Scribd
Upload
208 views

Correlation and Scatter Graphs

This document discusses correlation and scatter graphs. It defines positive, negative, and zero correlation between two variables based on how the variables change together. It also discusses line of best fit, rank correlation coefficient, and interpreting the magnitude and significance of correlation coefficients. Examples are provided to demonstrate calculating correlation and testing significance.

Uploaded by

kaziba stephen
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
208 views

Correlation and Scatter Graphs

This document discusses correlation and scatter graphs. It defines positive, negative, and zero correlation between two variables based on how the variables change together. It also discusses line of best fit, rank correlation coefficient, and interpreting the magnitude and significance of correlation coefficients. Examples are provided to demonstrate calculating correlation and testing significance.

Uploaded by

kaziba stephen
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 7

TOPIC 2: Correlation and scatter graphs

Correlation is the relationship between two variables. The measure of the


degree of the relationship between two variables the is correlation coefficient

Scatter graphs:

Is a graphical representation of two variables for a given set of data.

Types of correlation

a) Positive/direct correlation (A): If one variable, x, increases as the other,


y, increases, or, as x decreases, y also decreases then the relationship
between the two variables is said to positive.
b) Negative / inverse correlation(B): If one variable, x, increases as y
reduces, or, x reduces as y increases then the relationship between two
variables is said to be negative
c) Zero/no correlation (C ): If an increase or decrease in one variable does
not affect the other, then the variables are said to have a zero
correlation.
y y
y

x x x

Line of best fit


This refers to a line through a scatter plot of data points that best
expresses the relationship between these points. To draw this line the
following steps are followed:
Find the mean point M  X , Y  , where X   and Y  
X Y
i)
n n
Where n is the number of pairs of X and Y
ii) Draw the line through the mean point ensuring that there are
equal points left on either side of the line
Rank correlation coefficient:
It is the measure of the degree of the relationship between two
variables. It is determined on the basis of ranks. There many
correlation coefficients. Spearman’s correlation coefficient is
more preferred.
It is given by
6 d 2
  1

n n2  1 
Where n=number of pairs of the variables
d= difference between the ranks
Interpretation of magnitude of correlation coefficient:
a) Using magnitude and direction
Correlation coefficient comment

0 – 0.19 Very low

0.2 – 0.39 Low

0.40 – 0.59 Moderate

0.60 – 0.79 high

0.80 – 1.0 Very high

When giving a comment it should have magnitude and direction.


For example if   0.7 , the comment should be high (magnitude),
positive (direction) correlation.

b) Using level of significance:


After calculating the correlation coefficient,  C , the comment is
made basing on the standard value of significance levels for
correlation coefficient:( use uneb tables)
i) Determine the corresponding standard value of
Spearman’s correlation coefficient at either 5% or 1%
level of significance from the tables
ii) Compare the two values. If the magnitude of the
calculated value,  C , exceeds the standard value then
correlation is significant otherwise not.

Example 1:

The height (cm) and ages (years) of a random sample of ten farmers are given
in the table below.

Heights, 156 151 152 160 146 157 149 142 158 140
x(cm)

Ages, y 47 38 44 55 46 49 45 30 45 30
years

a) (i) Calculate the rank correlation coefficient


(ii) Comment on your result at 5% level of significance
b) Plot a scatter diagram for the data and comment on the relationship of
the data. Draw a line of best fit
c) Use your graph to find Y when X = 147
Solution:
a)
RX RY d d2

4 3 1 1

6 8 -2 4

5 7 -2 4

1 1 0 0

8 4 4 16
3 2 1 1

7 5.5 1.5 2.25

9 9.5 -0.5 0.25

2 5.5 -3.5 12.25

10 9.5 0.5 0.25

d 2
 41

6 d 2 6  41
i)  1  1  0.7515

n n 1
2
 10(10 2  1)
ii) Since  C  0.7515  T  0.65 , correlation is significant at 5%
Or There is a high positive Correlation between heights and ages.
 42.9 . Mean point X , Y   151.1,42.9
1511 429
b) X   151.1, Y 
10 10
Comment: There is a positive linear correlation between X and Y
c) Y= 37, X=147
Assignment 2.1.2

1. The table below shows scores by candidates in ICT and Mathematics.


ICT 𝐹9 𝑃8 𝐶6 𝐶4 𝐷1 𝐶5 𝐶4 𝐶4
Maths F E O D A E B C
a.) Calculate the rank correlation for the data.
b.) Test whether the correlation is significant at 5% level of significance.

2. Three examiners X, Y, and Z each marked the scripts of ten candidates who
sat Mathematics examination. The table below shows the examiner’s
ranking of the candidates.

CANDIDATE

EXAMINER A B C D E F G H I J

X 8 5 9 2 10 1 7 6 3 4

Y 5 3 6 1 4 7 2 10 8 9

Z 6 3 7 2 5 4 1 10 9 8

Calculate the coefficient of rank correlation of the rankings between

(i) X and Y
(ii) Y and Z.
iii) State with reason whether there is significant difference between
rankings of the three examiners.

3. The course work grades ranging from A to G and examinations marks of 8


candidates are given below.

Course A B C D E F G H
work
grade
Exam 92 75 63 54 48 45 34 18
mark
(a) Calculate the spearman’s rank correlation coefficient for the two sets of
data

(b) Test whether it is significant at 5% level.

You might also like