So the national average went up on this exam...

  • Thread starter Thread starter 78222
  • Start date Start date
This forum made possible through the generous support of SDN members, donors, and sponsors. Thank you.
7

78222

Doesn't this affect those of us who took the exam before the new test was administered? The average score on my exam was 218, now according to those who recently got their scores back it is 222. Won't this essentially mean that those of us who took the earlier test will have our scores "devalued"?`
 
Doesn't this affect those of us who took the exam before the new test was administered? The average score on my exam was 218, now according to those who recently got their scores back it is 222. Won't this essentially mean that those of us who took the earlier test will have our scores "devalued"?`

No I don't think this an accurate assumption. The standardization of the scores assures that a 240 on an exam taken 5/1/08 equates to the same level of competence as a 240 on an exam taken 6/15/08. The average is a reflection of the scores made by a certain cohort.

If the standardization is doing its job then the increasing avg means that for the cohort with the higher average the mean level of competence was greater than the cohort with the lower average.
 
No I don't think this an accurate assumption. The standardization of the scores assures that a 240 on an exam taken 5/1/08 equates to the same level of competence as a 240 on an exam taken 6/15/08. The average is a reflection of the scores made by a certain cohort.

If the standardization is doing its job then the increasing avg means that for the cohort with the higher average the mean level of competence was greater than the cohort with the lower average.

Sure, thats the idea of having a standardized test. However, it seems since they've significantly changed the exam recently by shortening it and adding these audio/visual questions, it's very difficult to claim it is still standardized to the previous exam.
 
Sure, thats the idea of having a standardized test. However, it seems since they've significantly changed the exam recently by shortening it and adding these audio/visual questions, it's very difficult to claim it is still standardized to the previous exam.

I see your point. But, I think in the end the change in format will not alter the way scores measure competence. Residency directors won't care whether the test was the "old" or "new" format. And programs will look at a 240 today the same way they did last year and all previous years (there may be more 240s to look at however).
 
Where is this information available? I can't seem to find any kind of general performance data past 2007...
 
I could be wrong, but I've previously been told that the 2 digit number more accurately compares how an individual performed in comparison to all previous examinees (years back). Whereas the 3 digit score compares you to your specific cohort.

Has anyone else heard this?
 
yea i dont know where to find it on the web, but I have some close friends who took the exam over the last couple years who showed me their reports.

The average and SD are in the score reports. I was looking at last year's data and for first attempt the data are the same mean of 222 SD 22. This comes from the report the NBME provides for each school the national data includes all exams for all of 2007.
 
I could be wrong, but I've previously been told that the 2 digit number more accurately compares how an individual performed in comparison to all previous examinees (years back). Whereas the 3 digit score compares you to your specific cohort.

Has anyone else heard this?

I doubt this is true. When the NBME decided to make the test harder to pass, they adjusted the 3 digit passline, but could not adjust the 2 digit score because by regulations its passline must always be 75. The 3 digit score is specifically designed to facilitate comparison between people of different cohorts; the 2 digit score only exists to meet the requirements of a few licensing boards that can't handle interpreting a 3 digit score.
 
I read the little paragaph talking about the 222 average and it says 1st time taker american and canadian medical student average. Maybe they arent including fmg, img, and do students (and they definitely arent including retaker scores either) in that average which might bring the overall average to around 218 which is in line with what most people say was the old average? so it might not have really changed at all. im just speculating though.
 
I doubt this is true. When the NBME decided to make the test harder to pass, they adjusted the 3 digit passline, but could not adjust the 2 digit score because by regulations its passline must always be 75. The 3 digit score is specifically designed to facilitate comparison between people of different cohorts; the 2 digit score only exists to meet the requirements of a few licensing boards that can't handle interpreting a 3 digit score.

Couldn't it exist for both reasons? It's definitely done for licensing reasons, since 75 is the minimum passing, but couldn't it also be used as a comparison among all examinees, historically? Any statisticians out there? I'm not implying that 90 means you're in the 90th percentile historically.

Eh, I have no idea.
 
The standard for passing has changed, so you can't make a comparison. For instance, someone who scored 75 10 years ago would have failed today with the same performance (and must score lower than 75)
 
I read the little paragaph talking about the 222 average and it says 1st time taker american and canadian medical student average. Maybe they arent including fmg, img, and do students (and they definitely arent including retaker scores either) in that average which might bring the overall average to around 218 which is in line with what most people say was the old average? so it might not have really changed at all. im just speculating though.

the old average (218) was also only for 1st time taker american and canadian medical students. fmg and img are synonymous terms.
 
Last edited:
Top