How to calculate spearman's rank coefficient of correlation.
PRELIMINARIES
This
post is designed to help students in their higher learning units that include
correlation as topic. The concept of correlation is tested in all examinations
national and international. This post is prepared by a professional Educator
and can be used as lecture notes.
What is
Correlation?
Correlation
is the degree of relationship between to variables say X and Y or Independent
and Dependent Variables.
Whenever
there is some definite connection existing between two or more groups, classes
or series, there is said to be correlation.
Why is it
important to study correlation?
1.
Correlation
shows the degree of relationship between variables such as demand and supply,
income and expenditure etc
2.
The
knowledge of correlation helps to locate critically important variables on
which others depend.
3.
Progressive
development of methods of science and philosophy has been characterized by the
increase in the knowledge of relationship among variable.
4.
It
clearly shows the cause effect relationship which has been used to solve many
problems and decision making.
Types
of correlation
1.
Positive
and negative correlations.
-
Correlation
is positive if two series /variables move in the same direction. For example,
supposing that Demand and supply are two variables, and an increase in demand
causes an increase in supply, we can conclude that demand and supply are
positively correlated.
-
However,
correlation is said to be negative if two series / variables move in the
opposite direction. For instance, the above mentioned scenario, an increase in
demand leads to a decrease in supply,
then we can conclude that demand and supply are negatively correlated.
2.
Linear
and non-linear correlations.
-
Correlation
is linear if the amount of change in one variable bears a constant change in
the amount of change in the other variable. However, correlation is Non–linear if
the amount of change in one variable does not bear a constant ratio to the
amount of change in the other variable.
3.
Simple,
partial and multiple correlations.
-
When
only two variables are studied it is a simple correlation problem.
-
When
more than two variables are studied, the problem is either partial or multiple correlations.
What are methods of studying correlation?
a)
Scatter
diagrams method
b)
Karl
Pearson’s coefficient of correlation.
c)
Spearman’s
rank correlation coefficient.
d)
Method
of least squares.
A)
SPEARMAN’S RANK
CORRELATION COEFFICIENT
This
method was developed by a British psychologist, Charles Edward Spearman in
1904.
- RANK
COEFFICIENT OF CORRELATION is defined as the correlation between the ranks
assigned to individuals in two characters
The
formula is: (insert formula)
Where
R= rank coefficient of correlation.
d = difference between ranks
n = number of pairs of ranks.
-
In rank correlation there are two types of problems.
a) Where actual ranks are given.
b) Where ranks are not given
1.
WHERE ACTUL RANKS ARE GIVEN.
ILLUSTRATION
Compute
the rank coefficient of correlation and comment on the value.
Employee Rank 1 Rank 2
A 10 9
B 2 4
C 1 2
D 4 3
E 3 1
F 6 5
G 5 6
H 8 8
I 7 7
J 9 10
SOLUTION
- Calculate the differences between
ranks ie d= (rank 1(R1) – rank 2(R2))
- Square the differences to obtain d²
- apply the formula to calculate R
THUS.
(d
= RI-R2) d²
1
1
-2 4
-1 1
1 1
2 4
1 1
-1 1
0 0
0 0
-1 1
∑d²=
14
B)
WHERE
RANKS ARE NOT GIVEN.
-Ranks can be assigned by giving the highest value 1 or the smallest value 1. In our illustrations we have assigned the smallest value 1.
ILLUSTRATION
1. The following are the scores awarded by two assessors to eight employees based on their work ethics.
Employee Assessor X Assessor Y
A 65 50
B 70 48
C 55 72
D 52 80
E 48 62
F 77 70
G 83 65
H 45 59
i)
Determine Spearman’s rank coefficient of correlation
ii)
Interpret the results in (i) above.
SOLUTION
-
Assign ranks to the scores by assigning the smallest value 1 (one)
Employee Assessor RANK 1
Assessor RANK 2
X (R1) Y (R2)
A 65 5 50 2
B 70 6 48 1
C 55 4 72 7
D 52 3 80
8
E 48 2 62
4
F 77 7 70
6
G 83 8 65 5
H 45 1 59 3
∑d²= 86


Comments
Post a Comment