Correlation and Chi-Square Test - LDR 280

A Review of
Widely-Used Statistical
Methods
REVIEW OF FUNDAMENTALS
When testing hypotheses,
all statistical methods will
always be testing the null.
Null Hypothesis?
No difference/no relationship
If we do not reject the null, conclusion?

Found no difference/no relationship
If we do decide to reject the null, conclusion?

A significant relationship/difference is found and reported
o The observed relationship/difference is too large to be
How do we decide to reject/not reject the null?

Statistical tests of significance always test the null and
always report
(sig. level)probability of erroneously rejecting a
true null based on sample data.
represents the odds of being wrong if we decide to
reject the null
the probability that null is in fact true and that
any apparent relationship/difference is a result
of chance/sampling error and, thus
the odds of being wrong if we report a significant
relationship/difference.
Rule of thumb for deciding to reject/not reject the null?
3
STATITICAL DATA ANALYSIS

COMMON TYPES OF ANALYSIS?
Examine Strength and Direction of Relationships
Bivariate (e.g., Pearson Correlationr)
Between one variable and another:
Y = a + b 1 x1
Multivariate (e.g., Multiple Regression Analysis)
Between one dep. var. and an independent variable, while

holding all other independent variables constant:
Y = a + b 1 x1 + b2 x 2 + b 3 x 3 + + b k x k
Compare Groups
Between Proportions (e.g., Chi Square Test 2)
H0 :
P 1 = P 2 = P3 = = P k
Between Means (e.g., Analysis of Variance)

H0 :
1 = 2 = 3 = = k
Lets first review some fundamentals.
Remember: Level of measurement determines choice of

statistical method.
Statistical Techniques and Levels of Measurement:
INDEPENDENT
DEPENDENT
NOMINAL/CATEGORICAL
N
O
M
I
N
A
L
M
E
T
R
I
C
* Chi-Square
* Fishers Exact Prob.
* T-Test
* Analysis of Variance
METRIC (ORDERED METRIC or

HIGHER)
* Discriminant Analysis
* Logit Regression
* Correlation (and Covariance) Analysis

* Regression Analysis
5
Correlation and Covariance:

Covariance Measures of Association
Between Two Variables
Often
Often we
we are
are interested
interested in
in the
the strength
strength and
and nature
nature of
of the
the
relationship
relationship between
between two
two variables.
variables.
Two
Two indices
indices that
that measure
measure the
the linear
linear relationship
between
continuous/metric
continuous/metric variables
variables are:
are:
a.
a.Covariance
Covariance
b.
b.Correlation
Correlation Coefficient
Coefficient (Pearson
(Pearson Correlation)
Correlation)
Covariance
Covariance
Covariance is
is aa measure
measure of
of the
the linear
linear association
association betw
betw
two
two metric
metric variables
variables (i.e.,
(i.e., ordered
ordered metric,
metric, interval,
interval, or
or ra
ra
variables).
variables).
Covariance
Covariance (for
(for aa sample)
sample) is
is computed
computed as
as follows:
follows:
( xi x )( yi y )
sxy
n 1
for
samples
Positive
Positive values
values indicate
indicate aa positive
positive relationship.
relationship.
Negative
Negative values
values indicate
indicate aa negative
negative (inverse)
(inverse) relationsh
relationsh
Covariance (sxy ) of Two Variables

Example: Golfing Study
A golf enthusiast is interested in investigating
the relationship, if any, between golfers
driving distance (x) and their 18-hole score (y).
He uses the following sample data (i.e., data
from n = 6 golfers) to examine the issue:
x =Average Drivingy = Golfers Average
18-Hole Score
Distance (yards.)
277.6
259.5
269.1
267.0
255.6
272.9
69
71
70
70
71
69
Covariance (sxy ) of two variables

x
277.6
259.5
269.1
267.0
255.6
272.9
sxy
( xi x )( yi y )
n 1
y (xi x) (yi y)
69 10.65
71 -7.45
70
2.15
70
0.05
71 -11.35
69
5.95
Average 267.0 70.0

Std. Dev. 8.2192.8944
-1.0
1.0
0
0
1.0
-1.0
(xi x)(yi y)
-10.65
-7.45
0
0
-11.35
-5.95
Total -35.40
n=6
(s
xn)1y635.4107.08
ii
xy
Covariance

Covariance:
What can we say about the relationship between the two variables?
The relationship is negative/inverse.
That is, the longer a golfers driving distance is, the lower (better)
his/her score is likely to be.
How strong is the relationship between x and y?
Hard to tell; there is no standard metric to judge it by!

