Scribd
Upload
51 views4 pages

Spearman's Rank Correlation Coefficient: Idea

Spearman's Rank Correlation Coefficient is a measure of the strength of the correlation between two variables. It ranges from -1 to 1, with -1 indicating a perfect negative correlation, 0 indicating no correlation, and 1 indicating a perfect positive correlation. The coefficient is calculated by first ranking the data values for each variable, then determining the differences between ranks and using this to calculate the coefficient value. Interpreting the size of the coefficient depends on the context, but generally values closer to -1 or 1 indicate stronger correlations while values closer to 0 indicate weaker or no correlation.

Uploaded by

Ali Sajjad
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOC, PDF, TXT or read online on Scribd
51 views4 pages

Spearman's Rank Correlation Coefficient: Idea

Spearman's Rank Correlation Coefficient is a measure of the strength of the correlation between two variables. It ranges from -1 to 1, with -1 indicating a perfect negative correlation, 0 indicating no correlation, and 1 indicating a perfect positive correlation. The coefficient is calculated by first ranking the data values for each variable, then determining the differences between ranks and using this to calculate the coefficient value. Interpreting the size of the coefficient depends on the context, but generally values closer to -1 or 1 indicate stronger correlations while values closer to 0 indicate weaker or no correlation.

Uploaded by

Ali Sajjad
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOC, PDF, TXT or read online on Scribd
You are on page 1/ 4

Spearmans Rank Correlation Coefficient

The Spearmans Rank Correlation Coefficient is a different way of describing the strength of the correlation between two quantities. Idea: Consider these three scatter graphs:
#$$ "$ !$ $ 2$ 2 ! " #$ #2 #

#$$ "$

& 2 # % #$ #% 2$ 2%

!$ $ 2$ 2 ! " #$ #2 #

'n each of these scatter graphs( y increases each time that x increases. Spearmans Rank Correlation Coefficient is defined to be # in each of these cases. )ow consider the following three scatter graphs:
#$$
"$ !$ $ 2$ 2 ! " #$ #2
!$ %$ $ &$ 2$ #$ 2 ! " #$ #2 #

"$ !$ $ 2$ 2 ! " #$

'n each of these( y decreases each time that x increases. Spearmans Rank Correlation Coefficient is defined to be *# in each case. Sometimes there is no connection between the two quantities( as in the following scatter graph
#$$ "$ !$ $ 2$ 2 ! " #$

'n this case( Spearmans Rank Correlation Coefficient would be close to $.

www.schoolworkout.co.uk

Calculating the value of Spearmans Rank Correlation Coefficient


There are two methods that can be used to find the +alue of Spearmans rank correlation coefficient ,rS-. They are best illustrated through an e.ample. Example: " students take tests in /rench and Science: /rench 0 Science 0 !" %# % 2! #1 &2 22 "% %$ !2 2% 12 2 &2 %1

Method 1: Step 1: R3)4 both sets of data. 5i+e the largest +alue rank #( the second largest +alue rank 2( etc. Step : Calculate the correlation coefficient of the ranked data

/rench 0 Science 0 /rench rank( x Science rank( y

!" %#

% 2!

#1 &2

22 "%

%$ !2

2%

12 2

&2 %1

x = x
2

y= y
S xy = S yy =
2

xy =
so S xx =

So Spearmans rank correlation coefficient rS 6 !ote: This +alue represents

!"#E: 7hen the data is ranked( you could gi+e the smallest +alue rank #( the second smallest rank 2( etc. 't doesnt matter which way you rank your data so long as you 89 T:; S3<; T:')5 /9R =9T: S;TS 9/ 83T3. !"#E : Interpreting the Si$e of the Correlation Coefficient
*# *$." *$.! *$. *$.2 $ >ittle correlation $.2 $. $.! 7eak positi+e correlation $." #

Strong negati+e 7eak negati+e correlation correlation

Strong positi+e correlation

:owe+er( you must interpret in the conte.t of the question.

www.schoolworkout.co.uk

Method : ?se of the formula: rS = # ! d2 n,n 2 #-

Example %continued& Step 1: R3)4 both sets of data. 5i+e the largest +alue rank #( the second largest +alue rank 2( etc. Step : Calculate the differences in the ranks( d. Step ': 7ork out the squares of the differences ,d2-. Step (: Calculate the sum of these 2 squared differences( d Step ): Spearmans Rank Correlation Coefficient is found by substituting this sum into the following formula: ! d2 # n,n 2 #where n is how many pairs of data you ha+e.
/rench 0 Science 0 /rench rank( x Science rank( y d

!" %#

% 2!

#1 &2

22 "%

%$ !2

2%

12 2

&2 %1

d2

! d2

n,n 2 #-

Example : 2 @udges mark the performances gi+en by 2 contestants in a talent competition: Contestant Audge # Audge 2 3 # #$ # #$ = #2 1 #2 1 C #2 #& #2 #& 1 # 8 1 # #% " #$ #% ; #% " ## #2 / #$ #% 5 ## #2

Audge # Audge 2 Audge # rank Audge 2 rank d

d2 So rS = # ! d2 *

n,n 2 #-

This +alue for the correlation shows that B www.schoolworkout.co.uk &

Examination style +uestion: 3s part of a tele+ision phone*in( +iewers were asked to rank 2 songs ,3( =( B( 5- in order of preference. Tanyas order of preference ,fa+ourite song first- was C( ;( =( 3( /( 8( 5. Aims order of preference was ;( 3( C( /( 8( =( 5. Calculate the +alue of Spearmans rank correlation coefficient between Tanyas and Aims rankings. Comment on its +alue. The +alue of Spearmans rank correlation coefficient between Tanyas and Colins rankings is *#. 7rite down Colins preference order for the 2 songs. Solution: Song Tanyas rank Aims rank 3 = C 8 ; / 5

www.schoolworkout.co.uk

You might also like