Rank correlation, Applied Statistics

Rank Correlation

Sometimes the characteristics whose possible correlation is being investigated, cannot be measured but individuals can only be ranked on the basis of the characteristics to be measured. We then have two sets of ranks available for working out the correlation coefficient. Sometimes the data on one variable may be in the form of ranks while the data on the other variable are in the form of measurements which can be converted into ranks.

Thus, when both the underlying variables are ordinal or when the data are available in the ordinal form irrespective of the type of variable, we use the rank correlation coefficient to find out the extent of relationship between the two variables.

This rank correlation coefficient is also known as Spearman's rank correlation coefficient. Rank correlation is a non-parametric technique for measuring strength of relationship between paired observations of two variables when the data are in a ranked form.

Rank Correlation measures the degree of agreement between the two sets of ranks.

Rank Correlation Coefficient R is

  1153_rank correlation.png

where,   n   =      Number of individuals ranked

             Di   =       Difference in the ranks of the ith individual

             i     =       1, 2, ........ n

 

             Di   =       Difference in the ranks of the ith individual

             i     =       1, 2, ........ n

Example

The ranks of 8 students are given according to their marks in English and History.

Student No.

1

2

3

4

5

6

7

8

English

2

8

1

7

6

4

5

3

History

7

3

2

6

5

8

4

1                                          

 

We want to know whether or not students who are good in English are also good in History. For this the calculations are:

1531_rank correlation1.png

 

Here

1509_rank correlation2.png

= 74 and n = 8

 

R

  706_rank correlation3.png
  = 0.119  

The correlation between the ranks is seen to be low.

Posted Date: 9/15/2012 4:59:18 AM | Location : United States







Related Discussions:- Rank correlation, Assignment Help, Ask Question on Rank correlation, Get Answer, Expert's Help, Rank correlation Discussions

Write discussion on Rank correlation
Your posts are moderated
Related Questions
A researcher hypothesized that the pulse rates of long-distance athletes differ from those of other athletes. He believed that the runners’ pulses would be slower. He obtained a ra

Test for Equality of Proportions For example, we may want to test whether the percentage of smokers (p 1 ) among the males equals the percentage of female smokers (p 2 ). W


how to interpret results, a good explanation to help me understand.

Determine the maximum weight in kN to one decimal point (1 DP) of the engine that can be supported without exceeding the tension given in  Parameter 1 (P1) in chain AB or 1.1 x P1

Regression Coefficient While analysing regression in two related series, we calculate their regression coefficients also. There are two regression coefficients like two regress

introduction of median

In PCA the eigknvalues must ultimately account for all of the variance. There is no probability,'no hypothesis, no test because strictly speaking PCA is not a statistical procedure

A study was designed to investigate the effects of two variables - (1) a student's level of mathematical anxiety and (2) teaching method - on a student's achievement in a mathemati

Consider the sample of 60 package design ratings given in the table below.                                    A Sample of Package Design Ratings                 (Composite S