-
CycleTrack and SDN are teaming up to make medical school admissions more transparent and accessible! Read the announcement to learn how we’re supporting the future of CycleTrack’s free application-tracking tools .
You are using an out of date browser. It may not display this or other websites correctly.
You should upgrade or use an alternative browser.
You should upgrade or use an alternative browser.
Percentile correlation with Percent Correct
Started by docjohn101
Get help with your application
Use all the free resources available to you from SDN: articles, guides, expert advising, forums discussions, and school research.
bump...anyone?
Also, anyone know how generous the curve is the the family medicine shelf or about how many points it add to your percent correct or how it is calculated?
Also, anyone know how generous the curve is the the family medicine shelf or about how many points it add to your percent correct or how it is calculated?
Does anyone know the correlation between the percentile and percent correct?
For our clerkships, we only need to be in the 4th percentile to pass, what percent correct does that correlate to?
The two are basically entirely unrelated. The tests are scaled quarterly with everyone else in the country who takes the exam during that time period. So if you take a shelf as the first block, you will have many more questions you can miss and get a good score. If you take it last, you won't have as big of a margin.
Don't worry about it. Do the best you can on every test and just try to get as many questions right as possible. Let the chips fall where they may.
Advertisement - Members don't see this ad
Does anyone know the correlation between the percentile and percent correct?
It is a positive correlation.
On my shelf score PDF, it states:
The test is scaled with a mean of 70. We have no idea what our raw scores are, so it's hard to say at minimum how many questions you need. If you apply statistics, you need a scaled score of ~56 or so to make 4th percentile comfortably.
NBME said:"The subject examination score is scaled to have a mean of 70 and a standard deviation of 8 for a scaling group of first-time takers from U.S. LCME-accredited medical schools who took this examination as a final clerkship examination under standard testing conditions. As a result, the vast majority of scores range from 45 to 95, and although the scores have the look and feel of percent-correct scores, they are not. This scale provides a useful tool for comparing your performance with that of a nationally representative group taking this examination as an end-of-clerkship assessment."
The test is scaled with a mean of 70. We have no idea what our raw scores are, so it's hard to say at minimum how many questions you need. If you apply statistics, you need a scaled score of ~56 or so to make 4th percentile comfortably.
Unfortunately they scaled those scores like 20 years ago, the averages are now quite higher...I think the average score for Psych for the quarter I took it in was like 85 or something.On my shelf score PDF, it states:
The test is scaled with a mean of 70. We have no idea what our raw scores are, so it's hard to say at minimum how many questions you need. If you apply statistics, you need a scaled score of ~56 or so to make 4th percentile comfortably.
My med school uses percentile minimums for clerkship honors. They mentioned what my score's percentile was, and it directly lines up with basic stats. Doesn't matter if the average is an 85% correct. The scale will recenter that at 70.
Then why was a "raw" score of 80 on the last psych NBME considered equivalent to the 53rd percentile, if the distribution is centered around a raw score of 70...? That should be around a 70th percentile based on the average * 70 plus/minus SD of 8 * system described above. My math senses are tingling...
The world may never know.
This conversion is simply one of the great mysteries of medical school that doesn't in any way have to be a mystery but is.
This conversion is simply one of the great mysteries of medical school that doesn't in any way have to be a mystery but is.
Then why was a "raw" score of 80 on the last psych NBME considered equivalent to the 53rd percentile, if the distribution is centered around a raw score of 70...? That should be around a 70th percentile based on the average * 70 plus/minus SD of 8 * system described above. My math senses are tingling...
The distribution isn't centered around a raw score of 70. The Shelf reported scores are scaled so that a score of 70 correlates to the 50th percentile of test-takers. In other words, if the raw score of the 50th percentile of test-takers is a 78, that 78 gets rescaled so that your NBME reported score is 70.