This preview shows pages 1–2. Sign up to view the full content.
CHAPTER 23 SOLUTIONS AND MINIPROJECT NOTES
CHAPTER 23
HYPOTHESIS TESTING—EXAMPLES AND CASE STUDIES
EXERCISE SOLUTIONS
23.1
a.
Null
: It is not the case that a majority of American Catholics favor allowing women to be priests.
Alternative
: A majority of American Catholics favor allowing women to be priests.
b.
Standard deviation is the square root of (0.5)(0.5)/507 = 0.022; standardized score is (0.59 
0.50)/0.022 = 4.05.
c.
Reject the null hypothesis.
We can conclude that a majority of American Catholics favor allowing
women to be priests.
23.2
Onesided, because the alternative includes values above 0.50 only.
23.3
Double the value to get 0.06.
23.4
a.
No, because 0.08 > 0.05.
b.
Yes, the
pvalue
would be 0.04.
23.5
Use
z
score = (sample mean  hypothesized mean)/SEM; find SEM using the standard deviation divided
by the square root of the sample size.
23.6
a.
2
×
0.015 = .03
b.
0.023
c.
2.16
×
10
5
= 0.0000216
23.7
a.
Null
: The training program has no effect on test scores.
Alternative
: Test scores will be higher after
the training program.
b.
Standardized score = (10  0)/4 = 2.5.
c.
From Table 8.1, the
pvalue
is between 0.01 and 0.005, so reject the null hypothesis and conclude that
scores would be higher after the program.
d.
The students had already taken the test once, before the program, so when they took it afterward they
were familiar with the test already.
e.
Randomly assign some students to take the program and others to not take it, using matched pairs if
possible.
Compare scores after the program.
23.8
A type 1 error would be that there is no relationship but that the study found one, in which case the law
discriminating by ages would have been upheld.
A type 2 error would be that there is a relationship but it
was not detected, in which case the law was not upheld, but perhaps should have been because young
males are more likely to drink and drive than young females.
23.9
The null hypothesis was that mental scores are not changed by taking estrogen, while the alternative is that
they are.
(It is not clear if the alternative was onesided, saying scores increase.)
The study found that the
null hypothesis could be rejected, so the
pvalue
must have been 0.05 or less.
23.10
a.
Null
: In the population of elderly people, there is no relationship between pet ownership and number
of doctor contacts.
Alternative
: In the population of elderly people, those with pets have fewer doctor
contacts than those without pets.
b.
Because we are told that the
pvalue
is less than 0.05, we can reject the null hypothesis.
We can
conclude that elderly people with pets have significantly fewer doctor contacts than those without pets.
23.11
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
This is the end of the preview. Sign up
to
access the rest of the document.
 '08
 SEIFRIEDTHOMASJ

Click to edit the document details