Allah is kind

Syl:- 3. Correlation Analysis: Meaning and concept,

Significance of the study of Correlation and
Causation; Types of Correlation; Scatter Diagram
Method; Karl Pearson’s Co-efficient Correlation;
Rank Correlation Co-efficient.
Formula:

(i) Coefficient of correlation:

( d x )(  d y )
 dxd y 
r= N
( d x ) 2
((  d y ) 2
d 
2
x d  2
y
N N

 x y
 xy  N
ii) r =
( x ) ( y )
2 2

{ x  2
}{  y  2
}
N N

Mizan sir 01911-971203, 01627-360073 1

 Correlation
Correlation : According to W.I. King.” Correlation means that between two series or
group of data there exists some causal connection.” Or
The statistical tool with the help of which these relationships between two or more than
two variables is studied is called correlation.
Types of correlation:
Correlation can be classified under the following heads according to nature and degree of
correlation:-
i) Perfect positive correlation
ii) Partial positive correlation
iii) Perfect negative correlation
iv) Partial negative correlation
v) Zero correlation
(i)Perfect positive correlation: If the relationship between two variables is such that
with an increase in the value of one, value of other increase in a fixed proportion,
correlation between them is said to be perfect positive correlation. In this case r=±1.

r=±1 0<r<1
Fig. Perfect positive correlation Fig. Partial positive correlation
ii)Partial positive correlation: If the relationship between two variables is such that with
an increase in the value of one, value of other increase not in a fixed proportion,
correlation between them is said to be partial positive correlation. In this case 0<r<1.
(iii) Perfect negative correlation: If the relationship between two variables is such that
with a decrease in the value of one, value of other decrease in a fixed proportion,
correlation between them is said to be perfect negative correlation. In this case r = -1.

Mizan sir 01911-971203, 01627-360073 2

 r= -1 -1<r<0
Fig. Perfect negative correlation Fig. Partial negative correlation
iv) Partial negative correlation: If the relationship between two variables is such that with
a decrease in the value of one, value of other decrease not in a fixed proportion,
correlation between them is said to be partial negative correlation. In this case -1<r<0.
v)Zero correlation: If the relationship between two variables is such that with a change
in the value of one, value of other not changed, correlation between them is said to be
zero correlation. In this case r=0.

r =0
Fig: Zero correlation
Properties of correlation coefficient:-
The following are the important properties of the coefficient of correlation:
i) The coefficient of correlation is independent of changes of scale and origin of the
variables x and y.
ii) The value of coefficient of correlation lies between -1 and +1.
iii) Coefficient of correlation is the geometric mean of two regression
coefficients.
Limitations of correlation analysis:-
(i) To determine the coefficient of correlation (r) we have to assume that there is a
linear relationship and not non-linear relationship.
(ii) It is valid when we have a random sample from a bivariate normal distribution.
(iii) If the sample size is small then it does not give us a better result to
determine the relation.
Q:-(2010). Discuss the degree of correlation.

Mizan sir 01911-971203, 01627-360073 3

=) The coefficient of correlation measures the degree of relationship between two
variables. Coefficient of correlation is denoted by ‘r’. The value of ‘r’ lies between -1 to
+1. r=+1 means perfect positive correlation exists the variables and r= -1 means the
perfect negative correlation. r=0 means no correlation between the variables. If the value
of r lies between 0 to ±1 then the relation express the correlation exist either partial
positive or partial negative.
Q(2009):- Briefly state the significance of correlation.
=) The study of correlation is of immense use in practical life because of the following
reasons-
(i) Most of the variables show some kind of relationship between price and supply,
income and expenditure, etc. with the help of correlation analysis we can
measure in one figure the degree of relationship existing between the variables.
(ii) The theory of correlation is very important for business and economic point of
view.
(iii) It is study is helpful for the purposes of prediction and forecasting.
(iv) It helps us in verifying the reliability and accuracy of the data.
* Q (2008,2012):-What is scatter diagram? How does it help in studying the
correlation between two variables.
=) Scatter diagram is the simplest method of finding out whether there is any relationship
present between two variables by plotting the values on a chart, known as scatter
diagram.
In this method, the given data are plotted on graph paper in the form of dots. X variables
are plotted on the horizontal axis and Y variables on the vertical axis. Thus we have the
dots and we can know the scatter or concentration of the various points. This will show
the type of correlation.

