The workbook covers how to work out averages and range using the theme of hand size, and then goes into a real life example of
distorted average using the hand size of basketball player (video link included) and when each type of average should be used.
Not exact matches
Tim Perhaps, I should have added to my last comment that I accept that I am not in a position to put forward alternative figures (partly because I consider that there is a lack of accurate empirical data on a number of aspects that affect the budget) but in general, I consider that the
use of
average figures is not simply an over simplification of the system but more significantly it disguises and
distorts what is in practice going on thereby leading to a failure to properly understand how the system works as a whole.
The method I
used was a way to compare long term data sets of different record lengths, and how adding stations
distorts the
average.
I reworked the idea of trend analysis
using less
distorting filters and reproduced the split in two groups even
using shorter filters (whereas M&F sufficiently mangled the data that the separation was not found
using their 15y sliding
averages).
It is very easy to
distort what is happening when
using averages without error bars.