Values of covariance depend on units of measurement for x and y.
WHAT DOES THIS MEAN?
(s
xn
)1
(y)6
3
5
.4
0
.10
7
8
x
yii
Covariance
Covariance
It means:
If driving distance (x) were measured in feet, rather than yards,
even though it is the same relationship (using the same data),
the covariance sxy would have been much larger. WHY?
Because x-values would be much larger, and thus ( xi x )
(
values will be much larger which,
in turn, will make
( xi x )( yi ymuch
)
larger.
SOLUTION: Correlation Coefficient comes to the rescue!
Correlation Coefficient (r) is a standard measure/metric for judging

strength of linear relationship that, unlike covariance,
is not affected by the units of measurement for x and y.
This is why correlation coefficient (r) is much more widely
used that covariance.
Correlation
(Pearson/simple correlatio
Coefficient rrxy
xy (Pearson/simple correlatio
is
is aa measure
measure of
of linear
linear association
association between
between two
two variabl
variab
ItIt may
may or
or may
may not
not represent
represent causation.
causation.
The
(for sample data) is
The correlation
correlation coefficient
coefficient rrxy
xy (for sample data) is
computed
computed as
as follows:
follows:
for
samples
rxy
s xy
sx s y
sxy = Covariance of x & y

sx = Std. Dev. of x
sy = Std Dev. of y
Correlation Coefficient = r
Francis Galton
(English researcher,
inventor of
fingerprinting, and
cousin of Charles
Darwin)
In1888, plotted lengths of forearms and head sizes to see to what

degree one could be predicted by the other.
Stumbled upon the mathematical properties of correlation plots
(e.g., y intercept, size of slope, etc.).
RESULT: An objective measure of how two variables are
co-related--CORRELATION COEFFICIENT (Pearson
Correlation), r.
r
Assesses the strength of a relationship based strictly on
empirical data, and independent of human judgment or opinion
13
Correlation Coefficient (Pearson Correlation) = r

What do you use it for?
Karl Pearson, a Galton

Student & the Founder
of Modern Statistics
To examine:
a. Whether a relationship exists between two metric
variables
e.g., income and education, or
workload and job satisfaction and
b. What the nature and strength of that relationship
may be.
Range of Values for r?
14
Correlation Coefficient (Pearson Correlation) rxy

-1
-1 <
< rr <
< +1.
+1.
r-values
r-values closer
closer to
to -1
-1 or
or +1
+1 indicate
indicate stronger
stronger linear
linear relationships
relationship
r-values
r-values closer
closer to
to zero
zero indicate
indicate aa weaker
weaker relationship.
relationship.
NOTE:
NOTE: Once
Once rrxyxy is
is calculated,
calculated, we
we need
need to
to see
see whether
whether itit is
is
statistically
statistically significant
significant (if
(if using
using sample
sample data).
data).
Null
Null Hypothesis
Hypothesis when
when using
using r?
r?
H
H00:: rr =
=0
0
There
There is
is no
no relationship
between the
the two
two varia
varia
16
Correlation Coefficient (Pearson

Correlation) rxy

A golf enthusiast is interested in investigating
the relationship, if any, between golfers
driving distance (x) and their 18-hole score (y).
He uses the following sample data (i.e., data
from n = 6 golfers) to examine the issue:
x =Average Driving
Distance (yards.)
277.6
259.5
269.1
267.0
255.6
272.9
y =Average
18-Hole Score
69
71
70
70
71
69

Correlation) rxy
x
277.6
259.5
269.1
267.0
255.6
272.9
y (xi x) (yi y)
69 10.65
71 -7.45
70
2.15
70
0.05
71 -11.35
69
5.95
Average 267.0 70.0

Std. Dev. 8.2192.8944
-1.0
1.0
0
0
1.0
-1.0
(xi x)(yi y)
-10.65
-7.45
0
0
-11.35
-5.95
Total -35.40
(s
xn)1
(y)6
3
5
.4
0
17
.0
8
ii
xy

Correlation) rxy
We had calculated sample Covariance sxy

to be:

Correlation)
sxy rxy
7.08
rxy
-.9631
sxsy (8.2192)(.8944)
Conclusion?
Not only is the relationship negative, but also extremely
strong!
Correlation Coefficient (Pearson Correlation):

r
( x x)( y y )
( x x)
( y y)
s xy
( s x ).( s y )
To understand the practical meaning of r, we can square it.

