Chronology | Current Month | Current Thread | Current Date |
[Year List] [Month List (current year)] | [Date Index] [Thread Index] | [Thread Prev] [Thread Next] | [Date Prev] [Date Next] |
Within any specific knowledge of what Blackboard is doing, I might
venture
a guess that Blackboard is reporting what is known as the standard
error,
which is (st_dev)/(n^1/2).
This seems to basically agree with your numbers: 13 / 24^(1/2) = 2.6
Blackboard reports a standard deviation of around 2.
My statistics calculator computes a value near 13.
The standard deviation is a measure of how similar the students within
a
section are. The standard error is an estimate of how similar other
sections should be (assuming student of the same general ability).
I.e., if
you gave the same exam to a different section (with the same general
ability), then you would expect the means of the sections to fall
within
about 2.6 of each other. As the difference gets larger, then you can
be
more confident that the two sections are different (e.g. better
students,
better teaching, they got the answers from previous sections, etc).
Personally, I'd rather have them report the standard deviation -
especially
if that is what they say they are reporting!
(This is a hunch. This is only a hunch. If this had been actual
knowledge,
you would have been informed where to tune for further information ;-)