OK, you passed - but how good you really were?

S

shootingstar

New member
Joined
Jun 18, 2026
Messages
0
Reaction score
0
What was your 40/60/80 score in the exam?
I checked mine (75.1%) at 300hours.com - I hope more people try this so that we all get nice analysis later on.
 
Kind of lame since we know that in the PM the 60% number has to be 66.67% and the 80% number has to be either 83% or 100%. Lazy oversight IMO.
At least in the majority of the item sets (the ones worth 18 points).
 
Systematic wrote:
Kind of lame since we know that in the PM the 60% number has to be 66.67% and the 80% number has to be either 83% or 100%. Lazy oversight IMO.
At least in the majority of the item sets (the ones worth 18 points).
Click to expand...
I made this adjustment to my own calc as well.
 
Systematic wrote:
Kind of lame since we know that in the PM the 60% number has to be 66.67% and the 80% number has to be either 83% or 100%. Lazy oversight IMO.
At least in the majority of the item sets (the ones worth 18 points).
Click to expand...
Well, you could have come up with a better system and implemented it on your website.
Let’s give him the credit for putting effort on this.
 
Another new way for me to burn up time. Any way, 66.7% overall and a pass.
I’m fine with the 40/60/80 methodology. It has obvious flaws, but it’s a reasonable way to assess relative scores.
 
Systematic wrote:
Kind of lame since we know that in the PM the 60% number has to be 66.67% and the 80% number has to be either 83% or 100%. Lazy oversight IMO.
At least in the majority of the item sets (the ones worth 18 points).
Click to expand...
Appreciate the comments. You are, of course, correct that for the PM paper the questions are fairly discrete so the rough 40/60/80 can be improved (as well as the min/max).
However when designing the tool I found the calculation to be resource intensive, and in favour of performance reliability I decided to make the trade-off. Clearly not up to AF’s members’ standards!
I will have another look tomorrow and see if I can solve this issue - with time I should be able to come up with an elegant solution. I will keep you guys updated.
 
Quant was not my strongest area, but I don’t think that the chance of given result in an item set is same for all values.
For instance, if a candidate is making six pure random guesses, there is a stronger likelihood of scoring a 2/6 than a 0/6.
 
I think the concept is that if an item set has 6 questions, then for 51%-70% the only combination is 4/6, which is 66.67% (as 3/6 will be 50%, and 5/6 will be 83.33%). Hence for the PM paper, 51%-70% should be weighted as 66.67%, not 60%.
Likewise with min/max, the min for 71%-100% for PM will be 83.33%, not 71%.
These limits and assumptions will vary across topics (as they have different point weghtings).
 
shootingstar wrote:
Systematic wrote:
Kind of lame since we know that in the PM the 60% number has to be 66.67% and the 80% number has to be either 83% or 100%. Lazy oversight IMO.
At least in the majority of the item sets (the ones worth 18 points).
Click to expand...
Well, you could have come up with a better system and implemented it on your website.
Let’s give him the credit for putting effort on this.
Click to expand...
Yes, I could have, but I’m lazy and I chose not to. Any questions?
 
I just built my own excel spreadsheet. It’s super easy and all you need is 2 sets of % (AM and PM) and 1 formula.
=IF(D3=”*”,C3*$D$35,IF(E3=”*”,C3*$E$35,C3*$F$35))
AM i stuck with 40/60/80
PM i used 42/67/83
 
Of course 40/60/80 is off – possibly way off – at the individual item set level. But if the upper two bands are too conservative, the bottom band could be too generous. So, you fix the methodology and come up with different score. Is it really any more valid?
Remember, some questions in the morning counted for nearly 10% of the overal exam score. An error in those sections makes a pretty big diff.

usj2 wrote:
I think the concept is that if an item set has 6 questions, then for 51%-70% the only combination is 4/6, which is 66.67% (as 3/6 will be 50%, and 5/6 will be 83.33%). Hence for the PM paper, 51%-70% should be weighted as 66.67%, not 60%.
Likewise with min/max, the min for 71%-100% for PM will be 83.33%, not 71%.
These limits and assumptions will vary across topics (as they have different point weghtings).
Click to expand...
 
40/60/80 was a 60.4% according to their website. Passed, so I really don’t care, but that’s the calculated score.
 
From the website, I got this:

- Your 40/60/80 score is: 69.4%
- Your minimum score is: 56.8%
- Your maximum score is: 85.3%
That is, I could still fail with the same reported result.
 
IMO, for the Level III PM paper, 40/67.7/ 85 should be an improvement on 40/60/80 and 40/67/83.
Why 67.7? Hank Moody, I take your point, but correcting this band still takes a step in the right direction. e.g. if a candidate scores all in the 51%-70% band for the PM, we know that the score has to be 67.7 - there are no other possibilities.
Why 85? Since 83.3 is the lower limit and 100% the upper limit for the >70% category, the ‘average’ level should be something in between. I can’t prove that e.g. 85 is better than 84, but it should be something in between, so I think 85 should suffice, but 83 would be a ‘minimum’ score rather than an ‘average’ estimate.
Why 40? Since wild guesses should average 33.3%, something between 33.3% and 50% would do it for me for that category.
Changing the values for min and max does make it theoretically more valid, i.e. 0/66.7/83.3 for min, 50/66.7/100 for max.
Iteracom: I didn’t say it was difficult, I said it was resource intensive, more so for Levels I and II rather than III. For each topic that has a different points weighting these values would have to be assigned uniquely, and incorporating that into a reliable system that can support multiple simultaneous queries didn’t seem like something I was equipped to deal with at the time. Learnt lots from the experience though so I may give it another go.
Having said that, I did manage to work out a solution for Level III and have incorporated it now. Hopefully this new set of values doesn’t confuse the majority even further. But if it doesn’t I’d be happier to roll with this as I think it’s closer to the ‘true’ score.
 
At the risk of rambling - for questions such as Ethics this year, since there are 2 item sets, again the whole set of values are different, since it’s a base of 12 questions rather than 6 for that category. Minimum will now be 0/58.3/75, and max will be 50/67/100.
 
You must log in or register to reply here.
Back
Top