I think the CFAI wants the average candidate to slightly overestimate the Minimum Passing Score. The CFAI makes public the figure of 70% times the average score of the top 1% but does not disclose how well the top 1% does or what the Minimum Passing Score was in the past. It’s possible that top 1% of candidates only averages 90%, in which case the Minimum Passing Score would be just 63%. There are two reasons that the CFAI wants the average candidate to slightly overestimate the Minimum Passing Score: 1. This encourages all candidates to aim high (70%+) and to learn the material well. The CFAI doesn’t want candidates to settle for mediocrity. 2. It’s good PR. The CFAI would rather have candidates think they failed and end up passing rather than have candidates think they passed and end up failing. Thus, most of the candidates who fail in band 10 are those who weren’t that confident to begin with, so they are encouraged to try again rather than give up in disgust. As a result of these policies, the CFAI minimizes discontent and the number of candidates who feel like they were ripped off. It’s a win-win for everyone - CFAI, CFAs, candidates, and test prep providers.
Interesting observation DVI, I hope you’re right. I have this theory that the MPS and # of candidates who pass is also dependent on how well the other two levels did. Think about this: July 28 = Level 1 results (just under 2 months to grade 240 M/C questions) August 18 = Level 2 results (about 2.5 months to grade 120 M/C questions) I think Level 1 & 2 score’s grading are completed at the same time. But the # of candidates in Level 1 who pass is adjusted (e.g. ethics adjustment, etc.) by how well those in Level 2 do. Level 2 & 3 are probably done the same way. They both release results on August 18, which makes sense for Level 3 due to the nature of grading the essay portion of the exam. Then the # of candidates in Level 2 who pass is adjusted by how well those in Level 3 do. Am I absolutely out in left-field here?
Both are very interesting points, and both sound like they could be plausible.
Interesting observations but I would highly doubt the minimum score is less then 67. There are so many people that take this exam, many of which are actuaries, PM’s, Analysts who have many years of experience and know this stuff like the back of their hand. I would not be surprised if the top % of scores is around 95.
If 128,000 examinees are supposed to appear then lets assume that 100,000 sat for the exam. 1% of that would be 1,000 people. So what do you think, 1000 people can get 95%? I dont think so.
But if I’ll score <=50% in one area (suppose Derivatives or Alternative Investments) I have some chance to pass if I will score 51-70% in other areas?
Ares77 Wrote: ------------------------------------------------------- > But if I’ll score <=50% in one area (suppose > Derivatives or Alternative Investments) I have > some chance to pass if I will score 51-70% in > other areas? Assuming the Minimum Passing Score is 75% (random number), you just need 75% on average to pass. Doesnt matter if you score 50% is a given section. This is what I understood.
kh.asif Wrote: > > Assuming the Minimum Passing Score is 75% (random > number), you just need 75% on average to pass. > Doesnt matter if you score 50% is a given section. > This is what I understood. Thank you.
@ Diehardvalueinvestor Nice thinking. I tend to beleive that MPS should be some percentage of maximum score in an exam. For me its difficult to beleive that top 1 percentage of serious candidates, can score more than 95% in all. So assuming 70% of 95% would mean 67% of total 240 questions leading to 160 questions in all should be correct, which effectively means 2/3rd of the questions should be ok.
I am also the belief the MPS is roughly around 65%. Diehardvalueinvestor summed up some of the reasons well. But if you really think about it and work out some general averages for those “top 1000” you begin to realize that the agnoff method would force the average below 70%, assuming that the CFAI doesn’t use Agnoff + Minimum Cutoff. It’s good news in one way but in the other its potential dilution waiting to happen. I would rather have a MPS of 80%, and a fewer amount of charter holders, then an MPS of 65% and everyone grabbing a piece of the pie.
Can we conclude from agnoff method that all questions don’t hold the same weightage in grading ?
4tay, I think that is a little too much conspiracy theory for me, it doesn’t make sense. The more I thought about it, the more I actually realized it doesn’t matter how many below average people take the test. A lot of people on here were all happy talking to people who didn’t know anything thinking they were helping the curve when in reality they likely did nothing to influence it at all. Think about this: If they score the test strictly based off of the old method of 70% of the top 1% of scores we are only competing against the top 1% people, no one else, meaning that if the top 1% get 95% it doesn’t matter that 60% of the people may get 50% on the test. If it is based off the Angoff method which seems to be consensus, we are again competing only against what the CFA institute deems to be an acceptable passing score, not based on a curve determined by unprepared candidates. If poor test takers really influenced it there would be a much higher passing rate because it would normally distribute and you would likely have 50-55% being the average(maybe not just using it as an example). So I think all we can hope for is either that the top 1% score at best 95(if this still factors into it) or that the CFA institute feels the test is harder than normal and sets the MPS lower. just my opinion of course. Ben
legacy01 Wrote: ------------------------------------------------------- > 4tay, I think that is a little too much conspiracy > theory for me, it doesn’t make sense. haha… I tried. That’s the one main thing I could think of for the big difference between Level 1 & 2 result date release.
I would think it is more likely due to the difference in formats and how long it takes them to establish a MPS, but that is just me. 