Perfect positive correlation Perfect negative correlation

r=+1 r=-1
Diagram-1 Diagram-2

Mizan sir 01911-971203, 01627-360073 4

If the plotted points from a straight line running from the lower left-hand corner to the
upper right-hand corner, then there is a perfect positive correlation(i.e., r=+1, Diagram-
1). On the other hand, if the points are in a straight line, having a falling trend from the
upper left-hand corner to the lower right- hand corner, it reveals that there is a perfect
negative or inverse correlation(i.e., r=-1, Diagram-2).
If the plotted points fall in narrow band, and the points are rising from lower left-hand
corner to the upper right-hand corner, there will be a high degree or positive correlation
between the variables(Diagram-3). If the points fall in a narrow band from the upper left-
hand corner to the lower right- hand corner, there will be a high degree or negative
correlation (Diagram-4). If the plotted points lie scatter all over the diagram, there is no
correlation between the two variables(Diagram-5).

0<r<1 -1<r<0 r=0

Partial positive correlation Partial negative correlation Zero correlation
Diagram-3 Diagram-4 Diagram-5
Over-all scatter diagram method help finding out nature of correlation between
two variables.
2006: Interpret: r = -1; r=0, r= 0.785; r= -.05.
Solution:
Interpretation: r = -1 a perfectly negative correlation between the variables.
Interpretation: r = 0 shows no correlation between the variables.
Interpretation: r = 0.785 shows a high degree of positive correlation between the
variables.
Interpretation: r = -0.05 shows a low degree of negative correlation between the
variables.
2015: (b) Interpret the notions: r=+1, r=-0.95, r=0. 3
Interpretation: r = 1 a perfectly positive correlation between the variables.
Interpretation: r = -0.95 shows a higher degree of negative correlation between the
variables
Interpretation: r = 0 shows no correlation between the variables.
Mizan sir 01911-971203, 01627-360073 5
1. Math:- Ten competitors in a beauty contest are ranked by three judges in the
following order:
1st Judge: 1 6 5 10 3 2 4 9 7 8
2nd Judge: 3 5 8 4 7 10 2 1 6 9
3rd Judge: 6 4 9 8 1 2 3 10 5 7
Use the rank correlation coefficient to determine which pair of judge has the nearest
approach to common tastes in beauty.
Solution: In order to find out which pair of judges has the nearest approach to common
tastes in beauty, we compare rank correlation between the judgment of
(i) 1st judge and 2nd judge
(ii) 2 nd judge and 3rd judge
(iii) 1st judge and 3rd judge.
Rank by 1st Rank by 2nd Rank by 3rd (R1-R2)2 (R2-R3)2 (R1-R3)2
judge (R1) judge (R2) judge (R3)
D2 D2 D2
1 3 6 4 9 25
6 5 4 1 1 4
5 8 9 9 1 16
10 4 8 36 16 4
3 7 1 16 36 4
2 10 2 64 64 0
4 2 3 4 1 1
9 1 10 64 81 1
7 6 5 1 1 4
8 9 7 1 4 1
N=10 N=10 N=10  D2=200  D2=214  D2=60