What would r2 mean/represent?
e.g., r = 0.96
r2 = 92%
r2 Represents the proportion (%) of the total/combined variation

in both x and y that is accounted for by the joint variation
(covariation) of x and y together (x with y and y with x)
r2 always represents a %
Why do we show more interest in r, rather than r2?
20
Correlation Coefficient: Computation

r2 = (Covariation of X and Y together) / (All of variation of X & Y combined)
r
Blood
Age Pressure
X
4
6
9
.
.
_
X=7
( x x)( y y )
( x x)
Y
XX
12
-3
19
-1
14
2
.
.
.
.
_
Y=16
YY
-4
3
-2
.
.
( y y)
_
(X X) (Y Y)
12
-3
-4
.
.
_
_
(X X) (Y Y)
s xy
( s x ).( s y )
_
(X X)2
9
1
4
.
.
_
(X X)2
_
(Y Y)2
16
9
4
.
.
_
(Y Y)2
NOTE: Once r is calculated, we need to see if it is statistically

significant (if sample data). That is, we need to test H0: r = 021
Correlation Coefficient?
Suppose the correlation between X (say, Students GMAT Scores)

and Y (their 1st year GPA in MBA program) is r = +0.48 and
is statistically significant. How would we interpret this?
a) GMAT score and 1st year GPA are positively related so that as
values of one variable increase, values of the other also tend to
increase, and
b) R2 = (0.48)2 = 23% of variations/differences in students GPAs
are explained by (or can be attributed to) variations/
differences in their GMAT scores.
Lets now practice on SPSS
Menu Bar: Analyze, Correlate, Bivariate, Pearson
EXAMPLE: Using data in SPSS File Salary.sav
we wish to see if beginning salary is related to seniority,
age, work experience, and education
22

COMMON TYPES OF ANALYSIS:
Between one variable and another:
Y = a + b 1 x1

Y = a + b 1 x 1 + b2 x 2 + b3 x 3 + + b k x k
Compare Groups
Between Proportions (e.g., Chi Square Test 2)
H 0:
P1 = P2 = P3 = = Pk
Between Means (e.g., Analysis of Variance)

H 0:
1 = 2 = 3 = = k
23

Chi-Square Test of Independence?
Developed by Karl Pearson in 1900.
Is used to compare two or more groups regarding a categorical

characteristic.
That is, to compare proportions/percentages:
Examines whether proportions of different groups of subjects
(e.g., managers vs professionals vs operatives) are equal/
different across two or more categories (e.g., males vs females).
Examines whether or not a relationship exists between
two categorical/nominal variables
(e.g., employee status and gender)
A categorical DV and a categorical IV.
EXAMPLE?
Is smoking a function of gender? That is, is there a difference

between the percentages of males and females who smoke? 24
Chi-Square Test of Independence

Research Sample (n=100):
ID
Gender Smoking Status
1
0 = Male
1 = Smoker
2
1 = Female
0 = Non-Smoker
3
1
1
4
1
0
5
0
0
.
.
.
.
.
.
.
.
.
100
1
0
Dependent variable (smoking status) and the independent variable (gender) are
both categorical.
Null Hypothesis?
H0: There is no difference in the percentages of males and females
who smoke/dont smoke (i.e., Smoking is not a function of gender).
QUESTION: Logically, what would be the first thing you would do?
25

H0: There is no difference in the percentages of males and females who
smoke (Smoking is not a function of gender).
H1: The two groups are different with respect to the proportions who smoke.
TESTING PROCEDURE AND THE INTUITIVE LOGIC:
Construct a contingency Table: Cross-tabulate the

observations and compute Observed (actual) Frequencies
(Oij ):
Male
Smoker
Nonsmoker
TOTAL
Female TOTAL
O11 = 15
O12 = 25
O21 = 5
O22 = 55
20
80
40
60
n = 100
26

Next, ask yourself: What numbers would you expect to find in the
table if you were certain that there was absolutely no difference between
the percentages of males and females who smoked (i.e., if you expected
the Null to be true)? That is, compute the Expected Frequencies (Eij ).
Hint:
What % of all the subjects are smokers/non-smokers?
Male
Smoker
Female
O11 = 15
TOTAL
O12 = 25
40
Nonsmoker
O21 = 5
O22 = 55
60
TOTAL
20
80
n = 100
27

