Page 165 - BA2 Integrated Workbook - Student 2017
P. 165
Risk 1: summarising and analysing data
2.2 Grouped distributions
Where we have more outcomes in our data, if we try to tally using each possible
measurement taken, then we will still have too many values to be really useful.
A far more sensible approach is to tally the number of values in a certain range or
class. The choice of classes is somewhat arbitrary, but should be such that they are
neither too narrow, which could result in most of the frequencies being zero, nor too
wide, which could produce only a small number of classes and thereby tell us little.
As a rough guide, between four and twelve groups are often used.
If our data was:
2 7 21 11 14 10 8 12 13 18 32 21 15
The range here is between 2 and 32. In this case it would be easier to group the
outcomes:
0 – 5
6 – 10
11 – 15
16 – 20
21 – 25
26 and over
When grouping data, take care when defining the classes so that it is clear where
each observation will fit, e.g. 11 – 15, 16 – 20 etc. Be careful of using confusing
groupings such as 10 – 15, 15 – 20 etc.
Go through illustration 2
Try TYU 2
157