Mizan sir 01911-971203, 01627-360073 6

 6  D2 6  200 1200
R (i&ii)= 1- 3 = 1- 3 = 1- = 1- 1.212 = -0.212
N N 10  10 990

6  D2 6  214 1284
R (ii&iii)= 1- = 1- 3 = 1- = 1- 1.297 = -0.297
N N
3
10  10 990

6  D2 6  60 360
R (i&iii)= 1- = 1- 3 = 1- = 1- 0.364 = 0.636
N N
3
10  10 990

Since coefficient of correlation is maximum in the judgment of the first and third judges,
we conclude that they have the nearest approach to common tastes in beauty.
2. Math:-(2011). The following data relate to the prices and supplies of a commodity
during a period of eight years:-
Price (Tk./kg) 10 12 18 16 15 19 18 17
Supply (100kg) 30 35 45 44 42 48 47 46
Calculate the co-efficient of correlation between the two series.
Solution: Let, price be denoted by x and supply by y.
Calculation table of coefficient of correlation
Price(x) Supply (y) dx=x-A dy=y-A dx2 dy2 dxdy
10 30 -5 -12 25 144 60
12 35 -3 -7 9 49 21
18 45 3 3 9 9 9
16 44 1 2 1 4 2
15=A 42=A 0 0 0 0 0
19 48 4 6 16 36 24
18 47 3 5 9 25 15
17 46 2 4 4 16 8
 dx=5  dy=1  dx2=73  dy2=283 
dxdy=139

Mizan sir 01911-971203, 01627-360073 7

 ( d x )(  d y ) 5 1
 dxd y  139 
We know, r = N = 8
( d x ) 2
((  d y ) 2 (5) 2 (1) 2
 d x2   d y2  73  283 
N N 8 8

139  0.625 138.38 138.38

= = = =0.98 (ans).
(73  3.125)(283  0.125) 69.88  282.88 140.60

3. Math:-(2010). The co-efficient of rank correlation between debenture prices and

share prices is found 0.143 and sum of squares of the difference is 48. Find out the
value of N( Number of pairs).
Solution: Given, Rank of correlation coefficient (R)= 0.143;  D2= 48

6  D2
We know, R= 1-
N3  N

6  48 288
=) 0.143= 1- =) 0.857 = 3
N N
3
N N

288
=) N3 – N = =) N3 – N =336
0.857

=) N3 – N - 336 = 0
=) N3 – 7N2 +7N2 - 49N +48N - 336 = 0
=) N2(N – 7) + 7N(N – 7) +48 (N- 7)= 0
=) (N-7)( N2+ 7N +48) = 0
=) (N-7) = 0 or ( N2+ 7N +48) = 0 is imposible.
 N = 7 (Ans)

4. 2011(H): Given the following pairs of values of the variables X and Y:

X 10 20 30 40 50 60
Y 25 50 75 100 125 150
Draw a scatter diagram. Is there any correlation between the variables X and Y?
Solution: Scatter Diagram:

Mizan sir 01911-971203, 01627-360073 8

Fig: Scatter Diagram:
x Y x2 y2 xy
10 25 100 625 250
20 50 400 2500 1000
30 75 900 5625 2250
40 100 1600 10000 4000
50 125 2500 15625 6250
60 150 3600 22500 9000

 x=210  y=525  x2=9100  y2=56875  xy=22750

210  525
22750 
r= 6
(210) 2 (525) 2
{ {9100  56875  }
6 6

22750  18375 4375 4375

=  = =1
(9100  7350)(56875  45937.5) 1750 10937.5 4375

Comment: r =+1 shows perfectly positive correlation between X and Y.

5. Math:-(2010). Calculate co-efficient of correlation by using following data and
give the opinion of the relation of the variables:-
X 127 135 150 175 194 228
Y 750 800 870 820 780 735

Solution: Calculation table of coefficient of correlation

Mizan sir 01911-971203, 01627-360073 9
x (x-A)=dx dx2 y (y-A)=dy dy2 dx dy
127 -48 2304 750 -50 2500 2400
135 -40 1600 800=A 0 0 0
150 -25 625 870 70 4900 -1750
175=A 0 0 820 20 400 0
194 19 361 780 -20 400 -380
228 53 2809 735 -65 4225 -3445

 dx=-41  dx2=7699  dy=-  dy2=12425  dx dy=-3175

45
( d x )(  d y )  41  45
 dxd y   3175 
We know, r = N = 6
( d x ) 2 ((  d y ) 2
(41) 2
(45) 2
 d x2   d y2  7699  12425 
N N 6 6

 3175  307.5  3482.5  3482.5

= = = = -0.37 (ans).
(7699  280.17)(12425  337.5) 7418.83  12087.5 9469.69

Comment: There is a lower degree of negative correlation between of the two variables.
6. Math (2009):- Find co-efficient of correlation between price and sales from the
following data:-
Price (in taka) 103 98 85 92 90 84 88 90 93 95
Sales (units) 500 610 700 630 670 800 800 750 700 680

Solution: Let, price be denoted by x and sales by y.

