CamScanner

CamScanner
CamScanner
CamScanner
CamScanner
CamScanner
CamScanner
CamScanner
CamScanner
CamScanner
Regression and Multiple
Correlation
11.2 COEFFICIENT OF MULTIPLE
CORRELATION

If information on two variables like height and weight, income and

expenditure, demand and supply, etc. are available and we want to study the
linear relationship between two variables, correlation coefficient serves our
purpose which provides the strength or degree of linear relationship with
direction whether it is positive or negative. But in biological, physical and
social sciences, often data are available on more than two variables and value
of one variable seems to be influenced by two or more variables. For
example, crimes in a city may be influenced by illiteracy, increased
population and unemployment in the city, etc. The production of a crop may
depend upon amount of rainfall, quality of seeds, quantity of fertilizers used
and method of irrigation, etc. Similarly, performance of students in
university exam may depend upon his/her IQ, mother’s qualification, father’s
qualification, parents income, number of hours of studies, etc. Whenever we
are interested in studying the joint effect of two or more variables on a single
variable, multiple correlation gives the solution of our problem.
In fact, multiple correlation is the study of combined influence of two or
more variables on a single variable.
Suppose, X1 , X2 and X3 are three variables having observations on N
individuals or units. Then multiple correlation coefficient of X1 on X2 and
X 3 is the simple correlation coefficient between X1 and the joint effect of
X2 and X 3 . It can also be defined as the correlation between X1 and its
estimate based on X2 and X3 .
Multiple correlation coefficient is the simple correlation coefficient between
a variable and its estimate.
Let us define a regression equation of X1 on X2 and X3 as

X1  a  b12.3X 2  b13.2 X3

Let us consider three variables x1 , x 2 and x 3 measured from their respective

means. The regression equation of x1 depends upon x 2 and x 3 is given by

x1  b12.3x2  b13.2 x3 … (1)

Where X1  X1  x1 , X 2  X2  x 2 and X3  X3  x3
 x 1   x 2   x 3  0
Right hand side of equation (1) can be considered as expected or estimated
value of x1 based on x 2 and x 3 which may be expressed as

x1.23  b12.3 x 2  b13.2 x3 … (2)

23
 r122  r132  2r12 r13 r23
R1.23 = … (3)
1  r232
which is required formula for multiple correlation coefficient.
where, r12 is the total correlation coefficient between variable
X1 and X2 , r23 is the total correlation coefficient between
variable X2 and X3
Now let us solve a problem on multiple correlation coefficients.
Example 1: From the following data, obtain R1.23 and R 2.13

X1 65 72 54 68 55 59 78 58 57 51
X2 56 58 48 61 50 51 55 48 52 42
X3 9 11 8 13 10 8 11 10 11 7

Solution: To obtain multiple correlation coefficients R1.23 and R 2.13 , we

use following formulae
r 2  r 2  2r r r
2
R 1.23  12 13 2 12 13 23 and
1  r23
r 2  r 2  2r r r
2
R  12 23 12 13 23
1  r132
2.13

We need r12 , r13 and r23 which are obtained from the following table:

S. No. X1 X2 X3 (X1)2 (X2)2 (X3)2 X1X2 X1X3 X2X3

1 65 56 9 4225 3136 81 3640 585 504

2 72 58 11 5184 3364 121 4176 792 638

3 54 48 8 2916 2304 64 2592 432 384

4 68 61 13 4624 3721 169 4148 884 793

5 55 50 10 3025 2500 100 2750 550 500

6 59 51 8 3481 2601 64 3009 472 408

7 78 55 11 6084 3025 121 4290 858 605

8 58 48 10 3364 2304 100 2784 580 480

9 57 52 11 3249 2704 121 2964 627 572

10 51 42 7 2601 1764 49 2142 357 294

Total 617 521 98 38753 27423 990 32495 6137 5178

Now we get the total correlation coefficient r12 , r13 and r23

N(X1X2 )  (X1 ) (  X 2 )
r12 
N( X )  ( X ) N( X )  (X
2
1 1
2 2
2 2
)2 
(10  32495)  (617)  (521)
r12 
(10 38753)  (617) (617)(10  27423)  (521)  (521)
41
Regression and Multiple 3493 3493
r12    0.80
Correlation
 4368.01

