my take on the 1998 vs. 2007 NBDE 1

This forum made possible through the generous support of SDN members, donors, and sponsors. Thank you.

USUaggie

bringing ugly back
10+ Year Member
5+ Year Member
15+ Year Member
Joined
Dec 20, 2004
Messages
581
Reaction score
2
Got my results today and there is definitely something fishy going on. I got an 84 and thought, crap, it's just an 84 (I was shooting for over 85 which I thought was average). But then I got looking closer and saw that I got 15.1 more questions right than the national mean. That's 3.75% more which should equate to a couple scaled points, so, I think it's safe to assume that the national average so far is 82 or really close to it.

Now it gets weird - on the back they have the breakdown of where they expect the scores to be and it says that the average should be an 85.
81-83 should be the 39.4th percentile,
84-86 should be the 60.6th percentile, and
87-89 should be the 79.1st percentile.

It just doesn't add up to me. I took it in late June so I think the national mean scores on my report are pretty representative of how the exam will end up. I just emailed Dr. Tsai to point out the discrepancy to him.

Oh, by the way I was really annoyed when everyone here kept saying there scores would have been 5 points higher using the 98 scale when nobody actually took the 98 exam and compared results. I took the 98 exam a few days before the real thing and got an 87. So, there is consistency in that I've been +2 from the national average on both exams.
 
Got my results today and there is definitely something fishy going on. I got an 84 and thought, crap, it's just an 84 (I was shooting for over 85 which I thought was average). But then I got looking closer and saw that I got 15.1 more questions right than the national mean. That's 3.75% more which should equate to a couple scaled points, so, I think it's safe to assume that the national average so far is 82 or really close to it.

Now it gets weird - on the back they have the breakdown of where they expect the scores to be and it says that the average should be an 85.
81-83 should be the 39.4th percentile,
84-86 should be the 60.6th percentile, and
87-89 should be the 79.1st percentile.

It just doesn't add up to me. I took it in late June so I think the national mean scores on my report are pretty representative of how the exam will end up. I just emailed Dr. Tsai to point out the discrepancy to him.

Oh, by the way I was really annoyed when everyone here kept saying there scores would have been 5 points higher using the 98 scale when nobody actually took the 98 exam and compared results. I took the 98 exam a few days before the real thing and got an 87. So, there is consistency in that I've been +2 from the national average on both exams.

I know you aren't to happy with your score, but nevertheless congratulations for passing, do keep us inform on what Dr. Tsia has to say about this matter....yeah that break down on your chart doesnt add up.
 
I agree with that 1998 comparsion talk being presumptuous. It has been clearly demonstrated that the averages of individual sections vary year to year, especially in the case of DAO. Which if you were examine past averages, could be anywhere from low 60's to mid 70's.
 
hopefully you get a response back from mr t that satisfies you, but here is what i think. there is a big difference between average and percentile. The reason your scores are above average and your percentile is not is because of that difference. When someone bombs the test and gets 40% of the ?'s correct that brings down the average tremendously, but when you look at percentile that person is still only one person and won't effect the percentile that much. on the other hand because the average is relatively high, when someone gets 90% of the questions correct (i think this probably happens less than someone getting 40%) it doesn't effect that average that much because that average is already pretty high. I did a really crappy job explaining that but hopefully that makes sense why your percentile and average don’t seem to be matching up.
 
hopefully you get a response back from mr t that satisfies you, but here is what i think. there is a big difference between average and percentile. The reason your scores are above average and your percentile is not is because of that difference. When someone bombs the test and gets 40% of the ?'s correct that brings down the average tremendously, but when you look at percentile that person is still only one person and won't effect the percentile that much. on the other hand because the average is relatively high, when someone gets 90% of the questions correct (i think this probably happens less than someone getting 40%) it doesn't effect that average that much because that average is already pretty high. I did a really crappy job explaining that but hopefully that makes sense why your percentile and average don't seem to be matching up.

Thanks for pointing this out to me. It took me a while but I get your point now. Say on a math test there were 20 people who got 86%, one 84%, and 5 who only got 10%. The mean score is 71% so the guy with 84 gets all worked up about being in a lower percentile when he's clearly beat the average score. (That's the example I had to think of to help me understand your point).

However, I really doubt this is the case with this exam. According to the percentile chart, the same number of people fail as get 96-99 - that's 2.9% in each case. If only 2.9% are getting a raw score that's somewhere below 200 (estimate), it would be pretty hard to bring the mean down that much. There were lots of easy questions so to get an abnormally low raw score (below 100) somebody would have to really not have a clue what they're doing or else be intentionally getting questions wrong. (my apologies to anyone reading this who got a raw score below 100).

Like I said, your explanation is a possibility because the numbers could come out that way, I just don't think that's what happened here. I would bet it has to do with what was discussed in this thread:
http://forums.studentdoctor.net/showthread.php?t=400563

The scoring of the exam is based on a SCALE and not a CURVE. Therefore, your score does not depend on how other students do. The scale for this exam was determined by the test makers, based on pilot exams and previous board exams. So essentially we are using a scale based on the curve of previous exams. That explains why people are scoring lower.
 
Got my results today and there is definitely something fishy going on. I got an 84 and thought, crap, it's just an 84 (I was shooting for over 85 which I thought was average). But then I got looking closer and saw that I got 15.1 more questions right than the national mean. That's 3.75% more which should equate to a couple scaled points, so, I think it's safe to assume that the national average so far is 82 or really close to it.

Now it gets weird - on the back they have the breakdown of where they expect the scores to be and it says that the average should be an 85.
81-83 should be the 39.4th percentile,
84-86 should be the 60.6th percentile, and
87-89 should be the 79.1st percentile.

Hi ,
I have the same problem ,
I did my exam late may and I got 85,
I'm an international dentist and I want to apply, is there any hope with this score for me?
thanks
 
Thanks for pointing this out to me. It took me a while but I get your point now. Say on a math test there were 20 people who got 86%, one 84%, and 5 who only got 10%. The mean score is 71% so the guy with 84 gets all worked up about being in a lower percentile when he's clearly beat the average score. (That's the example I had to think of to help me understand your point).

However, I really doubt this is the case with this exam. According to the percentile chart, the same number of people fail as get 96-99 - that's 2.9% in each case. If only 2.9% are getting a raw score that's somewhere below 200 (estimate), it would be pretty hard to bring the mean down that much. There were lots of easy questions so to get an abnormally low raw score (below 100) somebody would have to really not have a clue what they're doing or else be intentionally getting questions wrong. (my apologies to anyone reading this who got a raw score below 100).

Like I said, your explanation is a possibility because the numbers could come out that way, I just don't think that's what happened here. I would bet it has to do with what was discussed in this thread:
http://forums.studentdoctor.net/showthread.php?t=400563

i am in los angeles ,how long does it take to get our test results back?
 
Last edited:
Top