These are writing assignments for a graduate-level applied research methods statistics 1 class. Must be knowledgable in the system software stata. Please find attached.
Writing Assignment 4:
Bivariate Analysis of Public Opinion
Due date: Friday, February 12, 6:00 a.m.
Purpose: This assignment applies your understanding of contingency tables and bar charts to perform bivariate analyses. It reinforces concepts and coding skills that you will use repeatedly in the remainder of this class and in 8131. It also prepares you for the first paper assignment.
Knowledge: This exercise
Reinforces your understanding of contingency tables
Teaches you a little about public opinion
Skills: This exercise also
Uses Stata to create contingency tables and bar charts
Develops your ability to write about statistics
Task: It is the second day of your internship at the campaign of a Congressional candidate. Your supervisor wants a new analysis of public opinion on government spending. She asks you to look at the 2018 GSS again, but this time she wants you to focus on one of the government spending variables. She wants you to use contingency tables and bar charts to examine how three independent variables (personal characteristics) affect attitudes toward whether we should be spending more or less on that item.
Choose a government spending item (almost any of the variables that start with nat). You can use “describe nat*” or “tab1 nat*” to see what your options are.
Go to GSS Data Navigator (https://gssdataexplorer.norc.org/variables/vfilter), type in the variable name, and click on the variable name on the next screen to see the exact question wording. Include at least part of that wording, rather than just the variable label, in your paper.
For your bar chart(s), you will need to recode your dependent variable into a dummy variable.
Code it 0 for No and 100 for Yes. When you take the mean of that variable in the bar chart, it will represent the percentage who say “Yes.”
If you want to focus on what characteristics make a person more likely to want to increase spending on this item, code people who say “too little” as 100 and those who say “about the right amount” or “too much” as 0.
If you want to focus on what characteristics make a person more likely to want to cut spending on this item, code people who say “too much” as 100 and those who say “about the right amount” or “too little” as 0.
Do not make “about the right amount” a missing value. We want to know what percentage of the respondents want to increase or decrease spending, not what percentage of respondents who want changes in spending favor cuts or increases.
Choose three independent variables that you think might affect attitudes toward spending on this item. Obvious choices include sex, race, age, education (educ), religion (use relig), religious attendance (attend), party identification (partyid), and liberalism-conservatism (polviews).
Use codebook varname or tab varname, nolabel to make sure that you understand how each variable is coded; you may want to recode it.
One way to decide whether to recode an ordinal- or interval-level variable with many values (e.g., attend, polviews, party7, educ, age) is to run a crosstab using the original version of the independent variable. If you think you can make the pattern clearer without distorting the findings by using fewer values, recode your variable.
Create three contingency tables.
If your independent variable is the column variable, have Stata print only the column percentages. (Use the nofreq option.)
If your independent variable is the row variable, have Stata print only the row percentages. (Use the nofreq option.)
Create at least one bar chart showing the relationship between one of your independent variables and the spending item.
Your independent variable goes inside the parentheses after over and your recoded 0-100 dummy dependent variable goes after graph bar or graph hbar.
Write approximately one page (double-spaced, 12-point font, in Word) describing your findings.
First, describe public opinion for the data set as a whole. What percentage of the full sample thinks we should spend more, less, or the same amount?
Second, describe the findings from your three contingency tables.
If your independent variable is the column variable, compare column percentages in the top or bottom row.
If your independent variable is the row variable, compare row percentages in the first or last column.
What do the percentages show about the relationship between the independent and dependent variables?
If your independent variable is a dummy or nominal-level variable, how do the groups differ in the percentage who want to spend more or less?
If your independent variable is ordinal-level, does the percentage saying Yes increase or decrease steadily as the independent variable increases, or is the pattern more complex?
Third, interpret the findings from your bar chart(s).
What do the percentages/bar heights show about the relationship between the independent and dependent variables?
If you use a bar chart and a contingency table for the same independent variable, discuss both at the same time, rather than discussing the bar chart in a separate section.
Fourth, save all your commands in a do-file and copy it into your Word document at the end of the paper as an appendix.
You can insert your tables and bar chart at appropriate places in the text, or you can add them at the end of the paper but before the appendix.
Feel free to create better value labels to make your tables look better.
Criteria: This is a low-stakes exercise, worth 2% of your final grade, so it is primarily an opportunity to practice your understanding of this module. Writing assignments are worth 10 points. To get the full 10 points,
Recode your spending variable into a 0-100 dummy variable for your bar chart. Make clear in your write-up which of the original values got recoded 100 and which got recoded 0.
Create three contingency tables. The tables should only include the appropriate (row or column) percentages and no absolute frequencies.
Correctly interpret the percentages. Show the relationship between the independent and dependent variables by comparing appropriate percentages. Ignore absolute frequencies.
Correctly create at least one bar chart.
Correctly interpret the findings from your bar chart(s).
Provide an appropriate appendix.
I will also provide feedback on your writing.
