Geography Reference
In-Depth Information
Multivariate statistics
3.3
h e focus so far has been on methods for exploring single variables. In many applica-
tions, there is a need to consider how two or more variables are related to one another.
For example, what is the relationship between altitude and snowfall? Does snowfall
tend to be greater at high altitudes? h is section makes use of the data given in Table
3.1. h e i rst stage of an analysis may involve plotting one set of values against the
other (this is called a scatter plot—an example is given in Figure 3.3) and such an
analysis could be expanded using regression analysis, as detailed in this section.
Figure 3.3 indicates that small values of
z
(in this example, snowfall) tend to corre-
spond to small values of
y
(elevation) and that large values of one tend to correspond
Table 3.1
Sample data set for illustrating regression.
Variable 1 (
y
) in m
Variable 2 (
z
) in cm
12
6
34
52
32
41
12
25
11
22
14
9
56
43
75
67
43
32
80
70
60
50
40
30
20
10
0
0
10
20
30
40
50
60
70
80
y
Figure 3.3
Scatter plot. Elevation (
y
) in metres (m) against snowfall (
z
) in centimetres (cm).
Search WWH ::
Custom Search