Scribd
Upload
69 views

Asif Hanif: Analysis With Chi-Square

- Chi-square tests are used to determine if there is an association between qualitative variables. Pearson's chi-square is calculated for a 2x2 table by comparing observed and expected counts in each cell. Fisher's exact test is used when expected counts are less than 5. - Several examples are provided of chi-square tests conducted on medical and survey data to analyze relationships between variables like treatment outcomes, television ownership preferences, and malocclusion in infants. - Additional tests discussed include Kendall's tau-b for ordinal data, contingency coefficients for effect size, and Phi and Cramer's V for nominal associations.

Uploaded by

Asad Chaudhary
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 PPT, PDF, TXT or read online on Scribd
69 views

Asif Hanif: Analysis With Chi-Square

- Chi-square tests are used to determine if there is an association between qualitative variables. Pearson's chi-square is calculated for a 2x2 table by comparing observed and expected counts in each cell. Fisher's exact test is used when expected counts are less than 5. - Several examples are provided of chi-square tests conducted on medical and survey data to analyze relationships between variables like treatment outcomes, television ownership preferences, and malocclusion in infants. - Additional tests discussed include Kendall's tau-b for ordinal data, contingency coefficients for effect size, and Phi and Cramer's V for nominal associations.

Uploaded by

Asad Chaudhary
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 PPT, PDF, TXT or read online on Scribd
You are on page 1/ 14

ASIF HANIF

Analysis with Chi-Square


Pearsons Chi-Square: (2 X 2)
Chi-Square test is employed to determine if there
is an association between qualitative variables.
When the word association is used in the
statistical sense, a comparison is implied. For a 2
X 2 table chi-square statistic is calculated as:
( )
( )( )( )( )
2
2
n ad bc
a b c d a c b d
_

=
+ + + +
Example
The following represent mortality data for two
groups of patients receiving different
treatments, A and B.


Outcome
Dead Alive
Treatment A 41 216
B 64 180
Example
Classification Men Women Total

Want T.V 80 120 200
Dont Want T.V 170 130 300
Total 250 250 500
A random sample of 250 men and 250 women
were pooled as to their desire concerning the
ownership of television sets. The following data
results
Formation of Hypothesis
H
o
: The two variables of classification are
independent
H
1
: The two variables of classification are
not independent
Level of Significance
= 0.05
Decision
P-Value = .0000
Hence P- Value is less than , so we reject H
o
and
we may conclude that variables of classification
depends upon each other.


Example
The following data (as above) describe the state of grief of 66
mothers who had suffered a neonatal death. The table relates this to
the amount of support given to these women:

Support
Good Adequate Poor
Grief State
I 17 9 8

II
6 5 1

III
3 5 4

IV
1 2 5
Fishers Exact Test:
When expected count is less than 5 we use
this techniques
1 2 1 2
! ! ! !
'
! ! ! ! !
R R C C
Fisher s Exact Test
n a b c d
=
Example

The following data relate to suicidal feelings in
samples of psychotic and neurotic patients:

Psychotics Neurotics Total
Suicidal feelings
2 6 8
No suicidal feelings
18 14 32
Total
20 20 40
Example
The following data compare malocclusion of teeth
with method of feeding infants.

Normal teeth

Malocclusion

Breast fed

4 16 =20
Bottle fed

1
5
21 = 22
37 = 42
Pearsons Chi-Square: (RXC)
For a R x C table the Chi-Square Statistics can
be calculated as:



And Yates Corrected Chi-Square for R X C
contingency table can be calculated as:
( )
2
2
ij ij
i j
ij
O E
E
_

=

2
2
0.5
ij ij
ij
O E
E
_
(


=

Co-efficient of Contingency
Contingency coefficient. A measure of association based
on chi-square. The value ranges between zero and 1,
with zero indicating no association between the row and
column variables and values close to 1 indicating a high
degree of association between the variables.




2
2
C
n
_
_
=
+
Phi & Cramers V
The Phi coefficient is a degree of association
between two attributes and is calculated as:


( )( )( )( )
2
ad bc
Phi
n
a b c d a c b d
_
|

= = =
+ + + +
Kendalls Tau b
Kendall's tau-b: This test is used to measure strength of
correlation when we have ordinal data. The sign of the
coefficient indicates the direction of the relationship, and its
absolute value indicates the strength, with larger absolute
values indicating stronger relationships. Possible values
range from -1 to 1, but a value of -1 or +1 can only be
obtained from square tables.







Where S=P-Q
P=Concordant pairs of observation
Q=Discordant pairs of observation
m=min(r,c)

( )( )
0 0
b
S
P Q X P Q Y
t =
+ + + +
Example
An animal epidemiologist tested dairy cows for the presence of a
bacterial disease. The disease is detected by the analysis of
blood samples, and the disease severity for each animal was
classified as None (1), Low (2) and High (2). Moreover, the size
of the herd that each cow belongs to a category is classified as
Large (3), Medium (2) and Small (1). The number of animals in
each of the 9 cells are recorded as:
Size of the
herd
None (1) Low (2) High (3) Total
Small (1) 9 5 9 23
Medium (2) 18 4 19 41
Large (3) 11 88 136 235
Total 38 97 164 299

You might also like