>> The problem with distributing grades along a bell curve, like that, is
>> that in years where the class is full of genies, someone who has an
>> otherwise good grasp of the theory ends up with a C, and years when the
>> class is full of morons, you give an A to someone who doesn't merit one.
>
> Agreed if only a small number of people are taking the exam, a handful
> of geniuses could well upset the balance. However for national exams
> like the GCSEs in the UK there are 700k people taking them, having a
> fixed distribution is surely much fairer than trying to ensure the
> standard of questions is the same each year (whilst trying to not make
> them too similar to previous questions).

The 700k population of students perhaps doesn't fluctuate much, but the 
difficulty of the questions surely does. But how to control for that?

I still believe that a qualification - ANY qualification - should have a 
fixed set of requirements, and anybody who meets those gets the 
qualifications, regardless of what anybody else did in the same year.

(Not that I expect my opinion to influence anything...)

Post a reply to this message