Discuss the following:
Q1) Data from the 2002 GSS show that 84.3 percent (198 out of 235) of females and 75 percent (198 out of 264) of males are in favour of requiring gun permits. You wonder whether there is any difference between males and females in the population in their support for gun permits. Use a test of the difference between proportions when answering these questions.
a) What is the research hypothesis? Should you conduct a one- or a two-tailed test? Why?
b) Test your hypothesis at the .05 level. What do you conclude?
Q2) Now calculate the chi-square for this table to determine whether our previous advice to the neighbourhood clinic was correct.
|
Race and Sex of Respondent
|
|
|
|
|
|
|
White Males
|
Black Males
|
White Females
|
Black Females
|
Total
|
|
Confidence in Medicine
|
A great deal
|
135
|
17
|
136
|
25
|
313
|
|
|
Only some
|
162
|
23
|
214
|
47
|
446
|
|
|
Hardly any
|
33
|
6
|
41
|
11
|
91
|
|
Total
|
|
330
|
46
|
391
|
83
|
850
|
a) How many degrees of freedom does the table have?
b) Calculate chi-square for the table. What is expected number black females who have 'only some' confidence in medical care?
c) Test the hypothesis that confidence in medicine and gender/race are independent. What is the P value of your obtained chi-square? What do you conclude (alpha=.05)?
Q3) a) Calculate chi-square for the below table.
b) Test the null hypothesis that the two variables are independent at the .05 alpha level.
c) Which cell has the greatest difference between the expected value (fe ) and the actual value (f0)? What did you discover when you tried to answer this question?
|
|
|
Church Attendence
|
|
|
|
|
Never
|
Several Times a Year
|
Every Week
|
Total
|
|
Homosexual
|
Always wrong
|
43
|
23
|
53
|
119
|
|
Relations
|
Not wrong at all
|
42
|
22
|
10
|
74
|
|
|
|
|
|
|
|
|
Total
|
|
85
|
45
|
63
|
193
|