9.7. Tests about a Population Proportion
The general setting is the following:
: the population proportion ;
: the particular hypothesized value
The
statistical description for a testing problem is
vs. ![]()
vs.
,
and
vs.
.
General
Procedure: Large Sample Case (
) and
level of significance ![]()
.
(I):
vs.
,
Then,

In addition,
(II):
vs.
.
Then,

In addition,
![]()
(III):
vs. ![]()
Then,

In addition,
![]()
Example 6:
An official
of a large national union claims that the fraction of women in the union
is
not significant different from 0.5,
vs.
.
Using
the sample information reported below,
|
Sample size |
400 |
|
Women |
168 |
|
Men |
232 |
(a) Test the
hypothesis at 5% level of significance using a critical value (classical
approach).
(b)
Test the hypothesis at 10% level of significance using a p-value.
(c)
Using a confidence interval, test the hypothesis at the
5% level of significance.
[solutions:]
.
(a)
,
.
(b)
,
![]()
(c) A 95% confidence interval for
is
![]()
Since
, we reject
.
JavaStatSoft:
Z
test:
Statistics
-> Tests -> One Sample -> Proportion and Count