Scribd
Upload
14 views2 pages

Measure of Correlation

The document discusses the differences between Karl Pearson's method and Spearman's rank method for calculating correlation, highlighting their applicability to quantitative and qualitative data respectively. It outlines the limitations of the rank method, particularly in large datasets, and emphasizes the precision of Pearson's method. Additionally, it provides various formulas for calculating Pearson's coefficient of correlation.

Uploaded by

aakash kumar
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
14 views2 pages

Measure of Correlation

The document discusses the differences between Karl Pearson's method and Spearman's rank method for calculating correlation, highlighting their applicability to quantitative and qualitative data respectively. It outlines the limitations of the rank method, particularly in large datasets, and emphasizes the precision of Pearson's method. Additionally, it provides various formulas for calculating Pearson's coefficient of correlation.

Uploaded by

aakash kumar
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

Statistics for S

--
11.36
--------· ---~ftii"~
~
Demerits
• in a bivariate (grouped) fr
1· Rank method cannot be used for finding out correlation equet\~
distribution.
2 If difficult task to ascertain the trlnL_
· the number of values is quite large, it becomes a -~ at\q
. . thi5 method only when the nuinb
therr differences. That is why it is advisable to use er of
observations is less than 30.
3 Thi p n' 5 method. It uses ranks •
· s method lacks precision as compared to Kar1 earso lllstead
of the original values. •

Karl Pearson's Method Vs Spearman's Rank Method


1
The coefficient of correlation by both the methods ranges between - l and + • Still, there exist
fue following differences:
1· Karl Pearson's method of Correlation measures correlation for quantitative data, Whereas
Spearman's method of rank correlation measures coefficient of correlation for qualitative
data.
2· Karl Pearson's method calculates deviations from actual or assumed mean, whereas Spearman'8
method calculates the rank differences.
3. Rank correlation gives less importance to the extreme values because it gives them rank.
However, Karl Pearson's method of correlation gives more importance to extreme values as
it is based on actual values.
FORMULAE AT A ·GtANce· -
Karl Pearsons Coefficient of Correlation

r = Ixy - "i.xy Ixy


1. Actual Mean Method Nxaxxay N x ~ x {R- ✓I.x2xiy2
· \}N \}N
NIXY-IX.IY
2. Direct Method ✓ N~I~X2~-=(I=X~)2 ~x-✓rN=I~=-=(IY=)=2
r=r.

r= Nidxdy - Idx x Idy


3. Short- Cut Method
✓ Nidx2- (Idx)2 x ✓ Nidy2 _ (Idy)2

r= Nidx'dy' - Idx' x Idy'


4. Step Deviation Method
✓ Nidx' - (Idx')2 x ~ Nidy'2 _ (Idy')2
2
ot Equal

,cs are Equal


~ ~sn
f"

't'oOS used
~13 I
i8 "' l(arl Pearson's Coefficient
! of Correlation.
~~
Number of pair of observations
.
r. ~ Deviation of X series from
mean (X-X).
y ~ ~eviation ~f Y series from mean
('{ _ Y).
' '

\ :: Standard deviation of Xser


G){ ies•i e ~ tx2 .
' ' ., N

vy . Standard deviation of'{ series i e ~ ty2.


·" ·==
' ••, N

tdx :: Sum of deviations· of Xvalue


s from assumed mean.
tdy = Sum of deviations of Yvalue
s from assumed mean.
2
tc1x = Sum of squared deviations
of Xvalues fro m assumed mearu
tdy2. = Sum of squared deviatio
ns of Y. values from assumed me
an.
tdxdy = Sum of the products of
deviations dx and dy.
tdx' • = Sum of step deviations
of Xvalues from as s~ ed mean.
tdy' = Sum of step deviations of
Y. values from assumed mean.
Idx12 = Sum of squared step dev
iations of Xvalues from assumed
mean.
Idy'2 =Sum of squared step devia
tions of Y. values from assumed me
an.
tdx'dy' = Sum of the products of
step deviations dx' and dy'.
rk = Coefficient of rank
correlation.
I.D2 = Sum of square of rank
differences.
m = Number of times an item is assigned equal
rank.

You might also like