1

Chapter 5

CORRELATION AND REGRESSION

5.1: Introduction

So far we have confined our discussion to the distributions involving

only one variable. Sometimes, in practical applications, we might come

across certain set of data, where each item of the set may comprise of the

values of two or more variables.

Suppose we have a set of 30 students in a class and we want to

measure the heights and weights of all the students. We observe that each

individual (unit) of the set assumes two values – one relating to the height

and the other to the weight. Such a distribution in which each individual or

unit of the set is made up of two values is called a bivariate distribution. The

following examples will illustrate clearly the meaning of bivariate

distribution.

(i) In a class of 60 students the series of marks obtained in two

subjects by all of them.

(ii) The series of sales revenue and advertising expenditure of two

companies in a particular year.

(iii) The series of ages of husbands and wives in a sample of selected

married couples.

Thus in a bivariate distribution, we are given a set of pairs of

observations, wherein each pair represents the values of two variables.

In a bivariate distribution, we are interested in finding a relationship

(if it exists) between the two variables under study.

The concept of ‘correlation’ is a statistical tool which studies the

various methods and techniques used for studying and measuring the extent

of the relationship between the two variables.

“Two variables are said to be in correlation if the change in one of the

variables results in a change in the other variable”. 2

5.2: Types of Correlation

There are two important types of correlation. They are (1) Positive

and Negative correlation and (2) Linear and Non – Linear correlation.

5.2.1: Positive and Negative Correlation

If the values of the two variables deviate in the same direction i.e. if

an increase (or decrease) in the values of one variable results, on an average,

in a corresponding increase (or decrease) in the values of the other variable

the correlation is said to be positive.

Some examples of series of positive correlation are:

(i) Heights and weights;

(ii) Household income and expenditure;

(iii) Price and supply of commodities;

(iv) Amount of rainfall and yield of crops.

Correlation between two variables is said to be negative or inverse if

the variables deviate in opposite direction. That is, if the increase in the

variables deviate in opposite direction. That is, if increase (or decrease) in

the values of one variable results on an average, in corresponding decrease

(or increase) in the values of other variable.

Some examples of series of negative correlation are:

(i) Volume and pressure of perfect gas;

R=;

(iii) Price and demand of goods. 3

Graphs of Positive and Negative correlation:

Suppose we are given sets of data relating to heights and weights of

students in a class. They can be plotted on the coordinate plane using x –

axis to represent heights and y – axis to represent weights. The different

graphs shown below illustrate the different types of correlations.

Figure for positive correlation

x4

Figure for negative correlation

Note:

(i) If the points are very close to each other, a fairly good amount of

correlation can be expected between the two variables. On the

other hand if they are widely scattered a poor correlation can be

expected between them.

(ii) If the points are scattered and they reveal no upward or downward
trend as in the case of (d) then we say the variables are

uncorrelated.

(iii) If there is an upward trend rising from the lower left hand corner

and going upward to the upper right hand corner, the correlation

obtained from the graph is said to be positive. Also, if there is a

downward trend from the upper left hand corner the correlation

obtained is said to be negative.

(iv) The graphs shown above are generally termed as scatter

diagrams.

xx

x5

Example:1: The following are the heights and weights of 15 students of a

class. Draw a graph to indicate whether the correlation is negative or

positive.

Heights (cms) Weights (kgs)

170 65

172 66

181 69

157 55

150 51

168 63

166 61

175 75
177 72

165 64

163 61

152 52

161 60

173 70

175 72

Since the points are dense (close to each other) we can expect a high

degree of correlation between the series of heights and weights. Further,

since the points reveal an upward trend, the correlation is positive. Arrange

the data in increasing order of height and check that , as height increases, the

weight also increases, except for some (stray) cases..

EXERCISES

(1) A Company has just brought out an annual report in which the capital

investment and profits were given for the past few years. Find the

type of correlation (if it exists).

Capital Investment (crores) 10 16 18 24 36 48 57

Profits (lakhs) 12 14 13 18 26 38 62 6

(2) Try to construct more examples on the positive and negative

correlations.

(3) Construct the scattered diagram of the data given below and indicate

the type of correlation.

(Average Value in Lakhs of Rs.)

Years 1965 1970 1975 1980 1985 1990

Raw cotton import 42 60 112 98 118 132

Cotton manufacture exports 68 79 88 86 106 114

The correlation between two variables is said to be linear if the

change of one unit in one variable result in the corresponding change in the

other variable over the entire range of values.

For example consider the following data.

X 2 4 6 8 10

Y 7 13 19 25 31

Thus, for a unit change in the value of x, there is a constant change in

the corresponding values of y and the above data can be expressed by the

relation

y = 3x +1

In general two variables x and y are said to be linearly related, if

there exists a relationship of the form

y = a + bx

where ‘a’ and ‘b’ are real numbers. This is nothing but a straight line when

plotted on a graph sheet with different values of x and y and for constant

values of a and b. Such relations generally occur in physical sciences but are

rarely encountered in economic and social sciences. 7

The relationship between two variables is said to be non – linear if

corresponding to a unit change in one variable, the other variable does not

change at a constant rate but changes at a fluctuating rate. In such cases, if

the data is plotted on a graph sheet we will not get a straight line curve. For

example, one may have a relation of the form

y = a + bx + cx

or more general polynomial.