If there were absolutely no differences between the two groups
with regard to smoking, you would expect 40% of individuals
in each group to be smokers (and 60% non-smokers).
Compute and place the Expected Frequencies (Eij ) in the
appropriate cells:
Smoker
Nonsmoker
TOTAL
Male
O11 = 15
E11 = 8
Female
O12 = 25
E12 = 32
TOTAL
40
O21 = 5
E21 = 12
O22 = 55
E22 = 48
60
20
80
NOW WHAT? What is the next logical step?
n = 100
28

Compare the Observed and Expected frequenciesi.e.,
examine the (Oij Eij) discrepancies.
Smoker
Nonsmoker
TOTAL
Male
O11 = 15
Female
O12 = 25
E11 = 8
E12 = 32
O21 = 5
E21 = 12
O22 = 55
E22 = 48
20
80
TOTAL
40
60
n = 100
QUESTION: What can we infer if the observed/actual frequencies

happen to be reasonably close (or identical) to the expected
frequencies?
29

So, the key to answering our original question lies in the size of the
discrepancies between observed and expected frequencies.
If the observed frequencies were reasonably close to the expected

frequencies:
Reasonably certain that no difference exists between percentages of males
and females who smoke,
Good chance that H0 is true
That is, we would be running a large risk of being wrong if we

decide to reject it.
On the other hand, the farther apart the observed frequencies

happen to be from their corresponding expected frequencies:
The greater the chance that percentages of males and females who smoke
would be different,
Good chance that H0 is false and should be rejected
That is, we would run a relatively small risk of being wrong if

we decide to reject it.
30
What is, then, the next logical step?

Compute an Overall Discrepancy Index:
Index One way to quantify the total
discrepancy between observed (Oij) and expected (Eij) frequencies is to
add up all cell discrepancies--i.e., compute (Oij Eij).
Problem?
Positive and negative values of (Oij Eij) RESIDUALS for different cells
will cancel out.
Solution?
Square each (Oij Eij) and then sum them up--compute (Oij Eij)2.
Any Other Problems?

Value of (Oij Eij)2 is impacted by sample size (n).
For example, if you double the number of subjects in each cell, even though cell
discrepancies remain proportionally the same, the above discrepancy index will
be much larger and may lead to a different conclusion.
Solution?
31
Divide each (Oij Eij)2 value by its corresponding Eij value before
summing them up across all cells
That is, compute an index for average discrepancy per subject.
(Oij Eij)2
Eij
You have just developed the formula for 2 Statistic:
(Oij Eij)2
Eij
2 can be intuitively viewed as:

An index that shows how much the observed frequencies are in agreement
with (or apart from) the expected frequencies (for when the null is assumed
to be true).
So, lets compute statistic for our example:

2
32
Smoker
Nonsmoker
Male
O11 = 15
E11 = 8
Female
O12 = 25
E12 = 32
TOTAL
40
O21 = 5
E21 = 12
O22 = 55
E22 = 48
60
20
80
TOTAL
2 =
(15 8)2
8
(25 32)2
32
(5 12)2
12
n = 100
(55 48)2
48
= 12.76
33

Lets Review:
Obtaining a small 2 value means?
Observed frequencies are in close agreement with what we would expect
them to be if there were no differences between our comparison groups.
That is, there is a strong likelihood that no difference exists between the
percentages of males and females who smoke.
Hence, we would be running a significant risk of being wrong if we were
to reject the null hypothesis. That is, is expected to be relatively large.
Therefore, we should NOT reject the null.
NOTE: Smaller2 values result in larger levels (if n remains the same).
A large 2 value means?

34

A large 2 value means:
Observed frequencies are far apart from what they ought
to be if the null hypothesis were true.
That is, there is a strong likelihood for existence of a difference
in the percentages of male and female smokers.
Hence, we would be running a small risk of being wrong if we
were to reject the null hypothesis. That is, is likely to be small.
Thus, we should reject the null.
NOTE: larger2 values result in smaller levels (if n remains the same).
But, how large is large?

For example, does 2 = 12.76 represent a large enough departure (of
observed frequencies) from expected frequencies to warrant
rejecting the null? Check out the associated level!
35
reflects whether is large enough to warrant rejecting the null.

