Does the UWorld Self Assessment Overestimate Score?

[wannabie]

So here's the thing, I'm about one month out from CK. I took the UWorld Self Assessment a few months ago when I was 20% through UWorld (I know, I know, should have checked here: it's better to keep it for the last month)--and I got a 225, which a couple months out is completely fine by me. THEN I finished Uworld averaging 60% and am going through it a second time averaging 85%--both not great, but not the end of the world. However, I took NBME 6 a month ago and got a 204 when I was 75% through UWorld. THEN I took NBME 3 and I got a score unheard of on these boards--a 187. I'm raging here because I don't know what's going on.

With UWorld questions, I can use physical dx and the labs and the history to actively think through the question and determine the answer. I KNOW I'm terrible at straight up memory recall questions and NBME's have a lot of them, but I'm freaking out. I'm a month out; I'm going to work as hard as I can, but can anybody give me any advice? What the heck's going on here?

---

Members don't see this ad.

 

[normtheniner]

I had the opposite experience. NBME 4 before intense studying was 248 (550 I think?)... took UWSA after one week of studying and got a 240 predicted.

I'd like to think NBME is more predictive in my case although know you're hoping the opposite. I think the main point is you can't hinge everything on a practice test score and really just try to reflect on how "ready" you really feel.

 

[Bagheera]

UWSA is very predictive for step 2ck. It was 1 point off for me, and if you look at the threads, it seems 90% it is very accurate. NBME 4 is also accurate. Taking your NBME 4 score and UWSA may give you a better prediction. NBME 2 isn't bad either, but was a little harder than NBME 4 for me. The real deal felt like NBME 4 or slightly easier.

 

[thedoctor8706]

This is the question I have about UWSA 1 for Step II CK: Does it's accuracy in predicting scores correlate with how high you scored? Digging through threads, it appears that the higher you scored on the UWSA, the more accurate it gets, and even higher still, it over predicts (like in 260 range). But for those that score say in 220's or 210's, it seems to under predict when compared with NBME exams.

I felt USWA 1 was WAY harder than NBME 6. I answered 85% of questions correct on NMBE 6, equating to a 480 (233), and thus below the mean. However, with UWSA 1, I answered only 63% of questions correct, but still scored a 500 even (219), which is technically the mean, and yet it predicted over 10 points lower than the NBME did. I'll take NBME 4 on Sunday and see what happens. My exam is next Thursday. I've used UWorld (62% over the whole year used as a learning tool in timed tutor mode) and Secrets for review, after having done fairly well on shelf exams all year long (80's, with a bit lower on surgery).

I'm just not sure I trust the UWSA exams in terms of predictability; it said I'd get a 197 on Step 1, but I scored 30 points higher than that on the real deal. I just don't think their formula for coming up with a predictive score works well, though their questions seem more on par with real questions when compared to NBME exams. Go figure.

I know a lot of you high scoring players will say it's really good predictively, but what do some of the more average test takers think? Any other input?