Math and the new MBE format - A sneaky way to increase the cut score? Forum

[supa_mitsu]

I know this has been kind of mentioned but my technical point would be off topic in other threads, so here we are.

I was reading this article from Above the Law (see below) and the author says that even if the scored questions went from 190 to 175, "the difference in scores was expected to be negligible". This also seems the general consensus on the internet in general. How so? Now we have only 175 chances to hit our state cut score instead of 190 chances! For example, if in California I needed to get a 144, I could have got 46 of the scored questions wrong, but now I can't afford to make more than 31 scored mistakes. Doesn't that make a HUGE difference? We are talking 15 raw points.

I know that some of you will say "no, because you are not considering that now you have 25 unscored questions that you can get wrong instead of 10", but does that matter? If it does, then I suppose that if the unscored questions would be 50, getting a 144/150 would still be as hard (but clearly, it looks much harder).

The case is either (1) everybody ignored this and we are getting an even harder MBE to pass, or (2) I am missing a technical mathematical nuance. Meanwhile, the MBE scores hit a new low last February.

Please, give me your input and remember we are all lawyers here, it's ok to repulse math.


Here's the article from ATL:
http://abovethelaw.com/2017/04/mbe-scor ... -time-low/

[blaze1306]

I know this has been kind of mentioned but my technical point would be off topic in other threads, so here we are.

I was reading this article from Above the Law (see below) and the author says that even if the scored questions went from 190 to 175, "the difference in scores was expected to be negligible". This also seems the general consensus on the internet in general. How so? Now we have only 175 chances to hit our state cut score instead of 190 chances! For example, if in California I needed to get a 144, I could have got 46 of the scored questions wrong, but now I can't afford to make more than 31 scored mistakes. Doesn't that make a HUGE difference? We are talking 15 raw points.

I know that some of you will say "no, because you are not considering that now you have 25 unscored questions that you can get wrong instead of 10", but does that matter? If it does, then I suppose that if the unscored questions would be 50, getting a 144/150 would still be as hard (but clearly, it looks much harder).

The case is either (1) everybody ignored this and we are getting an even harder MBE to pass, or (2) I am missing a technical mathematical nuance. Meanwhile, the MBE scores hit a new low last Febryary.

Please, give me your input and remember we are all lawyers here, it's ok to repulse math.


Here's the article from ATL:
http://abovethelaw.com/2017/04/mbe-scor ... -time-low/

Seems like funny math to me...175 from 190 seems to be less margin for error to me.

[supa_mitsu]

Seems like funny math to me...175 from 190 seems to be less margin for error to me.

I agree, that's why I'm so surprised that nobody seems to have noticed... I'm a very critical thinker, so I'm just waiting for somebody to tell me why I am wrong.

[lawhopeful100]

My understanding is the your scaled MBE score accounts for the number of questions discarded. So even if you have less questions to get a higher raw score, your scaled score should not change.

[happyhour1122]

I don't know all this number calculations, but only that matters to me is....I just lost 25 questions I could have scored.

[dionysius1]

because it is percentages.... You have 15 less chances to score, but you also have 15 less questions to possibly get wrong. The 175 are weighted slightly higher than the 190, so it all evens out in the end...

[supa_mitsu]

because it is percentages.... You have 15 less chances to score, but you also have 15 less questions to possibly get wrong. The 175 are weighted slightly higher than the 190, so it all evens out in the end...

So you are saying that basically, if before (with a raw score of around 134) you used to get around 10 extra points, now you will get more?

[happyhour1122]

because it is percentages.... You have 15 less chances to score, but you also have 15 less questions to possibly get wrong. The 175 are weighted slightly higher than the 190, so it all evens out in the end...

ohhhhhhhhhhhhh
are you a law student? Good math!

[dionysius1]

It is my understanding that it is a weighted average. All questions are not given the same weight either. I was told that there are some questions, that based on their tracked level of difficulty, are weighted differently. If an extraordinate number of people get one wrong, it is thrown out as a bad question. Your final score is a number based on the number correctly answered with their weighted average score. While the total may be 10 more than you got correct, that is a fluctuating number, not set in stone.

