Testing a Population Proportion by Using the Chi-Square Goodness-of-Fit Test

as an Alternative Technique to the z Test

 Testing a Population Proportion by Using the Chi-Square Goodness-of-Fit Test
as an Alternative Technique to the z Test

We have discussed a technique for testing the value of a population proportion.

When sample size is large enough (n p >= 5 and n q>= 5), sample proportions are normally distributed
and the following formula can be used to test hypotheses about p.
A manufacturer believes exactly 8% of its products contain at least one minor flaw.
Suppose a company researcher wants to test this belief. The null and alternative hypotheses are
H0: p = .08
Ha: p = .08

3
The observed value of z is in the rejection region (observed z = 4.43> table z.05= +1.645), so the
business researcher rejects the null hypothesis.
The chi-square goodness-of-fit test can also be used to conduct tests about p; this situation can
be viewed as a special case of the chi-square goodness-of-fit test where the number of
classifications equals two (binomial distribution situation).
The observed chi-square is computed in the same way as in any other chi-square goodness-of-fit
test, but because the test contains only two classifications (success or failure), k = 2 and the
degrees of freedom are k - 1 = 2 - 1 = 1.
Working this problem by the chi-square goodness-of-fit test, we view it as a two category
expected distribution in which we expect .08 defects and .92 non-defects.

The observed categories are 33 defects and 200 - 33 = 167 non-defects. Using the total observed
items (200), we can determine an expected distribution as .08(200) = 16 and .92(200) = 184.
Shown here are the observed and expected frequencies.

5
6
 2 Test for the Difference Between Two Proportions

H0: π1 = π2 (Proportion of females who are left handed is equal to the proportion of
males who are left handed)
H1: π1 ≠ π2 (The two proportions are not the same – hand preference is not
independent of gender)

• If H0 is true, then the proportion of left-handed females should be the same as the proportion of left-handed
males
• The two proportions above should be the same as the proportion of left-handed people overall
 The Chi-Square Test Statistic
The Chi-square test statistic is:
( fo  fe )2
2
χ STAT  
all cells
fe

• where:
fo = observed frequency in a particular cell
fe = expected frequency in a particular cell if H0 is true

 STAT
2
for the 2 x 2 case has 1 degree of freedom

(Assumed: each cell in the contingency table has expected frequency of at least 5)
 2
Decision Rule
The χ STATtest statistic approximately follows a
chi-squared distribution with one degree of freedom

Decision Rule:
If χ
2
STAT
 χ α2 , reject H0,
otherwise, do not reject H0


0
Do not Reject H0 2
reject H0  2α
 The Chi-Square Test Statistic
Gender Hand Preference
Left Right
Female Observed = 12 Observed = 108 120

Male Observed = 24 Observed = 156 180

36 264 300
 The Chi-Square Test Statistic
Gender Hand Preference
Left Right
Female Observed = 12 Observed = 108 120

Expected = 14.4 Expected = 105.6

Male Observed = 24 Observed = 156 180

Expected = 21.6 Expected = 158.4

36 264 300

The test statistic is: fe = (RT*CT)/n

(f o  f e ) 2
χ 2STAT   f
all cells e

(12  14.4)2 (108  105.6)2 (24  21.6)2 (156  158.4)2

     0.7576
14.4 105.6 21.6 158.4
 Decision Rule
DOF: (No of Roes -1) (No of columns -1)

2
The test statistic is χ STAT  0.7576 ; χ 02.05 with 1 d.f.  3.841
Decision Rule:
2
If χ STAT > 3.841, reject H0, otherwise, do
not reject H0

0.05 Here,
2 2
χ STAT = 0.7576< χ 0.05 = 3.841,
so we do not reject H0 and
0
Do not reject H0 Reject H0 2 conclude that there is not sufficient
evidence that the two proportions
20.05 = 3.841 are different at  = 0.05
 2 Test for Differences Among More Than Two Proportions
• Extend the 2 test to the case with more than two
independent populations:

H0: π1 = π2 = … = πc
H1: Not all of the πj are equal (j = 1, 2, …, c)
 The Chi-Square Test Statistic
The Chi-square test statistic is:
( fo  fe )2
2
χ STAT  
all cells
fe
• Where:
fo = observed frequency in a particular cell of the 2 x c table
fe = expected frequency in a particular cell if H0 is true

χ 2STAT for the r x c case has (r - 1)(c - 1)  c - 1 degrees of freedom

(Assumed: each cell in the contingency table has expected frequency of at least 1)
 Computing the
Overall Proportion
The overall X1  X 2    Xc X
p 
proportion is: n1  n2    nc n

• Expected cell frequencies for the c

categories are calculated as in the 2 x 2
case, and the decision rule is the same:
2
Where χα is from the chi-squared
Decision Rule: distribution with c – 1 degrees of freedom
2
If χ STAT  χ α2 , reject H0,
otherwise, do not reject H0
Chap 11-16
In an area three are three hotels, the table shows response to a question, whether they will
select it again

Choice of hotel Golden Palm Royale Palm princes Total

palm
Yes 128 199 186 513
No 88 33 66 187
216 232 252 700

X1  X 2    Xc X
p  (128+199+186) / (216+232+252) = 513/ 700 = 0.733
n1  n2    nc n

17
 Expected values
Choice of hotel Golden palm Palm Royale Palm princes Total

Yes .733*216 =158.30 .733*232=170.02 .733*252=184.68 513

No .267*216=57.70 .267*232=61.98 .267*252=67.32 187
216 232 252 700

H0: π1 = π2 = … = π3
H1: Not all of the πj are equal (j = 1, 2, …, 3)

18