Who fails this exam?

Looking at the comments on the forum, I do not see who fails this exam…

Granted the sample is skewed to the right, but everyone (80%) seems to be getting mock scores in the mid 70s, have done virtually every exam since 2004 and get mid 60s (and are probably harsh on their grading, from what they are saying…). So the question is, who fails this exam?

  1. The people who run out of time - probably only 10-20% of them are in the above sample

  2. People who had a bad day… 5-10%?

  3. People who have done very little, but show up anyway… 5%?

  4. Under-reported people in the forums who don’t speak much because they are not doing that great on the prep… 20?

How many AFs are in total candidates population? I was lurking on prior years result reveal topics and had impression that about 80 % of AFs had passed level 3. Maybe my impressions are wrong.

Flashback if I were a betting man, I’d put money on you passing.

More likely for someone who passed to post his results…

People who leave their calculator in BGN mode and go into the AM and PM exam. 1%.

Be cool if someone who is taking it easy today went back and looked at last years mock results thread and then ran a comparison against the exam result thread…

Backfill bias :wink:

Unless all of the questions were around calculating future value of insurance premiums…

Thank you. We’ll take a virtual beer once we pass.:slight_smile:

Survivorship bias?

Everything is biased. Your real results will be biased too.:slight_smile:

Easy to understand. 1)AM grading is subjective, we have no idea. 2) Personally I finished about 3-5 minutes early on the AM I took (3 3 hour timed), add in crunch time and they gap closes to 1-0 minutes easy on exam day. 3) We all have weaknesses, wrong set of combo we get rocked.

I think best assumption is just take the avg pass rate and apply it to people on the forum. Because the majority of people register to AF don’t post or if they fail don’t say they failed. Definitely skewed to the right, also having PM a few charterholders who still contribute to the forum recommended not to be active on AF. Also my 3 Day schweser Tutor suggested the same thing saying 80% of the things on AF aren’t relevant because people keep straying away from what’s tested.

Don’t know about the forum but I know two types of losers in the exam entrance:

  1. Brought a wrong calculator

  2. Is reading notes while presenting passport.

About half of you will fail.

#goodluck

the way i figure it is that within a normal distribution… MOST candidates are scoring between 55-70% (a very tight range)

people in band 10 literally miss the MPS cuttoff by 1-2 multiple choice questions… band 9 3-4 questions and so on

the better question that is unknown and up for debate is what % of first timers/ retakers pass the exam…

i think it is something ike 70% pass rate for retakers and closer to 40% for first timers

the 40/60/80 analysis shows interquartile range of 60-68% (i.e 64% is a pass).

i.e. each band 5-10 is about 1%, or maybe less than 1% (say 4minutes), so every 4minutes you drop on an IPS is dropping a band…

good luck everybody!

I think I can help here. It would be nice if the distribution were normal, but I’m pretty sure it’s not. 1. Other exams like the bar exams that publish the full distributions aren’t close to normal. They are typically skewed. Almost no one gets close to a perfect score, but some real losers do awful. Thus, the mean is less than the median. 2. Though imprecise, the 300 hours distributions and the resource teepee scatter plots don’t look like normal distributions.

For every other professional exam that shows pass rates for first time and repeat takers, the repeat takers have lower pass rates, usually much lower.

I do think it is a pretty narrow passing band. However, in addition to weighting ethics heavier for marginal students, something else looks like it is going on. The 300 hours histograms make me suspect that the topic weights get Reweighted. I have at least three possible reasons: 1. People who do really bad in ethics are getting penalized. 2. Some questions are getting thrown out. Topics which were supposed to be x% of the exam are less than the stated %. This wouldn’t surprise me, since I’ve heard that a few questions might be tossed in a given year. 3. The minimum passing score under the Angoff method is not a single number, like 67%, but has something else, like higher weighting for some questions or topics. This is really interesting, but not supported by the Institute’s description of the process. The only place the Institute says it does something like this is in the Level 3 AM. If I were designing the scoring system, I would do this at all levels.

Did not know that. Wonder how well that predicts L3. Can you say that those professional exams have something like L2 preceding them, where a lot of serious people are weeded out? I’ve assumed the L3 exam isn’t a “weeding out” exam. I.e., if you have passed L2, you’re basically going to pass L3, provided you a) invest the hours to go deep on just about everything in L3; b) get a lot of experience writing AM papers under exam conditions. My position now is: I tried to do that, and if I fail, I’ll just have to do more of that.

Would love to know whether I’m wrong, and whether L3 truly does weed out people who have the skill to pass L2, but lack some special sauce needed for L3. I didn’t detect anything like that – other than one’s comfort in some nuances of English (which is outside of the curriculum).