Many of you have asked what was the sample size. While not quite in the same league as the CFA Institute candidate survey (around 30,000 participants), it’s still pretty robust. The conclusions drawn from the analyses are pretty solid and any limitations are highlighted (as you will see in the Level II MPS estimate). You’ll get a better idea of the sample size in a future post.
In this post, we dive deep into Level II, generally agreed to be the toughest level among the three. As with the Level I post, we cover the MPS and band differences here.
Again, before we explore this section, let me repeat an important disclaimer and caution: The information in this section really has no practical application in your CFA preparation. No matter what information is disclosed on the MPS, with the exam format as it stands, you should always aim for 70% and above, and for Level II you need to step it up a notch (or several notches) compared to Level I. No amount of speculating on the MPS is going to change that.
This analysis is just for fun – do NOT hinge your prep for Level II on this. You have been warned.
The MPS for Level II was a bit more difficult to analyze – an exact percentage wasn’t possible from the data sample. This was surprising since the sample was larger than the Level I sample. However, the possible range for the MPS that the analysis yielded was 58% to 63%. This range is lower than the estimated MPS for Level I.
Those familiar with the Angoff method used by CFA Institute to establish a basis for the MPS would not be surprised by a lower MPS. But it’s funny – Level II generally is agreed to be the toughest level around (and further confirmed by a Weekly Poll I ran some time ago), but it has a higher pass rate and lower MPS than Level I.
Band difference is about 3-5 questions per band.
Although the data for MPS was not ideally adequate, the data for band differences was pretty good.
The percentage differences between failing bands are much smaller compared to Level I. For Level I, the difference between bands was about 8-11 questions. For Level II, the difference between bands is only about 3-5 questions. This means that assuming you were around the mid-point of your band, you were just one or two questions away from reaching the next band.
Why is this?
Less questions, more points. There are less questions in Level II compared to Level I (half actually), so even if the percentage differences remain the same, the number of questions between bands would be halved.
Competition is fiercer. The variation of quality between Level II candidates is also much lower compared to Level I. Remember, all Level II candidates would have already passed Level I, so the more casual candidates would have been weeded out already. Since the bands are defined in relative terms (e.g. Band 10 meant you are in the top 10% of failing candidates and so on) this would mean the differences in bands would be much smaller.
Below is the graph showing the band difference estimation for Level II. For more detail on the methodology, see the Level I analysis.
Future posts (perhaps not immediately after this) will deal with Level III, which is a whole other interesting beast. Check back in the coming weeks to find out more.
Meanwhile, it would be great to know what you think. Shout out in the comments box below!
Like this post? Get our analysis report for all our analytics for all levels, or view our 10 Commandments for every level!