2

Answer:
Consult the table of probability distribution for 2 statistic to see
what the actual value of is (i.e., what is the probability that our
2 value is not large enough to be considered significant).
That is, look up the level associated with your 2 value (under
appropriate degrees of freedom).
Degrees of Freedom:
df = (r-1) (c-1)
df = (2 1) (2 1) = 1
where r and c are # of rows and columns of the contingency
table.
36
37

From the table, the level for 2 = 10.83 (with df = 1) is 0.001 .
Our 2 = 12.76 > 10.83
QUESTION: for our 2 = 12.76 will be smaller or greater than 0.001?
Smaller than 0.001
Therefore, If we reject the null, the odds of being wrong will be even
smaller than 1 in 1000.
Can we afford to reject the null? Is it safe to do so?

CONCLUSION?
% of males and females who smoke are not equal.
That is, smoking is a function of gender.
Can we be more specific?
Percentage of males who smoke is significantly larger than that of the
females (75% vs. 31%, respectively)
38
CAUTION: Select the appropriate percentages to report (Row% vs. Column%)

Column%
Smoker
Male
O11 = 15
Female
O12 = 25
TOTAL
40
Nonsmoker
O21 = 5
O22 = 55
60
20
80
TOTAL
15 / 20 = 75%
n = 100
25 / 80 = 31%
Phi (a non-parametric correlation for categorical data):

=
2 / N
12.76 / 100 = 0.357 (Note: sign is NA)
39

VIOLATION OF ASSUMTIONS:
2 test requires expected frequencies (Eij) to be reasonably large. If

this requirement is violated, the test may not be applicable.
SOLUTION:
For 2 x 2 contingency tables (df = 1), use the Fishers Exact

Probability Test results (automatically reported by SPSS).
That is, look up of the Fishers exact test to arrive at your conclusion.
For larger tables (df > 1), eliminate small cells by combining
their corresponding categories in a meaningful way.
That is, recode the variable that is causing small cells into a
new variable with fewer categories and then use this new
variable to redo the Chi-Square test.
40

Lets now use SPSS to do the same analysis!
Menu Bar: Analyze, Descriptive
Statistics, Crosstabs
Statistics: Chi-Square, Contingency
Coefficient.
Cells: Observed, Row/Column
percentages (for the independent
variable)
SPSS File: smoker

SPSS File: GSS93 Subset
41

Suppose we wish to examine the validity of the gender
gap hypothesis for the 1992 presidential election between
Bill Clinton, George Bush, and Ross Perot.
SPSS File: Voter
42
Correlation Coefficient (Pearson Correlation) = r

What do you use it for?
Karl Pearson, a Galton

Student & the Founder
of Modern Statistics
To examine whether a relationship exists between

two metric variables (e.g., income and education, or
workload and job satisfaction) and what the nature
and strength of that relationship may be.
Range of Values for r?
-1 < r < +1
Null Hypothesis when using r?
r = 0 (There is no relationship between the two variables.)
43
44
Correlation Coefficient:
To understand the practical meaning of r, we can square it.
What would r2 mean/represent?
r2 Represents the proportion (%) of the total/combined variation in

both x and y that is accounted for by the joint variation
(covariation) of x and y together (x with y and y with x)
How is it calculated?
r2 = (Covariation of X and Y together) / (Total variation of X & Y combined)
How do we measure/quantify variations?
r2
( x x)( y y ) / n 1
[ ( x x) 2 / n 1][ ( y y ) 2 / n 1]
( x x)( y y)
( x x) ( y y )
2
r2 always represents a %
2
Why do we show more interest in

r , rather than r2?
45
Correlation Coefficient: Computation

r2 = (Covariation of X and Y together) / (All of variation of X & Y combined)
r
(x
(x
X
4
6
9
.
.
.
_
X=7
x)( y y )
x) 2
(y
Y
XX
12
-3
19
-1
14
2
.
.
.
.
.
.
_
Y=16
YY
-4
3
-2
.
.
.
y) 2
(X X) (Y Y)
12
-3
-4
.
.
.
_
_
(X X) (Y Y)
_
(X X)2
9
1
4
.
.
.
_
(X X)2
_
(Y Y)2
16
9
4
.
.
.
_
(Y Y)2
NOTE: Once r is calculated, we need to see if it is statistically

significant (if sample data). That is, we need to test H0: r = 046
Correlation Coefficient?
Suppose the correlation between X (say, Students GMAT Scores)

and Y (their 1st year GPA in MBA program) is r = +0.48 and is
statistically significant. How would we interpret this?
a) GMAT score and 1st year GPA are positively related

so that as values of one variable increase, values of
the other also tend to increase, and
b) 23% of variations/differences in students GPAs are
explained by (or can be attributed to) variations/
differences in their GMAT scores.
Lets now practice on SPSS
Menu Bar: Analyze, Correlate, Bivariate, Pearson
Using data in SPSS File Salary.sav we wish
to see if beginning salary is related to seniority,
age, work experience, and education
47

