That only goes part way. Even though you might be able to say a subject gets tested 50% or 80% of the time, you need to translate that into a rule that says how the different ratings are weighted to fill any 'spare' time slots.
Suppose, for example, you have six 'spare' time slots and six subjects, rated at 100%, 95%, 90%, 85%, 80% and 75%, respectively (let's not worry about subjects with equal ratings yet), do you put:
• each of these subjects into one of the time slots;
• the 100% subject in two slots and skip the 75% subject's re-test;
• the 100% subject in three slots and skip the 80% & 75% subjects' re-tests;
• the 100% & 90% subjects in two slots each and skip the 80% & 75% subjects' re-tests; or
• something else.
What is the rule?
Another possible complication is that different subjects might be worth different test scores. So a ferquently-occurring test might not be as valuable as some of the less-frequent ones. Is this a possibility and, if so, how would you handle that?
__________________
Cheers,
Paul Edstein
[Fmr MS MVP - Word]
|