CFA Level 1 – some observations & guesstimations….

Question

Some interesting results from guesstimating exam weightings versus length and depth of syllabus.....

(1)  20% of readings account for 41% of the test...

Note that this might not actually be the case.  It's based on some rough guesstimating, so read into it if you wish.

The result is based on the idea that if you assign an exam weighting per-reading (i.e., the # practice questions vs. total topic practice questions, multiplied by the subject weighting) you find that the "top 20" readings (from the total of 63) account for an estimated 54.6% of the tested content.

Using this over-simplified logic, the "top 12" readings (i.e., ~20% of the CFAI readings) account for 40.8% of the test.

What's even more interesting is that the areas with highest weightings are predominantly conceptual, and less formula and calculator-driven.

I'm not about to suggest that focusing all energies on these areas is the key to any form of success, but I did find it interesting.

FYI "Top 5" Readings (in chronological order):

#2 - Ethics: Guidance to Standards (***deadly boring, but by far the biggest marks for the reading material)
#7 - Quant: Statistical Concepts & Market Returns
#54 - Fixed Income: Intro to Fixed Income Valuation
#55 - Fixed Income: Understanding Fixed-Income Risk & Return
#63 - Alternative Investments: Intro to Alternative Investments

(2) The length of readings doesn't indicate exam values.

Of course this goes without saying, but for the heck of it I took a look at how many pages of reading is required for each 1% (rough estimate) of exam importance (using reading weightings above versus pages of reading).

This doesn't mean to avoid these areas, as some of the reading is long but really basic.  But it's the kind of stuff you can manage with a beer in the hand without worrying that you're missing something.

The largest volume of page-to-exam weighting ratio was by far Reading 42 (Portfolio Management: An Overview).  With the CFAI material it's 270 or so pages of material for what might be 1% of the total exam.  Second was Reading #5 (Quant: TVM).

The 12 readings with the largest volume of reading-to-exam weighting account for 11.8% of the exam and around 1,400 pages of reading (so about 40% of the CFAI Level 1 curriculum).

Again, I ran these numbers out of interest only.  Obviously some of this high-volume stuff is easy to skim through, or is required knowledge to actually understand other readings that build upon them.  Personally I am going to be using this as a guide to areas that I might (i.e.: should) be able to speed-read (1,400 @ 3ppm; ~ 8hrs) and get it out of the way.

edulima · Answer

Very interesting, @mitch895, thank you for sharing your analysis.

I found some of your conclusions very unexpected (and sometimes a bit bizarre!) For example, two of the five "most valuable" readings come from a Fixed Income study session and one is an Alternative Investments reading.

If you don't mind I'd like to ask you a few questions:

1) When estimating contribution to each reading to the test, what kind of sample did you use as far as the test goes? Was it the mock test provided by the CFAI, the volume of EOC problems, or something else?

2) I'm having a bit of trouble matching some of your numbers. For example, Reading 42 (Portfolio Management: An Overview); you mentioned 270 pages of reading in the CFAI materials; however, in my count I have only 30 pages (excluding the summary). How did you come up with 270? Note that in my count, the grand total number of pages (again, excluding summaries & EOC problems) is about 2,540.

Anyway, it is an interesting exercise the one you've done (in its own right, it gets you some familiarity with the structure of the body of material, the reading titles, etc.), and you have very fairly added enough caveats to render its spirit as in "trust these results at your own risk". So depending on the reliability of the results, I'd caution anyone who may think about using these conclusions to dictate their approach to studying for the exam.

mitch895 · Answer

The sample was based on Volume of EOC questions.

I.e.,

Number of EOC questions (reading) divided by total EOC questions for Subject, multiplied by exam weighting for Subject.

For example, say there are 10 EOC questions for reading #100, and the reading was part of a subject that had 50 questions across multiple Readings (so, Reading #100 = approx. 10/50 = 20% of subject content).  Then let's say that the subject was worth 10% of the exam scores; in that case the (very, very rough guesstimate) would be that Reading #100 might be worth somewhere in the order of 2% of the total exam content (20% x 10%).

Re: the number of pages.... god damn it, you're right!  I was calculating the number of pages based on each subject starting at page 1, which meant that I've totally miscalculated the number of pages for any of the books that have "doubled up" on subjects! Phew!  That's a little less reading then!

Correcting the numbers it now looks like Economics is the most heavy-handed when it comes to reading, though the heaviest by far looks to be reading #41 Corporate Finance - The Corporate Governance of Listed Companies: A manual for investors.

As for the total number of pages, I tallied up the total pages in each book and I think it came to a bit over 3,500 pages.  This was before actually reading any of the content, so some of this is going to be glossaries and indexes.  I'll be happy for it to be 2,500 though!

Really this exercise was just one of those things I sort of ran through out of interest (and thanks to Excel only takes a few minutes to do).