COMMON TYPES OF ANALYSIS:
Between one variable and another: Y = a + b 1 x1

Y = a + b 1 x 1 + b2 x 2 + b 3 x 3 + + b k x k
Compare Groups
Proportions (e.g., Chi Square Test 2)
Means (e.g., Analysis of Variance)
48

Chi-Square Test of Independence?
To examine whether proportions of different groups
of subjects (e.g., managers vs operatives) are
equal/different across two or more categories (e.g.,
males vs females).
To examine whether or not a relationship exists
between two categorical/nominal variables (e.g.,
employee status and gender)--categorical dependent
variable, categorical independent variable.
EXAMPLE?
Is smoking a function of gender? That is, is there a

difference between the percentages of males and
females who smoke?
49

Research Sample:
ID
Gender Smoking Status
1
0 = Male
1 = Smoker
2
1 = Female
0 = Non-Smoker
3
1
1
4
1
0
5
0
0
.
.
.
.
.
.
.
.
.
100
1
0
dependent variable (smoking status) and the independent variable (gender) are
both categorical.
Null Hypothesis?
H0: There is no difference in the percentages of males and females
who smoke (Smoking is not a function of gender).
50
QUESTION: Logically, what would be the first thing you would do?

H0: There is no difference in the percentages of males and females who
smoke (Smoking is not a function of gender).
H1: The two groups are different with respect to the proportions who smoke.
TESTING PROCEDURE AND THE INTUITIVE LOGIC:

Construct a contingency Table: Cross-tabulate the
observations and compute Observed (actual) Frequencies (Oij ) :
Smoker
Nonsmoker
TOTAL
Male
O11 = 15
O21 = 5
Female
O12 = 25
O22 = 55
TOTAL
40
60
20
80
n = 100
51

Next, ask yourself: What numbers would you expect to find in
the table if you were certain that there was absolutely no
difference between the percentages of males and females who
smoked? That is, compute the Expected Frequencies (Eij ).
Hint: What % of all the subjects are smokers/non-smokers?

Male
Smoker
Female
O11 = 15
TOTAL
O12 = 25
40
Nonsmoker
O21 = 5
O22 = 55
60
TOTAL
20
80
n = 100
52

If there were absolutely no differences between the two groups
with regard to smoking, you would expect 40% of individuals
in each group to be smokers (and 60% non-smokers).
Compute and place the Expected Frequencies (Eij ) in the
appropriate cells:
Smoker
Nonsmoker
TOTAL
Male
O11 = 15
E11 = 8
Female
O12 = 25
E12 = 32
TOTAL
40
O21 = 5
E21 = 12
O22 = 55
E22 = 48
60
20
80
NOW WHAT? What is the next logical step?
n = 100
53

Compare the Observed and Expected frequenciesi.e.,
examine the (Oij Eij) discrepancies.
Smoker
Nonsmoker
TOTAL
Male
O11 = 15
Female
O12 = 25
E11 = 8
E12 = 32
O21 = 5
E21 = 12
O22 = 55
E22 = 48
20
80
TOTAL
40
60
n = 100
QUESTION: What can we infer if the observed frequencies happen

to be reasonably close (or identical) to the expected frequencies?
54

If the observed frequencies were reasonably close to the expected
frequencies:
Reasonably certain that no difference exists between percentages of
males and females who smoke,
Good chance that H0 is true
That is, we would be running a large risk of being wrong if

we decide to reject it.
On the other hand, the farther apart the observed frequencies happen to be
from their corresponding expected frequencies:
The greater the chance that percentages of males and females who smoke
would be different,
Good chance that H0 is false and should be rejected
That is, we would run a relatively small risk of being wrong

if we decide to reject it.
So, the key to answering our original question lies in the size of the
discrepancies between observed and expected frequencies.
What is, then, the next logical step?
55

Compute an Overall Discrepancy Index: To quantify the overall
discrepancy between observed (Oij). and expected (Eij). frequencies,
we can add up all our cell discrepancies--i.e., compute (Oij Eij).
Problem?
Positive and negative values of (Oij Eij) RESIDUALS for different cells will
cancel out.
Solution?
Square each (Oij Eij) and then sum them up--compute (Oij Eij)2.
Any Other Problems?