N(X1X3 )  (X1 )(X3 )

r13 
  
N(  X12 )  (  X1 ) 2 N(  X 32 )  (  X 3 ) 2 

(10  6137)  (617)  (98)
r13 
(10  38753)  (617  617)(10  990)  (98  98)
904 904
r13    0.64
 1423.00

and
N(X 2 X3 )  ( X 2 )( X 3 )
r23 
  
N(  X 22 )  ( X 2 ) 2 N(  X 32 )  (X 3 ) 2 

(10  5178)  (521)  (98)
r23 
(10  27423)  (521 521)(10  990)  (98 98)
722 722
r23    0.79
 908.59

Now, we calculate R1.23

We have, r12  0.80 , r13  0.64 and r23  0.79 , then

r 2  r 2  2r r r
2
R  12 13 12 13 23
1  r232
1.23

0.802  0.642  2  0.80  0.64  0.79


1  0.792
0.64  0.41 0.81

1  0.62
0.24
2
R 1.23   0.63
0.38
Then
R1.23  0.79 .
r 2  r 2  2r r r
2
R  12 23 12 13 23
1  r132
2.13

0.802  0.792  2  0.80  0.64  0.79


1  0.642
0.64  0.62  0.81

1  0.49
0.45
  0.88
42 0.51
 Multiple Correlation
Thus,
R 2.13 = 0.94
Example 2: From the following data, obtain R1.23 , R 2.13 and R3.12
X1 2 5 7 11
X2 3 6 10 12
X3 1 3 6 10

Solution: To obtain multiple correlation coefficients R1.23 R 2.13 and R3.12,

we use following formulae
r 2  r 2  2r r r
2
R 1.23  12 13 2 12 13 23 ,
1  r23
r 2  r 2  2r r r
2
R 2.13  12 23 12 13 23 and
1  r132
2 r2  r2  2r r r
R3.12  13 23 12 13 23

1 r122

We need r12 , r13 and r23 which are obtained from the following table:

S. No. X1 X2 X3 (X1)2 (X2)2 (X3)2 X1X2 X1X3 X2X3

1 2 3 1 4 9 1 6 2 3

2 5 6 3 25 36 9 30 15 18

3 7 10 6 49 100 36 70 42 60

4 11 12 10 121 144 100 132 110 120

Total 25 31 20 199 289 146 238 169 201

Now we get the total correlation coefficient r12 , r13 and r23
N(X1X2 )  (X1 )( X 2 )
r12 
N(X )  (X ) N(X )  (X
2 2 2
)2 

  1 1 2 2

  (4  238)  (25)  (31)

r12    (25)(4  289)  (31)  (31)
 (4 199)  (25)
 

r  177  177  0.0.97
 182.61


12

N(X1X3 )  (X1 )( X 3 )
r13 
 
N(  X12 )  (X 1 ) 2 N(  X 32 )  ( X 3 ) 2 

43
Regression and Multiple (4 169)  (25)  (20)
r13 
Correlation
(4 199)  (25 25)(4 146)  (20 20)
176 176
r13    0.99
 177.38

and
N( X 2 X3 )  ( X 2 )(X3 )
r23 
 
N( X 22 )  ( X 2 ) 2 N( X 32 )   X ) 
3
2


(4  201)  (31)  (20)
r23 
4  289)  (31 31)(4 146)  (20  20)
184 184
r23    0.97
 189.42

Now, we calculate R1.23

We have, r12  0.97 , r13  0.99 and r23  0.97 , then

r 2  r 2  2r r r
2
R 1.23  12 13 12 13 23
1  r232
0.972  0.992  2  0.97  0.99 0.97

1 0.972

0.058
  0.98
0.059
Then
R1.23  0.99 .
r 2  r 2  2r r r
2
R 2.13  12 23 12 13 23
1  r132
0.972  0.972  2  0.97  0.99 0.97

1  0.992

0.19
  0.95
0.20
Thus,
R 2.13 = 0.97
2 r2  r2  2r r r
R3.12  13 23 12 13 23
1 r122
0.992  0.972  2  0.97  0.99 0.97

1 0.972
44
 Multiple Correlation
0.58

0.591
 0.981
Thus,
R3.12 = 0.99

Example 3: The following data is given:

