^are u a mathematician by any chance
If you collect enough of the score matrices… You can extrapolate the highest and lowest Min/Max passing/failing scores…Not that complicated
Not by a long shot
I agree that you can do this, and it’s not highly complicated, but do you really think they can resolve it down to a 4 point spread? There’s definitely a bias in their sample, and again, they don’t mention any validation techniques. If the 300 Hours pass rate was 50% using their MPS versus an actual pass rate of 42%, would you still have faith in their estimation technique (just as an example)? I think someone could more reasonably give you an eight to ten point spread for the MPS, given the data (but maybe not even that narrow).
You also need to remember that we are given qualitative data (that they attempt to turn it into quantitative data), and the results of their analysis hinge on their assumptions , which they haven’t defended. They could be highly reasonable assumptions, but the authors haven’t presented any argument.
So again, I don’t believe anyone has anything more substantial than a grainy, low-resolution answer as to the value of the MPS. This doesn’t seem very different than what you said about (essentially) a rough range.
Here is my opinion regarding pass rates and MPS. Let’s say you get 4/6 on every item set, that gets you to 66.67%, 80 correct. One less, 79 correct, and you’re at 65.83%. My guess is, inclusive of any adjustments, one of these two is the ultimate cutoff dependent on the candidate pool performance and where it falls relative to the 46% historical and 44% 10 year average pass rates. To imply that MPS is driven by revenue is laughable. MPS is driven by the exclusivity of the charter and rightfully so.
There’s no way I got 4/6 on every set. There’s just no way… Guess I’m a retaker.
In the same line of reasoning… there’s just no way 45% of candidates score 4/6 on every item set… Remember that CFAI makes adjustments to the raw score to reflect questions that were not worded correctly, or had multiple correct answers, or where deemed as unfair… That’s part of the reason it takes 6 weeks to grade. I get that some people score 80%+ but realistically 67% is just too high for the 45% that pass… I think the “raw” MPS is closer to 60% and when they report the matrix… It includes the adjusted score
^ completely agree.
seems unreasonable that 45% - essentially half of all candiates score a raw 67%.
i’d say raw mps is more like 62-64% with adjustments bringing up to 67%
I had passed in 2014. I am damn sure i had scored some around 60-62% and no way 67%.Theanalysis of 300 hrs is suffering by self reporting bias.Also there is a lot of debate going on whether cfai decides cut offs with an estimate of % of candidates to pass.I dont think this is true.The same is decided by statistical analysis as well as experts opinion on minimum level to pass. Statistically the population of cfa exam candidates must be similar when they come from same process hence similar pass rates.I think while 67-68% is requirement for l1 for l2 it should be most times less then 65%