Instructions

This assignment asks you to apply what we discussed in Modules 10 and 11 to conduct tests of significance and analyze data from the RCPA3 package in R. Your work will be graded on accuracy opposed to completion and you are not allowed to work with others to complete this assignment. Please read the instruction below closely. Questions 1, 2, and 3 will require you to analyze data in R. Question 4 requires you to provide all of the R code you used for this assignment.

Question 1 (7 points)

Let's compare the proportions of men and women who strongly favor the death penalty for murder. Opinions about the death penalty are recorded by the nes dataset variable "death.penalty" and respondents' gender is recorded by the "gender" variable. When you do this analysis, don't forget to weight observations using the dataset's "wt" variable.

1. What is the proportion of men who strongly favor the death penalty for murder?
2. What is the proportion of women who strongly favor the death penalty for murder?
3. What is the observed difference of proportions between men and women?
4. What is the standard error of the difference of proportions?
5. What is the test statistic (z score) and its corresponding p-value for the difference of proportions?
6. Suppose someone hypothesized that there is no difference between men and women in support of the death penalty for murder. Based on the results of the test you just performed and reported above, would reject or fail to reject this hypothesis? Explain your answer.

Question 2 (7 points)

For this exercise, you'll use a difference of means test to evaluate whether voter turnout is different in Southern states and non-Southern states. For this analysis, think about turnout in 2020 as the sample of voter turnout we might observe in the US states. Do not assume equal variance. Conduct a two-sample difference of means test comparing the mean value of the states dataset's "vep20.turnout" in Southern and non-Southern states. You can use the states dataset's "south" variable as the independent variable in this analysis.

1. If someone were to hypothesize that voter turnout is greater in Southern states compared to non-Southern states, what is the unstated null hypothesis?
2. According to your analysis, what is the mean voter turnout in non-Sothern states?
3. According to your analysis, what is the mean voter turnout in Southern states?
4. What is the observed difference of means and corresponding 95percent confidence interval between non-Southern and Southern states?
5. Is the hypothesis that voter turnout is greater in Southern states compared to non-Southern states supported by this analysis? Explain your answer.

Question 3 (9 points)

In this exercise, you will analyze factors that influence public opinion on abortion policy. Is there a significant gender different with respect to abortion policy? A basic comparison of public opinion on abortion policy may reveal some gender differences, with women tending to be less supportive or abortion restrictions than men. However, it is also reasonable to hypothesize that the relationship between gender (IV) and opinion about abortion (DV) will not be the same for all partisan identifications (control variable). If we control for party id, the gender difference may be spurious, or limited to certain party identifications.

The nes dataset contains the variable "abortion.legal" which will be the dependent variable in this analysis. The independent variable will be "gender" and the control variable will be "partyid3". Use the "wt" variable to weight observations. Use the crosstabC function to compare opinions about abortion policy by gender, controlling for party identification. Include the appropriate arguments to receive chi-square test statistics. Additionally, to make you table easier to read, include the following argument: "dvlabs = c("Never", "Rare Exceptions", "Sometimes", "Always", "Other")"

1. Complete the controlled cross-tabulation table below.
2. What is the chi-square statistics and corresponding p-value for Democrats?
3. What is the chi-square statistics and corresponding p-value for Independents?
4. What is the chi-square statistics and corresponding p-value for Republicans?
5. Based on the chi-square tests, the relationship between gender and abortion.legal variables is statistically significant for which of the following party identifications, Democrats, Independents, Republicans? Explain your answer.

Abortion Stance	Men Dem	Women Dem	Men Ind.	Women Ind.	Men Rep.	Women Rep.	Totals
Never
Rare Exceptions
Sometimes
Always
Other
Totals

Question 4 (2 points)

Provide all of the R code you used to complete this assignment.