One of the most widely used statistics is the coefficient of correlation

‘r’ which measures the degree of association between the two values of

related variables given in the data set. It takes values from + 1 to – 1. If two

sets or data have r = +1, they are said to be perfectly correlated positively if

r = -1 they are said to be perfectly correlated negatively; and if r = 0 they

are uncorrelated.

The coefficient of correlation ‘r’ is given by the formula

( )( )

2222

n x ( x) n y ( y)

nxyxy

∑−∑∑−∑

∑ ∑− ∑

The following example illustrates this idea.

Example:2: A study was conducted to find whether there is any

relationship between the weight and blood pressure of an individual.

The following set of data was arrived at from a clinical study. Let us

determine the coefficient of correlation for this set of data. The first

column represents the serial number and the second and third columns

represent the weight and blood pressure of each patient. 8

S. No. Weight Blood Pressure

1. 78 140

2. 86 160
3. 72 134

4. 82 144

5. 80 180

6. 86 176

7. 84 174

8. 89 178

9. 68 128

10. 71 132

Solution:

xyx

xy

78 140 6084 19600 10920

86 160 7396 25600 13760

72 134 5184 17956 9648

82 144 6724 20736 11808

80 180 6400 32400 14400

86 176 7396 30976 15136

84 174 7056 30276 14616

89 178 7921 31684 15842

68 128 4624 16384 8704

71 132 5041 17424 9372

796 1546 63,776 243036 1242069

Then
[ ][ ]

22

10( ) 63776 (796) 10( () 243036) (1546)

10(124206) (796 () 1546)

−−

r =9

0 5966.

(1144 () 40244)

11444

5.4: Rank Correlation

Data which are arranged in numerical order, usually from largest to

smallest and numbered 1,2,3 ---- are said to be in ranks or ranked data..

These ranks prove useful at certain times when two or more values of one

variable are the same. The coefficient of correlation for such type of data is

given by Spearman rank difference correlation coefficient and is denoted

by R.

In order to calculate R, we arrange data in ranks computing the

difference in rank ‘d’ for each pair. The following example will explain the

usefulness of R. R is given by the formula

( )1

()
16

−=

nn

Example:3: The data given below are obtained from student records.

Calculate the rank correlation coefficient ‘R’ for the data.

Subject Grade Point Average (x) Graduate Record exam score (y)

1. 8.3 2300

2. 8.6 2250

3. 9.2 2380

4. 9.8 2400

5. 8.0 2000

6. 7.8 2100

7. 9.4 2360

8. 9.0 2350

9. 7.2 2000

10. 8.6 2260

Note that in the G. P. A. column we have two students having a grade

point average of 8.6 also in G. R. E. score there is a tie for 2000. 10

Now we first arrange the data in descending order and then rank

1,2,3,---- 10 accordingly. In case of a tie, the rank of each tied value is the

mean of all positions they occupy. In x, for instance, 8.6 occupy ranks 5 and
6. So each has a rank ;5.5

56

Similarly in ‘y’ 2000 occupies ranks 9 and 10, so each has rank

5.9.

9 10

Now we come back to our formula

( )1

−=

nn

We compute ‘d’ , square it and substitute its value in the formula.

Subject x y Rank of x Rank of y d d

1. 8.3 2300 7 5 2 4
2. 8.6 2250 5.5 7 -1.5 2.25

3. 9.2 2380 3 2 1 1

4. 9.8 2400 1 1 0 0

5. 8.0 2000 8 9.5 -1.5 2.25

6. 7.8 2100 9 8 1 1

7. 9.4 2360 2 3 -1 1

8. 9.0 2350 4 4 0 0

9. 7.2 2000 10 9.5 0.5 0.25

10. 8.6 2260 5.5 6 -0.5 0.25

So here, n = 10, sum of d

= 12. So

1 .0 0727 .0 9273

10 (100 )1

12(6 )

=−=

R=−

Note: If we are provided with only ranks without giving the values of x and

y we can still find Spearman rank difference correlation R by taking the

difference of the ranks and proceeding in the above shown manner.

11

EXERCISES

1. A horse owner is investigating the relationship between weight

carried and the finish position of several horses in his stable.

Weight Carried Position Finished

110 2

113 6

120 3

115 4

110 6

115 5

117 4

123 2

106 1

108 4

110 1

110 3

2. The top and bottom number which may appear on a die are as

follows

Top 1 2 3 4 5 6

bottom 5 6 4 3 1 2

Calculate r and R for these values. Are the results surprising?

3. The ranks of two sets of variables (Heights and Weights) are given

below. Calculate the Spearman rank difference correlation

coefficient R.