Value of (Oij Eij)2 is impacted by sample size (n).
For example, if you double the number of subjects in each cell, even though cell
discrepancies remain proportionally the same, the above discrepancy index will
be much larger and may lead to a different conclusion.
Solution?
56
Divide each (Oij Eij)2 value by its corresponding Eij value before
summing them up across all cells
That is, compute the total discrepancy per subject index.
(Oij Eij)2
Eij
You have just developed the formula for 2 Statistic:
(Oij Eij)2
Eij
2 can be intuitively viewed as an index that shows how much the

observed frequencies are in agreement with (or apart from) the
expected frequencies (when the null is assumed to be true).
So, lets compute 2 statistic for our example:
57
Smoker
Nonsmoker
Male
O11 = 15
E11 = 8
Female
O12 = 25
E12 = 32
TOTAL
40
O21 = 5
E21 = 12
O22 = 55
E22 = 48
60
20
80
TOTAL
2 =
(15 8)2
8
(25 32)2
32
(5 12)2
12
n = 100
(55 48)2
48
= 12.76
58

Lets Review:
Obtaining a small 2 value means?
Observed frequencies are in close agreement with what we would expect
them to be if there were no differences between our comparison groups.
That is, there is a strong likelihood that no difference exists between the
percentages of males and females who smoke.
Hence, we would be running a significant risk of being wrong if we were to
reject the null hypothesis. That is, is expected to be relatively large.
Therefore, we should NOT reject the null.
A large 2 value means?
59

A large 2 value means:
Observed frequencies are far apart from what they ought
to be if the null hypothesis were true
That is, there is a strong likelihood for existence of a difference
in the percentages of male and female smokers.
Hence, we would be running a small risk of being wrong if we
were to reject the null hypothesis. That is, is likely to be
small.
Thus, we should reject the null.
But, how large is large?

For example, does 2 = 12.76 represent a large enough departure (of
observed frequencies) from expected frequencies to warrant
rejecting the null?
60

Answer:
Consult the table of probability distribution for 2 statistic to
see what the actual value of is (i.e., what is the probability
that it is not large enough to be considered significant).
That is, look up the level associated with your 2 value
(under appropriate degrees of freedom).
Degrees of Freedom:
df = (r-1) (c-1)
df = (2 1) (2 1) = 1
where r and c are # of rows and columns of the contingency
table.
61
62

From the table, the level for 2 = 10.83 (with df = 1) is 0.001 .
Our 2 = 12.76 > 10.83
QUESTION: If we decide to reject the Null, will be smaller or greater than
0.001?
Smaller than 0.001
Therefore, If we reject the null, the odds of being wrong will be even
smaller than 1 in 1000.
Can we afford to reject the null? Is it safe to do so?

CONCLUSION?
% of males and females who smoke are not equal.
That is, smoking is a function of gender.
Can we be more specific?
Percentage of males who smoke is significantly larger than that of the
females (75% vs. 31%, respectively)
63
CAUTION: Select the appropriate percentages to report (Row% vs. Column%)
Smoker
Male
O11 = 15
Female
O12 = 25
TOTAL
40
Nonsmoker
O21 = 5
O22 = 55
60
20
80
TOTAL
15 / 20 = 75%
n = 100
25 / 80 = %31
Phi (a non-parametric correlation for categorical data):

=
2 / N
64

VIOLATION OF ASSUMTIONS:
2 test requires expected frequencies (Eij) to be reasonably large.