X1 60 68 50 66 60 55 72 60 62 51

X2 42 56 45 64 50 55 57 48 56 42

X3 74 71 78 80 72 62 70 70 76 65

Obtain R1.23 , R 2.13 and R3.12

Solution: To obtain multiple correlation coefficients R1.23 , R 2.13 and R3.12

we use following formulae:
r 2  r 2  2r r r
2
R 1.23  12 13 2 12 13 23 ,
1  r23
r 2  r 2  2r r r
2
R 2.13  12 23 12 13 23
1  r132 and
2 r  r  2r r r
2 2

R3.12  13 23 12 13 23

1 r122

We need r12 , r13 and r23 which are obtained from the following table:
S. X1 X2 X3 d1= d2 = d3= (d1)2 (d2)2 (d3)2 d1d2 d1d3 d2d3
No. X1 -60 X2 -50 X3 -70
1 60 42 74 0 -8 4 0 64 16 0 0 -32
2 68 56 71 8 6 1 64 36 1 48 8 6
3 50 45 78 -10 -5 8 100 25 64 50 -80 -40
4 66 64 80 6 14 10 36 196 100 84 60 140
5 60 50 72 0 0 2 0 0 4 0 0 0
6 55 55 62 -5 5 -8 25 25 64 -25 40 -40
7 72 57 70 12 7 0 144 49 0 84 0 0
8 60 48 70 0 -2 0 0 4 0 0 0 0
9 62 56 76 2 6 6 4 36 36 12 12 36
10 51 42 65 -9 -8 -5 81 64 25 72 45 40
Total 4 15 18 454 499 310 325 85 110

45
Regression and Multiple Here, we can also use shortcut method to calculate r12, r13 & r23,
Correlation
Let d1 = X1− 60
d2 = X2− 50
d3 = X1− 70
Now we get the total correlation coefficient r12 , r13 and r23

N(d1d2 )  (d1 )( d 2 )
r12  N(d )  ( d ) N(d )  (d ) 
2
1 1
2 2
2 2
2

r12  (10 325)  (4) (15)

(10  454)  (4) (4)(10  499)  (15)  (15)

r  3190  3190  0.69
12
 4642.94
N(d1d3 )  (d1 )(d3 )
r13 
  
N(  d12 )  (  d1 ) 2 N(  d 32 )  (  d 3 ) 2 

(10  85)  (4)  (18)
r13 
(10  454)  (4  4)(10  310)  (18  18)
778 778
r13    0.22
 3543.81

and
N( d 2 d 3 )  (  d 2 )(d3 )
r23 
  
N( d 22 )  ( d 2 ) 2 N( d 32 )  ( d 3 ) 2 

(10 110)  (15)  (18)
r23 
(10  499)  (1515)(10  310)  (1818)
830 830
r23    0.23
 3636.98

Now, we calculate R1.23

We have, r12  0.69 , r13  0.22 and r23  0.23 , then

r 2  r 2  2r r r
2
R  12 13 12 13 23
1  r232
1.23

0.692  0.222  2  0.69  0.22  0.23


1  0.232
0.4547
  0.4801
0.9471
Then
46 R 1.23  0.69
 r 2  r 2  2r r r Multiple Correlation
2
R  12 23 12 13 23
1  r132
2.13

0.692  0.232  2  0.69  0.22  0.23


1  0.222
0.4592
  0.4825
0.9516
Thus,
R 2.13 = 0.69
2 r2  r2  2r r r
R3.12  13 23 12 13 23
2
1 r12
0.22  0.232  2  0.69  0.22  0.23
2


1  0.692
0.0315
  0.0601
0.5239
Thus,
R3.12 = 0.25
Now let us solve some exercises.
E1) In bivariate distribution, r12  0.6, r23  r31  0.54, then calculate
R1.23 .
E2) If r12  0.70, r13  0.74 and r23  0.54 , calculate multiple correlation
coefficient R 2.13 .
E3) Calculate multiple correlation coefficients R1.23 and R 2.13 from the
following information: r12  0.82, r23   0.57 and r13   0.42 .
E4) From the following data,

X1 22 15 27 28 30 42 40
X2 12 15 17 15 42 15 28
X3 13 16 12 18 22 20 12

Obtain R1.23 , R 2.13 and R3.12

E5) The following data is given:

X1 50 54 50 56 50 55 52 50 52 51
X2 42 46 45 44 40 45 43 42 41 42
X3 72 71 73 70 72 72 70 71 75 71

By using the short-cut method obtain R1.23 , R 2.13 and R3.12

47