how are scores calculated?

[rachmoninov3]

---

Members don't see this ad.

Sorry if this has already been posted--I did search, honest!

Anyway, how are the scores for the USMLE calculated? According to Kaplan's test bank, a 50% is 211, 60% 226, 70% 242.
Basic math tells us that 186/350 is 53%

I've got 5 months, and I'm anywhere from 40 to 60% depending on which qbank I'm using. Any advise on how this is actually calculated would be helpful, so I know how much time to spend now, or if I should just wait until this last semester is over and then really focus for 6 weeks.

thanks

 

[DrTennis]

Sorry if this has already been posted--I did search, honest!

Anyway, how are the scores for the USMLE calculated? According to Kaplan's test bank, a 50% is 211, 60% 226, 70% 242.
Basic math tells us that 186/350 is 53%

I've got 5 months, and I'm anywhere from 40 to 60% depending on which qbank I'm using. Any advise on how this is actually calculated would be helpful, so I know how much time to spend now, or if I should just wait until this last semester is over and then really focus for 6 weeks.

thanks

Skim this article. It's a bit dated, but it some info on how USMLE Step 1 and 2 scores are calculated.

http://www.academicmedicine.org/pt/...PGkVpGfZGFxTG1mgm!383905440!181195628!8091!-1

 

[psychobob]

I've seen anywhere from 50% to 70%, but that's just for passing. There's also an unknown amount of 'experimental' questions which apparently are not counted. There's just too many unknown factors, so It's probably best not to think about it.

I'm assuming the scores you presented from Kaplan are just simply 'estimates' of your score based on Qbank performance. I have the qbook, and they just present the passing range of 60-70%.

 

[tarsuc]

Im sorry im bumping this old thread, but im trying to actually figure this out.
I understand how scores are calculated(See the following example).


[Lets take a theoretical student called Bob, who scored 242/98 on USMLE Step 1.

Let's take Bob's 3-digit score of 242 and subtract the mean:

242 - 216 = 26

Bob is 26 points above the mean.

Now divide this number by 1 Standard Deviation:

26 / 24 = 1.08

This number 1.08 is called the z-score (pronounced zee score). It tells you the number of standard deviations above or below the mean that Bob's score was. He was 1.08 SDs above the mean.

Now we have to use a z-score statistical table, which is a constant comparison table. You can use any z-score table in any statistics book.

Read to where it lists the number 1.00 then read accross to where it adds 0.08 (= 1.08).

In this column it says 0.8599 which is roughly ~0.86

So now Bob knows he is in the 86th percentile. This means that in comparison to the recent test takers in US/Canada, he scored better than 85% of people, and only 14% of test takers scored better than him. Well Done Bob!!]


but does this mean that our performance on each individual qn does not matter, and all that matters is the total 3 digit score? I used to think, that if we get a qn wrong that everyone else gets right, then our score would drop drastically.

Any inputs?