CORRELATION

Correlation is defined as the relationship between two or more variables. Two variables are said to be
correlated if the change in one variable results in a corresponding change in the other variable.

For example: when price of a commodity rises supply for that commodity also rises.

Different kinds of correlation

1. Positive and negative correlation

Positive correlation:

Two variables are moves in the same direction then the correlation is said to be positive correlation.
That is an increase in the value of one variable causes an increase in the value of other variable or a
decrease in the value of one variable causes a decrease in the value of the other variable.

for example: height, weight

Negative correlation:

Two variables moves in the opposite direction then the correlation is said to be negative correlation.
That is an increase in the value of one variable causes a decrease in the value of the other variable
or a decrease in the value of one variable causes an increase in the value of other variable.

For example: price, demand

2. Linear and non-linear correlation

Linear correlation:

When the amount of change in one variable leads to a constant ratio to the change in the other
variable, then the correlation is said to be linear

x 5 10 15 20
y 10 20 30 40

Ratio 1:2

Graph of linear correlation will be a straight line.

Non-linear correlation (curve linear correlation)_

 When the amount of change in one variable does not leads to a constant ratio to the change in the
other variable, then the correlation is said to be non- linear.

x 5 10 15 20
y 12 23 8 25

Graph of non-linear correlation will be a curve.

3. Simple, partial and multiple correlation

Simple correlation

In the study of relationship between variables , if there are only two variables, the correlation is
said to be simple.

Example: price, demand

Partial correlation

In partial correlation we study the relationship between any two variables , and the third variable
remains constant, if there are three variables.

Example: pressure, volume and temperature

Multiple correlation

In multiple correlation we study the relationship between one variable on one side and the
remaining variables on the other side.

Example: yield, rainfall and temperature

METHODS OF STUDYING CORRELATION

1. Scatter diagram

This is a graphical method of studying correlation between two variables. One of the variables is
shown on the X – axis and the other on the Y- axis. Each pair of value is plotted on the graph by
means of a dot mark. After all the items are plotted we get as many dots on the graph paper as the
number of points. If this points show some upward trend ( left bottom to right top) , then the
correlation is said to be positive. If this points show some downward trend (left top to right bottom)
then the correlation is said to be negative. If the plotted point do not show any trend then there is
no correlation.
 Positive correlation

No correlation

Negative correlation

2. Coefficient of correlation

Correlation coefficient is denoted by ‘r’. the value of correlation coefficient lies between -1 and+1.

When r= +1 , There is perfect positive correlation.

When r =-1, T here is perfect negative correlation.

When r=0, There is No correlation.

 Methods of measuring correlation coefficient

1. Karl pearson’s coefficient of correlation

Correlation coefficient=r=

n xy  ( x  y )
n x 2    x  n y 2   y 
2 2

n  dxdy  ( dx  dy )
r
n  dx 2    dx  n  dy 2    y 
2 2

cov( x, y)
Karl Pearson’s correlation co.efficient=
 x  y
,

Product moment correlation coefficient =

 ( x  x )( y  y )
n x   y

2. Spearman’s Rank correlation coefficient

6 D 2
Spearman’s rank correlation coefficient = 1 
n(n2  1)

m3  m
6[ D  (
2
)]
In case of tied ranks, Spearman’s rank correlation coefficient = 1  12
n(n 2  1)

2. Concurrent deviation method

2C  N
Concurrent deviation is  ( )
N
1. Karl person’s co. efficient of correlation

Ex. Calculate co. efficient of correlation

X 2 3 4 5 6 7 8

Y 4 5 6 12 9 5 4

Solution:

X Y XY X2 Y2

2 4 8 4 16

3 5 15 9 25

4 6 24 16 36

5 12 60 25 144

6 9 54 36 81

7 5 35 49 25

8 4 32 64 16

35 45 228 203 343

n XY    X . Y 
r=
n  X 2    X  n Y 2   Y 
2 2

7  228  (35  45)

=
7  203  35  7  343  45 
2 2

1596  1575
=
1421  1225 2401  2025
 21
=
196 376
21
=
196 376

Correlation co. efficient ‘r’=0.077

2).Spearman’s rank correlation co. efficient

6 D 2
Spearman’s rank correlation coefficient = 1 
n(n2  1)

Ex.1. The ranking of 10 individuals at the start and at the finish of a course of training are as follows.

