Place to discuss CAS Exams. Their construction, administration, grading, etc.
Please remember that this overall process is done by volunteers and let’s be respectful of their willingness to give up their time to administer the process.
NOTE: I’ve moved a large block of discussion from an Exam 7 discussion. If any need to be moved back, please notify my of which post needs attention (you can also report it with a statement of moving it back to the Exam 7 discussion).
Exams hard. Likes at the bottom right please.
I can 100% guarantee they didn’t do this 
Do you really think the CAS is some evil organization out to keep people from getting their credentials?
More like a flawed organization that lacks transparency & execution, but sometimes it’s easier on the brain to simplify the name-calling 
They say they are very transparent.
They’re not evil. They’re just selfish and as SkolChicago said lacking in execution. I’ll add competency to the list of things they’re lacking in. Possibly ethics too. Can’t convince me those pass rates and artifical twists they put in exams aren’t manufactured based on how many people they want in the profession.
Their idea of transparency is to just repeat to us what their process is, which is not transparency to me.
I don’t think they’re targeting set numbers, but I do think they write questions to make the exams artificially hard, rather than differentiating the qualified vs not qualified actuaries.
I think that’s the motivation behind not releasing exams, which is really frustrating TBH.
I should show you my appeals letter which goes on for two pages saying how the appeals process is so transparent and then there is a one liner, your appeal was not accepted.
The CAS management sees “growth of credentialed CAS actuaries” as a sign of their success. I can absolutely guarantee they are not attempting to artificially restrict the number of newly minted ACAS or FCAS. Also, most of the leadership is older, senior in their area, and hope to hire new fellows – their jobs aren’t threatened by new fellows, quite the contrary.
Are there problems with execution? Sure.
It’s actually very hard to tell how difficult a new question will be. I once wrote a question that I was quite proud of, that I thought would get at understanding of a fundamental concept. But it turned out that I worded it badly, and a reasonable interpretation of the question led to an easy solution that showed at best shallow understanding. I had to give full marks for both answers. 
Misjudgements happen in the other direction, too.
But the goal of not releasing the questions is to build up a bank of proven questions of known difficulty. Questions that were “clean” and tested what they were supposed to test and are unambiguous and didn’t generate any complaints can be re-used in the future.
The downside is less feedback, both for the candidates and for the CAS. But the upside is potentially much better, more predictable, reliable exams in the future.
Having graded, and dealt with a few appeals… By far the most common appeal is “I did X. Why didn’t I get credit?” To which the response is “That is not a valid appeal.”
My appeals were not like this.
A few examples:
We realize the CAS is a non-profit, volunteer based organization right? I think selfish is incredibly harsh.
People have complained for ages that we don’t have multiple upper level sittings, now the CAS is doing that in the only way that makes sense within the constraints of a volunteer employee base and people are pissed at how it’s being done. Damned if you do, damned if you don’t.
Volunteer 1 creates a question that is reviewed by volunteer 2. Are you volunteering to fund the $200+/hr we’d have to pay for the level of competency you’re describing?
After I get FCAS, I’d gladly make exam questions at half that rate
Ask someone on the SOA side if they have to deal with the same issues regarding the exam process. As an example, SOA was faced with the same COVID circumstances and yet had their results release date scheduled far in advance. Not to mention their exams based on feedback from past candidates are far less controversial and more intelligently worded.
While I’ve not written exam questions for the CAS, I have done so for academic settings (college math class with common finals). I will tell you that it’s easier said than done . . . especially when you need to factor in how others who don’t think like you are likely to read the Q.
And it’s this latter part that I think that the CAS doesn’t put enough effort into for their calibration. More specifically, the main part of their calibration is to take uber-testers as the benchmark for whether or not an Exam is “too long” . . .
Yeah, that makes sense. I completely agree the CAS has a tendency to err on the side of the exams being too long. “This person with 200 IQ could finish the exam in 3 hours easily, so 4 should be plenty of time for most candidates!”
I will say the CBT environment made time much less of an issue this sitting imo.
Maybe the issue is also that uber test takers are also the ones creating the questions… I am not sure I would want Amp to create the questions. Amp is way too smart…