Calculation table of coefficient of correlation
Price(x) Sales (y) dx=x-A dy=y-A dx2 dy2 dxdy
103 500 13 -200 169 40000 -2600
98 610 8 -90 64 8100 -720

Mizan sir 01911-971203, 01627-360073 10

85 700= A -5 0 25 0 0
92 630 2 -70 4 4900 -140
90= A 670 0 -30 0 900 0
84 800 -6 100 36 10000 -600
88 800 -2 100 4 10000 -200
90 750 0 50 0 2500 0
93 700 3 0 9 0 0
95 680 5 -20 25 400 -100
 dx=18  dy=160  dx2   dxdy=
=336 dy2=76800 -4360
( d x )(  d y ) 18  160
 dxd y   4360 
We know, r = N = 10
( d x ) 2
((  d y ) 2 (18) 2 (160) 2
 d x2  d  2
y
336  76800 
N N 10 10

 4360  288  4648  4648

= = = = -0.98
(336  32.4)(76800  2560) 303.6  74240 4747.55

7. Math (2008):- Phychologist wanted to compare two methods A and B of teaching.

He selected a random sample of 22 students. He grouped them into 11 pairs so that
the students in a pair one student was taught by method A and the other by method
B and examined after the course. The marks obtained by them are tabulated below:-
Pair 1 2 3 4 5 6 7 8 9 10 11
A 24 29 19 14 30 19 27 30 20 28 11
B 37 35 16 26 23 27 19 20 16 11 21

i)Find the correlation coefficient between the two sets of scores.

(ii)Find the rank correlation coefficient between coefficient between the two sets of
scores.

Mizan sir 01911-971203, 01627-360073 11

Solution : (i) Calculation of coefficient of correlation
Test No. Method dx=x-A d x2 Method dy=y-A dy2 dxdy
A (x) B (y)
1 24 5 25 37 17 289 85
2 29 10 100 35 15 225 150
3 19 0 0 16 -4 16 0
4 14 -5 25 26 6 36 -30
5 30 11 121 23 3 9 33
6 19=A 0 0 27 7 49 0
7 27 8 64 19 -1 1 -8
8 30 11 121 20=A 0 0 0
9 20 1 1 16 -4 16 -4
10 28 9 81 11 -9 81 -81
11 11 -8 64 21 1 1 -8

 dx=42   dy=31   dx
2
dx =602 dy2=723 dy=137

( d x )(  d y )
 dxd y 
We know, coefficient of correlation(r) = N
( d x ) 2
((  d y ) 2
 d x2   d y2 
N N

42  31
137 
= 11
(42) 2 (31) 2
602  723 
11 11

137  118.36 18.64 18.64

= = = =0.035 (Ans).
(602  160.36)(723  87.36) 441.64  635.64 529.83

(i) Calculation of Rank correlation coefficient

Mizan sir 01911-971203, 01627-360073 12

Ranking No. Method A Rank Method B(y) Rank (Rx- Ry)
(x) according to according to
method method D2
A(Rx) B(Ry)
1 24 6 37 11 25
2 29 9 35 10 1
3 19 3.5 16 2.5 1
4 14 2 26 8 36
5 30 10.5 23 7 12.25
6 19 3.5 27 9 30.25
7 27 7 19 4 9
8 30 10.5 20 5 30.25
9 20 5 16 2.5 6.25
10 28 8 11 1 49
11 11 1 21 6 25

N= 11  D2=225

We know, Rank of correlation coefficient (R)

1 1 1
6{ D 2  (m13  m1 )  (m23  m2 )  (m33  m3 )
= 1- 12 12 12
N N
3

1 3 1 1 1 1 1
6{225  (2  2)  (2 3  2)  (2 3  2) 6(225   6   6   6)
= 1- 12 12 12 =1- 12 12 12
11  11
3
1331  11