1 2 3 4 5 6 7 8 9 10

Heights 2 6 8 4 7 4 9.5 4 1 9.5

Weights 9 1 9 4 5 9 2 7 6 3 12

5.5: Regression

If two variables are significantly correlated, and if there is some

from the other. This observation leads to a very important concept known as

‘Regression Analysis’.

Regression analysis, in general sense, means the estimation or

prediction of the unknown value of one variable from the known value of the

other variable. It is one of the most important statistical tools which is

extensively used in almost all sciences – Natural, Social and Physical. It is

specially used in business and economics to study the relationship between

two or more variables that are related causally and for the estimation of

demand and supply graphs, cost functions, production and consumption

functions and so on.

Prediction or estimation is one of the major problems in almost all the

spheres of human activity. The estimation or prediction of future production,

consumption, prices, investments, sales, profits, income etc. are of very great

importance to business professionals. Similarly, population estimates and

population projections, GNP, Revenue and Expenditure etc. are

indispensable for economists and efficient planning of an economy.

Regression analysis was explained by M. M. Blair as follows:

“Regression analysis is a mathematical measure of the average relationship

between two or more variables in terms of the original units of the data.”

5.5.1: Regression Equation

Suppose we have a sample of size ‘n’ and it has two sets of measures,

denoted by x and y. We can predict the values of ‘y’ given the values of ‘x’

by using the equation, called the REGRESSION EQUATION.

y* = a + bx

where the coefficients a and b are given by 13

n
ybx

nxx

n xy x y

∑−∑

∑−∑

∑∑−∑

22

()()

( () )

The symbol y* refers to the predicted value of y from a given value of

x from the regression equation.

Example: 4 : Scores made by students in a statistics class in the mid -

term and final examination are given here. Develop a regression equation

which may be used to predict final examination scores from the mid – term

score.

STUDENT MID – TERM FINAL

1. 98 90

2. 66 74

3. 100 98

4. 96 88

5. 88 80

6. 45 62

7. 76 78
8. 60 74

9. 74 86

10. 82 80

Solution:

We want to predict the final exam scores from the mid term scores. So

let us designate ‘y’ for the final exam scores and ‘x’ for the mid – term exam

scores. We open the following table for the calculations. 14

Stud x y X

xy

1 98 90 9604 8820

2 66 74 4356 4884

3 100 98 10,000 9800

4 96 88 9216 8448

5 88 80 7744 7040

6 45 62 2025 2790

7 76 78 5776 5928

8 60 74 3600 4440

9 74 86 5476 6364

10 82 80 6724 6560

Total 785 810 64,521 65,071

Numerator of b = 10 * 65,071 – 785 * 810 = 6,50,710 – 6,35,850 = 14,860

Denominator of b = 10 * 64, 521 – (785)

= 6,45,210 – 6,16,225 = 28,985

Therefore, b = 14,860 / 28,985 = 0.5127

Numerator of a = 810 – 785 * 0.5127 = 810 – 402.4695 = 407.5305

Therefore a = 40.7531

Thus , the regression equation is given by

y* = 40.7531 + (0.5127) x

We can use this to find the projected or estimated final scores of the

students.

For example, for the midterm score of 50 the projected final score is

y* = 40.7531 + (0.5127) 50 = 40.7531 + 25.635 = 66.3881

which is a quite a good estimation.

To give another example, consider the midterm score of 70. Then the

projected final score is

y* = 40.7531 + (0.5127) 70 = 40.7531 + 35.889 = 76.6421,

which is again a very good estimation. 15

This brings us to the end of this chapter. We close with some problems for

you.

EXERCISES

1. The data given below are obtained from student records. Calculate the

regression equation and compute the estimated GRE scores for GPA = 7.5,

8.5..

Subject Grade Point Average (x) Graduate Record exam score (y)

11. 8.3 2300

12. 8.6 2250

13. 9.2 2380

14. 9.8 2400

15. 8.0 2000

16. 7.8 2100

17. 9.4 2360

19. 7.2 2000

20. 8.6 2260

2. A study was conducted to find whether there is any relationship between

the weight and blood pressure of an individual. The following set of data

was arrived at from a clinical study.

S. No. Weight Blood Pressure

1. 78 140

2. 86 160

3. 72 134

4. 82 144

5 80 180

6. 86 176

7. 84 174

8. 89 178

9. 68 128

10. 71 132 16

3. A horse was subject to the test of how many minutes it takes to reach a

point from the starting point. The horse was made to carry luggage of

various weights on 10 trials.. The data collected are presented below in the

table.

Trial No. Weight (in Kgs) Time taken (in mins)

1 11 13

2 23 22

3 16 16

4 32 47

5 12 13
6 28 39

7 29 43

8 19 21

9 25 32

10 20 22

Find the regression equation between the load and the time taken to reach

the goal. Estimate the time taken for the loads of 35 Kgs , 23 Kgs, and 9

Kgs. Are the answers in agrrement with your intuitive feelings? Justify.

### $$$ ###