This perfectly illustrates why feedback questions must be treated with caution.
They are totally reliant on the memory of the students who sit the exams. It also shows how vulnerable we can all be to not reading the question properly, and in this respect I must offer my apologies to OB.
The answers to these feedback questions are also subject to how the person providing the answers reads the question, and in the original case reported by PP it would seem that two people interpreted this question differently in the two sets of feedback papers.
We are all in this together, and I wasn't trying to points score in my original contibution to this thread.
Basic DI drift calculations are quite easy if a set methodology is maintained, but so much can depend on how any particular question is read.
One last thing is the CAA responses to any challenges made. They will accept it when a question is deemed incorrect or unfair, but they are very careful to avoid giving any clue as to just how a question is phrased.
BTW: When 6000 question are compiled in the CQB they have stated that the CQB will be made public. What's the betting they stop adding questions at number 5,999!!!