Page 71 - MATLAB Recipes for Earth Sciences

4.2 Pearson·s Correlation Coeffi cient 63

Bivariate Scatter Bivariate Scatter
120 20
r = 0.96 r = -0.97
100 0
−20
80
−40
y 60 y
−60
40
−80
20 −100

0 −120
0 5 10 15 20 0 5 10 15 20
x x
a b
Bivariate Scatter Bivariate Scatter
20
r = 0.36 r = 0.95
15 15
Outlier
y 10 y 10

5 5
Random bivariate
data cluster
0 0
0 5 10 15 20 0 5 10 15 20
x x
c d
Bivariate Scatter Bivariate Scatter
2500 600
r = 0.96 r = 0.38
2000 500
1500 400

y 1000 y 300
500 200

0 100
−500 0
0 5 10 15 20 −10 −5 0 5 10
x x
e f

Fig. 4.2 Pearson·s correlation coeffi cent r for various sample data. a-b Positive and negative
linear correlation, c random scatter without a linear correlation, d an outlier causing a
misleading value of r, e curvilinear relationship causing a high r since the curve is close to a
straight line, f curvilinear relationship clearly not described by r.