If no one has ever gotten a 100%, what do you think the top score ever on the LI is? 1,025,407 people took the exam before yesterday. With that many, given that no one gotten the 100%, it is extremely unlikely that there are many scores over 98%
I think they just say that you make us feel relaxed. I am sure there are a few people every year who get a perfect score. Its a multiple choice exam! You are telling me no one has been able to get all 240 questions correct?
thought the same thing so 70% of top marks cant really be >95
Even if you only guess on 10 questions the odds of you getting a perfect score are less then 1 in a million.
f2d Wrote: ------------------------------------------------------- > Even if you only guess on 10 questions the odds of > you getting a perfect score are less then 1 in a > million. The exam is not a four sided die where the probability of getting a certain sequence of answers (240 to be exact) is 1 in 10 million. If you happen to have studied the sections pertaining to the exam question you have a 100% probability of getting the answer correct. Based on that I think if you studied enough you should be able to get a perfect score.
I’m saying if you guess on just TEN questions the odds of getting those 10 right is less then 1 in a million. How many test takers here guessed on 10 questions or more? I probably took educated guesses at half the questions. And for some of the questions, even if you had the book in front of you and infinite time, you still have to guess.
ZeroBonus Wrote: ------------------------------------------------------- > f2d Wrote: > -------------------------------------------------- > ----- > > Even if you only guess on 10 questions the odds > of > > you getting a perfect score are less then 1 in > a > > million. > > > The exam is not a four sided die where the > probability of getting a certain sequence of > answers (240 to be exact) is 1 in 10 million. > > If you happen to have studied the sections > pertaining to the exam question you have a 100% > probability of getting the answer correct. Based > on that I think if you studied enough you should > be able to get a perfect score. I think if that were true than someone would have done it by now
f2d Wrote: ------------------------------------------------------- > I’m saying if you guess on just TEN questions the > odds of getting those 10 right is less then 1 in a > million. How many test takers here guessed on 10 > questions or more? I probably took educated > guesses at half the questions. > Your analysis is good. Consider the consequences of your logic. If you are saying that the BEST test takers get 230/240 cold, the highest typical score should be 232.5/240. That must make the 99% ~228/240 (95%) or so. Pretty hard exam!
ZeroBonus Wrote: ------------------------------------------------------- > f2d Wrote: > -------------------------------------------------- > ----- > > Even if you only guess on 10 questions the odds > of > > you getting a perfect score are less then 1 in > a > > million. > > > The exam is not a four sided die where the > probability of getting a certain sequence of > answers (240 to be exact) is 1 in 10 million. > > If you happen to have studied the sections > pertaining to the exam question you have a 100% > probability of getting the answer correct. Based > on that I think if you studied enough you should > be able to get a perfect score. Huh? Your statement doesn’t really dispute his argument.
Isura, how did it go?
which brings me to another point. How many do you think you guessed? im thinking about 20-30 questions i had NO idea at and guessed. id say about 30-40 that could have gone either way - hopefully thats enough to get me through!
f2d Wrote: ------------------------------------------------------- > I’m saying if you guess on just TEN questions the > odds of getting those 10 right is less then 1 in a > million. How many test takers here guessed on 10 > questions or more? I probably took educated > guesses at half the questions. > > And for some of the questions, even if you had the > book in front of you and infinite time, you still > have to guess. Your analysis assumes that you are completely incapable of recognizing any of the incorrect answers. But isn’t CFAI changing to only 3 options next year, because almost nobody actually picks the 3rd distractor? For a lot of the questions, at least 2 answers were obviously wrong (2 increase vs. 2 decrease, for instance - you know it must increase, even though you don’t know how to do the math). If you can eliminate 2 possible answers on the questions, then the odds are just 1:1,024 - very plausible.
Yes, I made the assumption that there were 10 Qs that you have no idea of the answer to. But it was purely for an example. The majority of test takers do NOT know 230 of the 240 questions like the back of their hand. Most people are guessing on at least 20 - 30 questions Assuming you narrow your choices down to 2: 2^10 is 1024 2^20 is > 1 million 2^30 is > 1 billion How many people know the material so well that they only have to guess between 2 choices on 10 problems? Maybe 1 or 2 a year maximum. Sure, if they give out the exams for long enough, sooner or later someone will get a perfect score. But given the qty of material and the qty of questions, it’s not surprising that nobody’s ever gotten a perfect. Even if someone has worked in the investment profession their whole life, they wouldn’t come close to a perfect on this. The CFAI questions/answers are not all the real life equivalent.
TheAliMan Wrote: ------------------------------------------------------- > Isura, how did it go? Pretty sure I passed. But not too confident about the PM score. How about you Ali?
f2d Wrote: ------------------------------------------------------- > Yes, I made the assumption that there were 10 Qs > that you have no idea of the answer to. > > But it was purely for an example. The majority of > test takers do NOT know 230 of the 240 questions > like the back of their hand. Most people are > guessing on at least 20 - 30 questions > > Assuming you narrow your choices down to 2: > > 2^10 is 1024 > 2^20 is > 1 million > 2^30 is > 1 billion > > How many people know the material so well that > they only have to guess between 2 choices on 10 > problems? Maybe 1 or 2 a year maximum. > > Sure, if they give out the exams for long enough, > sooner or later someone will get a perfect score. > But given the qty of material and the qty of > questions, it’s not surprising that nobody’s ever > gotten a perfect. > > Even if someone has worked in the investment > profession their whole life, they wouldn’t come > close to a perfect on this. The CFAI > questions/answers are not all the real life > equivalent. I think the this analysis is useful to get a sense of how hard it is to get a perfect score. Remember, that all of this also assumes that all problems that you did answer were answered correctly… no misreads, misinterpretations, etc… So all that, AND you guessed just right on the remainder. It’s pretty hard to get 100%, and I’m not too surprised that no one has done it.
Even if you know the material inside and out it’s pretty much impossible to get a perfect. After staring at these questions all day, everyone’s bound to make mistakes.
F2d, that’s one of the more simple yet insightful analyses I’ve seen on the subject.
If most people, say >70%, get a question wrong is the question then removed and the total score lowered?
Can I just add, that I think it states on the CFAI website that there has never been a perfect score. So assuming that NOONE has every got 100%, I think it would be quite a small probability also that anyone would get 99% or 98%?
Well, I think it’s not that hard to get the perfect score. There was only question for which I didn’t know where the actual answer would be stated in the books. Don’t get me wrong, I had to guess quite a bit too, but at least I knew where to look up the answer. And I think there are many guys out there who studied really hard and got the 100%. But again, I just wasn’t willing to study a year for that thing and therefore I will be miles away from a perfect score (and maybe from a pass too).