/bibliography/julia001.pdf

I don't know why so many future physicians have a such a pathetic fetish with rejecting statistics and fixating on individual and flukish anecdotes. You can probably find someone in the last ten years who had an 8 verbal and earned a 240 USMLE, but the overwhelming majority of native English speakers who got an 8 verbal and actually managed to get admitted to med school had sub-average (likely, grossly sub-average) USMLE scores.

 

[Idiopathic]

Let's flip this question around---what is the LOWEST step I score you've heard of? I've heard that the vast majority of failers fail by only a handful of points, but that if you score below 165 or so, your likelihood of passing it the second time is very low, and you only get three chances total.

The vast majority of failers fail by a handful of points because that is what the standard curve looks like. That doesnt mean that one question would put anyone over the top necessarily. Are you sure you only get three chances? I thought that was school-specific. Our school only gives you three chances (you have to repeat the second year after failing twice), but I didnt think that was universal.

 

[Smurfette]

bl2seq: 🙄

1. The MCAT has changed quite a bit since the '93/'94 entering students have taken it. They changed how the questions are structured and also added the lame writing sample section as well. For several years after this change (possibly still going on, but I don't know as I am several years out from my own MCATs), the highest score on the verbal was/is "13-15" b/c there aren't enough hard questions on the new version to truly delineate the 13s from the 15s in a statistically significant manner, which makes it harder to try and correlate scores.

2. The USMLE has also been changing over the past several years. Ten years ago, a score above 230-240 was not as "easy to achieve" (I use this phrase lightly) as it is today and scores 250+ were just about unheard of. Now, PDs are using scores of 240 as cutoffs for interviews in certain competitive programs. My point here is that the score ranges have been widened as a result of changing the level of difficulty of the questions, which has enabled them to further determine different high scores as being truly different from one another. The same number of students may be getting 230+, but the highest score may be 274 now instead of 250. The overall average score on the USMLE has creeped up over the years as well. The focus has been shifting more and more toward patient-centered vignettes and less and less off the basic science rote memorization-type questions. The authors of First Aid have recognized this and are trying to 'tweak' information presented in FA as a result.

IF someone got in with borderline MCAT scores, yes, odds are that they will not do as well on the USMLE as someone with MCAT=38. (there may be a difference in abilities to take standardized tests between the two people as well as basic fund of knowledge). But will MCAT=38 get a significantly different USMLE score that MCAT=35? I doubt it. A lot of the USMLE is luck in what your exam focus is. If your weakness is pharm, and you get a test really heavy in pharm, your score may be a bit lower than if your test was heavier in micro.