If this requirement is violated, the test may not be applicable.
SOLUTION:
For 2 x 2 contingency tables (df = 1), use the Fishers Exact

Probability Test results (automatically reported by SPSS).
That is, look up of the Fishers exact test
For larger tables (df > 1), eliminate small cells by combining
their corresponding categories in a meaningful way.
65

Lets now use SPSS to do the same analysis!
Menu Bar: Analyze, Descriptive
Statistics, Crosstabs
Statistics: Chi-Square, Contingency
Coefficient.
Cells: Observed, Row/Column
percentages (for the independent
variable)
SPSS File: smoker

SPSS File: GSS93 Subset
66

Suppose we wish to examine the validity of the gender
gap hypothesis for the 1992-93 presidential elections
between Bill Clinton, George Bush, and Ross Perot.
SPSS File: Voter
67
Assignment #3
68
Assignment #3
69
Assignment #3
NOTE:
If you examine the value labels for the variable daysofwk, you will see that it is coded
as 1=Sunday, 2=Monday, 3=Tuesday, 4=Wednesday, 5=Thursday, 6=Friday, and
7=Saturday. Therefore, for part (b), you will need to create a new variable--i.e., Recode
daysofwk into a new dichotomous variable (say, deathday), that would represent death
during Mondays, Tuesdays, and Wednesdays vs. Fridays, Saturdays and Sundays.
Notice that the subjects who died on Thursdays should not be included in the analysis
(i.e., should not be represented in any of the two categories of days represented by the
new variable) Also, make sure you properly define the attributes (e.g., label, value
label, etc.) of this new variable (i.e., deathday).
REMINDERS:
For each analysis, include the Notes in the printout. Also, edit the first page of your
first analysis output to include your name. Make sure that on your printout you
explain your findings and conclusions. Be specific as to what parts of the output you
have used, and how you have used them, to reach your conclusions.
Make sure that you tell the whole story and that your explanations of the findings are
complete. For example, it is not enough to say that there is a significant relationship
between characteristic A and characteristic B. You have to go on to indicate how the
two characteristics are related and what that relationship really means.
70
HYPOTHESIS TESTING
QUESTIONS OR
COMMENTS
?
71

Correlation and Chi-Square Test - LDR 280

Uploaded by

Copyright:

Available Formats

Correlation and Chi-Square Test - LDR 280

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Correlation and Chi-Square Test - LDR 280

Uploaded by

Copyright:

Available Formats

A Review of

If we do not reject the null, conclusion?

If we do decide to reject the null, conclusion?

How do we decide to reject/not reject the null?

STATITICAL DATA ANALYSIS

Multivariate (e.g., Multiple Regression Analysis)

Between one dep. var. and an independent variable, while

Between Means (e.g., Analysis of Variance)

Lets first review some fundamentals.

Remember: Level of measurement determines choice of

* Fishers Exact Prob.

METRIC (ORDERED METRIC or

* Correlation (and Covariance) Analysis

Correlation and Covariance:

Covariance (sxy ) of Two Variables

Covariance (sxy ) of two variables

Average 267.0 70.0

Example: Golfing Study

Hard to tell; there is no standard metric to judge it by!

Correlation Coefficient (r) is a standard measure/metric for judging

sxy = Covariance of x & y

In1888, plotted lengths of forearms and head sizes to see to what

Correlation Coefficient (Pearson Correlation) = r

Karl Pearson, a Galton

Correlation Coefficient (Pearson Correlation) rxy

Correlation Coefficient (Pearson

Example: Golfing Study

Correlation Coefficient (Pearson

Average 267.0 70.0

Correlation Coefficient (Pearson

We had calculated sample Covariance sxy

Correlation Coefficient (Pearson

Correlation Coefficient (Pearson Correlation):

To understand the practical meaning of r, we can square it.

r2 Represents the proportion (%) of the total/combined variation

Correlation Coefficient: Computation

NOTE: Once r is calculated, we need to see if it is statistically

Suppose the correlation between X (say, Students GMAT Scores)

STATITICAL DATA ANALYSIS

Multivariate (e.g., Multiple Regression Analysis)

Between Means (e.g., Analysis of Variance)

STATITICAL DATA ANALYSIS

Developed by Karl Pearson in 1900.

Is used to compare two or more groups regarding a categorical

Is smoking a function of gender? That is, is there a difference

Chi-Square Test of Independence

Chi-Square Test of Independence

TESTING PROCEDURE AND THE INTUITIVE LOGIC:

Construct a contingency Table: Cross-tabulate the

Chi-Square Test of Independence

Chi-Square Test of Independence

NOW WHAT? What is the next logical step?

Chi-Square Test of Independence

QUESTION: What can we infer if the observed/actual frequencies

Chi-Square Test of Independence

If the observed frequencies were reasonably close to the expected

That is, we would be running a large risk of being wrong if we

On the other hand, the farther apart the observed frequencies

That is, we would run a relatively small risk of being wrong if