Individuals : A B C D E F G H I J

Rank before : 1 6 3 9 5 2 7 10 8 4

Rank after : 6 8 3 2 7 10 5 9 4 1

Rank Before Rank after Rank Difference (D) D2

1 6 5 25

6 8 2 4

3 3 0 0

9 2 7 49

5 7 2 4

2 10 8 64

7 5 2 4

10 9 1 1

8 4 4 16

4 1 3 9

Sum=176
 6 D 2
Rank correlation coefficient = 1 
n  n 2  1

6  176
=1 
10 100  1
=1 - 1.07

Rank correlation co efficient= -0.07

Ex2. Ten competitors in a beauty contest are ranked by three judges in the following order.

First judge :1 6 5 10 3 2 4 9 7 8

Second judge :3 5 8 4 7 10 2 1 6 9

Third judge :6 4 9 8 1 2 3 10 5 7

First judge Second judge Rank difference (D) D2

1 3 2 4

6 5 1 1

5 8 3 9

10 4 6 36

3 7 4 16

2 10 8 64

4 2 2 4

9 1 8 64

7 6 1 1

8 9 1 1

TOTAL 200
 6 D 2
Rank correlation coefficient = 1 
n  n 2  1

6  200
=1 
10(100 1)

=1-1.21

Rank correlation coefficient , r = -0.21

First judge Third judge Rank difference (D) D2

1 6 5 25

6 4 2 4

5 9 4 16

10 8 2 4

3 1 2 4

2 2 0 0

4 3 1 1

9 10 1 1

7 5 2 4

8 7 1 1

TOTAL 60

6 D 2
Rank correlation coefficient = 1 
n  n 2  1

6  60
=1 
10(100 1)

=1-0.364
Rank correlation co.efficient , r = 0.636

Second judge Third judge Rank difference (D) D2

3 6 3 9

5 4 1 1

8 9 1 1

4 8 4 16

7 1 6 36

10 2 8 64

2 3 1 1

1 10 9 81

6 5 1 1

9 7 2 4

TOTAL 240

6 D 2
Rank correlation coefficient = 1 
n  n 2  1

6  214
=1 
10(100 1)

= 1-1.30

Rank correlation co.efficient , r = -0.30

The rank correlation coefficient in the case of first and third judges is greater than other two pairs.
There for first and third judges have highest similarity of thought and have the nearest approach in
common taste in beauty.

REPEATED RANK

m3  m
6[ D 2  ( )]
In case of tied ranks, Spearman’s rank correlation coefficient = 1  12
n(n 2  1)
Obtain the rank correlation coefficient for the following data

X 68 64 75 50 64 80 75 40 55 64

Y 62 58 68 45 81 60 68 48 50 70

Solution:

X Rank of X Y Rank of Y D2

68 4 62 5 1

64 6 58 7 1

75 2.5 68 3.5 1

50 9 45 10 1

64 6 81 1 25

80 1 60 6 25

75 2.5 68 3.5 1

40 10 48 9 1

55 8 50 8 0

64 6 70 2 16

TOTAL 72

75 Occurs 2 times , m = 2 m3  m = 6

64 occurs 3 times , m= 3 m 3  m = 24

68 Occurs two times, m = 2 m3  m = 6

m3  m
6[ D 2  ( )]
Spearman’s rank correlation coefficient = 1  12
n(n 2  1)
 36 
6   72  
 12 
= 1
10(100 1)
  6  75 
=1   
 990 

Rank correlation coefficient , r = 0.54

2. Concurrent deviation method

2C  N
Concurrent deviation is r=  ( )
N

Ex. Calculate the co.efficient of concurrent deviation from the following data.

X: 20 25 30 15 28 32 35 17 29

Y: 30 18 25 10 30 25 15 30 27

X Y Direction of Direction of Dx DY
change (Dx) change of Y(Dy)
20 30 ................ ……… ………
25 18 + - -
30 25 + + +
15 10 - - +
28 30 + + +
32 25 + - -
35 15 + - -
17 30 - + -
29 27 + - -

C= no. of positive signs in Dx.DY column=3

N = No .of pairs compared= 8

2C  N
Concurrent deviation is r=  ( )
N

=-.5

Note:

When 2C>N, r is +ve

When 2C<N , r is –ve

 Properties of correlation coefficient.

1).correlation coefficient has a well defined formula

2).It lies between -1 and +1

3).it is a pure number and is independent of the units of measurement.

4).correlation co .efficient does not change with reference to change of origin or change of scale.

5).coefficient of correlation between X and Y is same as that of Y and X.

Probable error

Probable error is used to measure the reliability and dependability of the value of correlation
coefficient . If probable error is added or subtracted from the value of correlation coefficient , we get 2
limits within which the value of correlation coefficient may expected to lie.

0.6745(1  r 2 )
P.E= Probable error = 
n

(1  r 2 )
S.E =
n

Coefficient of determination = explained variance/ total variance