Exam Feedback November 2016 Part 1 Exam Feedback

[Nicole Seaman]

Just saw on LinkedIn
-----------------------------------------------------------------------------------
November 2016 FRM Exam pass rates -- Part I: 44.8% | Part II: 54.3%

Thanks. That's the lowest ever Nov passing rate for Part II and the second lowest ever Nov passing rate for Part I.

I just wanted to point out that David has updated our historical pass rate graph and posted in the forum here: https://www.bionicturtle.com/forum/threads/what-is-the-pass-rate-for-the-frm.10093/#post-47498.

Nicole

 

[perviz]

Hi Nicole. Thanks for info. Is there difference Nov or May session. what is purpose of grouping may and nov exam? is nov harder than may or vice versa?
thanks

 

[David Harper CFA FRM]

@MWDerbyshire if you see Nicole's link to our updated graph, I don't quite see the recent pass rates as the lowest/second lowest. For P1, I see four exams with worse pass rates, and for P2 I see three exams with worse pass rates (my count begins on May 2010). I do observe that both P1 and P2 are at least 2.0% below the long-term average.

@perviz The syllabus does not change from May to November which leads to a theory that November takers might do better because they have more time to prepare for the given syllabus (see my note here https://www.bionicturtle.com/forum/threads/may-2016-part-1-exam-feedback.9609/page-13#post-43494), but that "November bump" did not happen for P1 takers this time

 

[brin_page]

Failed
4-2-2-3

 

[berrymucho]

@MWDerbyshire if you see Nicole's link to our updated graph, I don't quite see the recent pass rates as the lowest/second lowest. For P1, I see four exams with worse pass rates, and for P2 I see three exams with worse pass rates (my count begins on May 2010). I do observe that both P1 and P2 are at least 2.0% below the long-term average.

@perviz The syllabus does not change from May to November which leads to a theory that November takers might do better because they have more time to prepare for the given syllabus (see my note here https://www.bionicturtle.com/forum/threads/may-2016-part-1-exam-feedback.9609/page-13#post-43494), but that "November bump" did not happen for P1 takers this time

I think @MWDerbyshire was referring specifically to the November pass rates. That is correct, the latest (Nov) P2 pass rate was the lowest November pass rate since the split in 2010. Besides dipping under the long-term average overall, the "seasonality" of the pass rates might be diminishing.

 

[David Harper CFA FRM]

@berrymucho ahh, yes, I see thank you! Yes, agreed, it looks like when isolating only within November exams @MWDerbyshire is exactly correct

 

[deepak.batra]

Hi All, I passed FRM Part I.........1,1,1,1 ........just bought part 2 materials from GARP...will register for part 2 in May-2017 for Nov-2017 exams since I do not want to spend USD 150 for extending the date to Nov-17 by registering now for May-17!...anyhow.......my question is that I do not see end of chapter problems in part 2 materials of GARP...is it the case actually or am I missing something....thanks for your reply...

 

[chintanudeshi]

Hi, Would anyone in this forum be able to give a rough estimate on the total number of qualitative and quantitative questions which were asked in the L1 Nov 2016 attempt? I guess this would just give a better about the trend, to those candidates who are appearing for L1 in May 2017. Thanks!

 

[emilioalzamora1]

First of all I do not (and suppose my peers neither) remember (I am not counting questions during the exam) how the split was between qual. vs. quant questions (I took level II but this assumption holds for Level I as well). A rough guess is 50/50.
Even if one question seems to be qualit in nature, it could be the case that it involves quant logic/thinking. And what is more, what happened in the past (Nov 2016) may not be a good indicator of the future.

 

[punx120]

Failed. Got 3,3,2,2.

How can you fail with 3322 ? someone else passed with 3422 itn the same exam ?? I am preparing for May 2017, I was just checking the result for Nov 2016

 

[David Harper CFA FRM]

@punx120 I shared a worksheet over at https://www.bionicturtle.com/forum/threads/quartiles-and-weights-exam-results.9883/#post-45570 with an extreme example comparing a hypothetical student who generates quartiles of 3rd/3rd/2nd/2nd compared to a student who scores 2nd/2nd/3rd/3rd, yet it's possible for the former to pass with 62 but the latter to fail with 36 (!); i.e.,

@hellohi

Okay I built this small spreadsheet (because trying to talk about this soon becomes a word soup imo), please see https://www.dropbox.com/s/ikhbkm0571fdchh/1019-garp-frm-scoring-exam.xlsx?dl=0
... and below is a snapshot. This is just for P1 and you only input (change) the yellow cells, the rest is calculated.

Notice how I input an extreme version of the one you referenced in your email; i.e., Student #1 earns 3/3/2/2 and passes while Student #2 earns 2/2/3/3 but fails. I was deliberately provocative: notice how the seemingly subtle difference can lead to a difference between a final score of 62 and 36 (wow!). Caveat: I'm not sure my quantiles are exactly calibrated, but they can't be too far off. I hope that clarifies!

 

[punx120]

Thank you David, I completely understand your example, it is different.
We have 3322 and 3422.
They have exactly the same result, only the result in Quants are différents .
The guy who failed has better result (3) in quants than who succeed (4).
The results are the same everywhere else.
How that possible?

 

[emilioalzamora1]

Well, actually you just get the quartile but no one knows the upper or lower bound and whether this has any impact on the pass/fail outcome.
Let's say it could be the case that two candidates have the same quartiles (for a partcular section, say 'Fin. Products') and one fails while the other one passes. This could stem from the fact that the one who fails had quartile results disproportionately at the lower bound of all four (or majority of) quartiles. And I think here is the difference. There must a difference between candidates who score at the upper bound of the 1st quartile (being among the top 10 candidates) against someone who is in the 1st quartile as well but very close to the 2nd quartile.

 

[David Harper CFA FRM]

@punx120 It's exactly what @emilioalzamora1 says, in my opinion; and his hypothesis is interesting! Here is the same XLS but I quickly added your example but artifically motivated to maximize the discrepancy: https://www.dropbox.com/s/29g80b6rqdhsbi5/0517-garp-frm-scoring-exam.xlsx?dl=0 (screenshot below). I am able to artificially create a gap of 17 points between the fail and unexpected passer, although interestingly, I did need to lower the pass rate to 57; i.e., it is unexpected to me that a 3/4/2/2 is a pass. However, as we've discussed on this forum a lot, there is not a knowable ex ante passing score (including, it would be variable from exam to exam). The pass is calibrated after scores are tabulated. But when a 3/4/2/2 passes, that at least suggests a very difficult test, with the prospect for passing in spite of getting less than 60% of the questions correctly answered. Thanks,

 

[punx120]

It is very clear now. Thanks a lot David.