[InterAlia1961]

My calculations show that to get to the 70% threshold with the new 175-question format, you have to get 123 questions correct before any scaling. On another thread, someone purported to have learned that for the Feb. 2017 MBE, the magic number is 119 correct questions. When the number of questions that counted was 190, my math shows you had to get to 133 (70%) correct questions before scaling in order to pass. So, with the new format, you have a better chance of passing. You only have to answer 123 unscaled questions correctly as compared to 133. So, as long as the MBE wasn't more difficult this time, you have a better chance of passing it than you did last July. That's my story, and I'm sticking with it. I can't math any more today.

[Estecontre]

I know this has been kind of mentioned but my technical point would be off topic in other threads, so here we are.

I was reading this article from Above the Law (see below) and the author says that even if the scored questions went from 190 to 175, "the difference in scores was expected to be negligible". This also seems the general consensus on the internet in general. How so? Now we have only 175 chances to hit our state cut score instead of 190 chances! For example, if in California I needed to get a 144, I could have got 46 of the scored questions wrong, but now I can't afford to make more than 31 scored mistakes. Doesn't that make a HUGE difference? We are talking 15 raw points.

I know that some of you will say "no, because you are not considering that now you have 25 unscored questions that you can get wrong instead of 10", but does that matter? If it does, then I suppose that if the unscored questions would be 50, getting a 144/150 would still be as hard (but clearly, it looks much harder).

The case is either (1) everybody ignored this and we are getting an even harder MBE to pass, or (2) I am missing a technical mathematical nuance. Meanwhile, the MBE scores hit a new low last Febryary.

Please, give me your input and remember we are all lawyers here, it's ok to repulse math.


Here's the article from ATL:
http://abovethelaw.com/2017/04/mbe-scor ... -time-low/

I'll tell you why you're wrong. You're doing a straight swap and doing basic math. Assuming 190 questions are in play in order to achieve the 144 "needed" you're just subtracting 190 minus 46 in order to get 144. If you were to miss the 46 questions you wouldn't get a score of 144, each question isn't worth one point to begin with (the maximum points possible is 200 not 190). If there were 200 questions in play then yes, they would each be worth one point, but since there's is less questions technically each question should be worth more than one point. Even if each questions is worth just a small decimal point more you don't get 144 if you only missed 46 questions. Let's say each question is actually worth 1.05 points. Multiply that by the number you got right (144) and you come out with 151.2. If your magic number was to try and hit 144 raw then you would actually only need to get 137 questions right leaving you with the possibility that you could miss 53 questions in totally and still hit the 144 score range. Moreover, even if you did just miss 46 questions there is no way your score would be 144 because the MBE isn't based on raw score but a scaled score so it would technically be higher than 144 depending on the average.

Now since there are less questions in play, technically speaking each questions is worth more points. Again, even if it's a small decimal point more it matters. Let's say that the questions are now worth 1.15 because there are less questions that count but the maximum possible score stays the same. Now in order to get the 144 needed you only need to get 125 correct in order to get a 143.75 which would more than likely be scaled to beyond the threshold needed.

Even I'm making the math too simple and assuming that it's just basic math.

[ur_hero]

The way the math works out for me, assuming 119 is the "pass" score....

Before: 128/190= .674 "passing".

Compared to now: 119/175 = .68 "passing".


Regardless, doesn't scaling just account for it? The NCBE and Calbar told me scaling fixes everything. Right?

[ur_hero]

I know this has been kind of mentioned but my technical point would be off topic in other threads, so here we are.

I was reading this article from Above the Law (see below) and the author says that even if the scored questions went from 190 to 175, "the difference in scores was expected to be negligible". This also seems the general consensus on the internet in general. How so? Now we have only 175 chances to hit our state cut score instead of 190 chances! For example, if in California I needed to get a 144, I could have got 46 of the scored questions wrong, but now I can't afford to make more than 31 scored mistakes. Doesn't that make a HUGE difference? We are talking 15 raw points.