6(225  0.5  0.5  0.5)

=1-
1320

1359
= 1- = 1- 1.029 = -0.029
1320
Mizan sir 01911-971203, 01627-360073 13
8. 2013: (a) Find the co-efficient of correlation from the following table and
interpret it:-

Salary 100 101 103 102 104 99 97 98 96 96

Cost of 98 99 99 97 95 92 95 94 90 91
living
Solution: Let, Salary=x and Cost of living=y

Calculation for Correlation Coefficient

Salary(x ) Cost of living(y ) x2 y2 xy
100 98 10000 9604 9800
101 99 10201 9801 9999
103 99 10609 9801 10197
102 97 10404 9409 9894
104 95 10816 9025 9880
99 92 9801 8464 9108
97 95 9409 9025 9215
98 94 9604 8836 9212
96 90 9216 8100 8640
96 91 9216 8281 8736

 x =996  y =950  x 2 =99276  y 2 =90346  xy =94681

 x y
 xy  N
r=
( x ) 2
( y ) 2

{ x 
2
}{  y  2
}
N N

996  950
94681 
10 61 61
= =  =0.72
9962
950 2
(74.4)(96) 85.5127
(99276  )(90346  )
10 10

Comment: r =+0.72 shows perfectly positive co-rrelated between salary and

leaving cost.
9. 2014: (a) Consider the following index numbers of security prices at Dhaka
Stock Exchange :- 4

Mizan sir 01911-971203, 01627-360073 14

Year 1995 1996 1997 1998 1999

Debenture price 99.2 98.8 98.3 98.4 97.1

Share price 85.8 78.9 75.8 77.2 83.8

Determine the rank correlation co-efficient (or spearman’s rank correlation co-
efficient) between debenture and share price.

Solution: Calculation for Rank correlation

Year Debenture Share (y) Rank x Rank y D =Rx-Ry D2

(x)
1995 99.2 85.5 5 5 0 0
1996 98.8 78.9 4 3 1 1
1997 98.3 75.8 2 1 -1 1
1998 98.4 77.2 3 2 1 1
1999 97.1 83.8 1 4 -3 9
D 2
=12

6  D2 6  12 72
We know, R= 1- = 1- 3 =1- = 1- 0.6 = 0.4
N N
3
5 5 120

here R= 0.4 it has lower degree of positive correlation between debenture and share.

10. (2016) :-Find Rank coefficient of correlation between marks in accountancy and marks
in statistics for the following data:- 4

Marks in 15 20 30 27 40 30 50 58
Accountancy

Marks in Statistics 40 35 48 58 38 40 48 40

Solution: Calculation for Rank of correlation coefficient

Mizan sir 01911-971203, 01627-360073 15
Marks in Marks in Rank of x Rank of y D =R1-R2 D2
Accountancy Statistics (R1) (R2)
(x) (y)
15 40 1 4 -3 9
20 35 2 1 1 1
30 48 4.5 6.5 -2 4
27 58 3 8 -5 25
40 38 6 2 4 16
30 40 4.5 4 0.5 0.25
50 48 7 6.5 0.5 0.25
58 40 8 4 4 16
D 2
=71.50
We know, Rank of correlation coefficient (R)
1 1 1
6{ D 2  (m13  m1 )  (m23  m2 )  (m33  m3 )}
= 1- 12 12 12
N N
3

The item 30 is repeated 2 times in series x and hence m1=2. In series y, the item 40
occurs 3 times and 48 item 2 times and hence m2=3 and m3=2.
1 3 1 1 1 1 1
6{71.5  (2  2)  (33  3)  (2 3  2)} 6(71.5   6   24   6)
= 1- 12 12 12 =1- 12 12 12
8 8
3
512  8
6(71.5  0.5  2  0.5)
=1-
504

447
= 1- = 1- 0.89 = 0.11
504

Here R= 0.11 it has lower degree of positive correlation between Marks in Accountancy
and Marks in Statistics.

Ok-12.02.2020

Mizan sir 01911-971203, 01627-360073 16