How to find out the representive mark from a list?
In what way do you consider that AVERAGE and MODE are "not a very
representative" in this case?
What do you want to do to make it more "representative"? If you don't want
to minimise the total difference, do you want a least squares fit, or do you
want to minimise
sigma[(x i-x rep)^n],
where you define your favourite value of n?
Another option is MEDIAN.
--
David Biddulph
"Eric" wrote in message
...
My class has 30 students, their marks are shown below
100,97,96,93,88,85,84,77,76,71,
66,62,62,55,53,49,48,46,46,43,
43,43,43,43,35,33,31,31,31,13
I would like to determine the most representative mark for this class for
comparing with another class.
I think the functions - average, mode are not a very representative in
this
case.
Does anyone have any suggestions on how to define the most representative
mark for this class?
Thank you in advance
Eric
|