I know that some of you will say "no, because you are not considering that now you have 25 unscored questions that you can get wrong instead of 10", but does that matter? If it does, then I suppose that if the unscored questions would be 50, getting a 144/150 would still be as hard (but clearly, it looks much harder).

The case is either (1) everybody ignored this and we are getting an even harder MBE to pass, or (2) I am missing a technical mathematical nuance. Meanwhile, the MBE scores hit a new low last February.

Please, give me your input and remember we are all lawyers here, it's ok to repulse math.


Here's the article from ATL:
http://abovethelaw.com/2017/04/mbe-scor ... -time-low/

I get where you're trying to go with this, but that's some pretty bizarre math. Sorry, haha.

[Estecontre]

I know this has been kind of mentioned but my technical point would be off topic in other threads, so here we are.

I was reading this article from Above the Law (see below) and the author says that even if the scored questions went from 190 to 175, "the difference in scores was expected to be negligible". This also seems the general consensus on the internet in general. How so? Now we have only 175 chances to hit our state cut score instead of 190 chances! For example, if in California I needed to get a 144, I could have got 46 of the scored questions wrong, but now I can't afford to make more than 31 scored mistakes. Doesn't that make a HUGE difference? We are talking 15 raw points.

I know that some of you will say "no, because you are not considering that now you have 25 unscored questions that you can get wrong instead of 10", but does that matter? If it does, then I suppose that if the unscored questions would be 50, getting a 144/150 would still be as hard (but clearly, it looks much harder).

The case is either (1) everybody ignored this and we are getting an even harder MBE to pass, or (2) I am missing a technical mathematical nuance. Meanwhile, the MBE scores hit a new low last February.

Please, give me your input and remember we are all lawyers here, it's ok to repulse math.


Here's the article from ATL:
http://abovethelaw.com/2017/04/mbe-scor ... -time-low/

I get where you're trying to go with this, but that's some pretty bizarre math. Sorry, haha.

It's not bizarre. It's just over-simplified and way too basic.

Sorry I don't mean for that to come off as insulting, but you over-simplified the math and are not taking into account other factors.

[supa_mitsu]

Yes, I absolutely over-simplified. It wasn't really clear to me that questions don't have the same "point value". Thanks everybody for your input, I haven't had such an a-ha moment since the day I understood what's a measuring life in the rule against perpetuities.

[Bluem_11]

It may increase the variance since the sample size is smaller, i.e., you'll have slightly larger or smaller pass rates than before, but not the cut-off grade. However I have no idea how the examiners scale things, so it's possible that is accounted for.

The math works out the same. Yes a wrong question counts for more wrong, but a right question counts for more right.

I swear I heard same logic from my LS classmates who preferred big classrooms because there were more A's to go around even though it was still only 5% no matter the class.

[dionysius1]

I may be aging myself, but there was a movie called "Spinal Tap" where the lead singer in the band loved his new amp because the volume level on most amps only went from 1 to 10. His new one went from 1 to 11...so it must be louder.

https://www.youtube.com/watch?v=4xgx4k83zzc

[supa_mitsu]

I may be aging myself, but there was a movie called "Spinal Tap" where the lead singer in the band loved his new amp because the volume level on most amps only went from 1 to 10. His new one went from 1 to 11...so it must be louder.

https://www.youtube.com/watch?v=4xgx4k83zzc

Lol, speaking of over-simplifying

[Estecontre]

I may be aging myself, but there was a movie called "Spinal Tap" where the lead singer in the band loved his new amp because the volume level on most amps only went from 1 to 10. His new one went from 1 to 11...so it must be louder.

https://www.youtube.com/watch?v=4xgx4k83zzc

I know it's kind of an old movie, but wasn't it meant as a mockumentary so it's supposed to be funny that his amp goes to 11? Could've sworn I heard this on Howard Stern. . .

[udonisandtrinity]

The cut-off remains the same. The path to reaching that cut-off becomes harder, albeit it is very negligible, as each question carries more weight to the overall scaled score.

The issue is how they will manipulate the 25 experimentals to making the exam more difficult. They could include more of the easy questions based on exam results into the experimentals thereby taking away points previous admin test takers scored. Lots of different ways to screw test takers with this increase in experimentals.