This assignment is designed to develop your skill in analyzing survey data. You’re asked to analyze data from the General Social Survey.
Collecting Data from the General Social Survey
You will investigate how (1) the frequency of spending time with friends and (2) the frequency of spending time with relatives each influence marital satisfaction, using data from the General Social Survey for the years 2004 to 2014 (combined) on married respondents ages 18 to 44.
Married respondents were asked about their marital happiness, as measured by the statement: Taking all things together, how would you describe your marriage? Would you say that your marriage is very happy, pretty happy, or not too happy?
Responses were 1 = very happy, 2 = pretty happy, 3 = not too happy. We will combine response categories 2 and 3.
Respondents were also asked, How often do you spend a social evening with friends outside your neighborhood? [ variable name: socfrend]
and in another question, How often do you spend a social evening with relatives? [socrel]
Responses to both questions were 1 = almost every day, 2 = once or twice a week, 3 = several times a month, 4 = about once a month, 5 = several times a year, 6 = about once a year, 7 = never.
We will combine response categories 1 and 2, 3, 4 and 5, and 6 and 7. In your paper, you may refer to these four categories as: very often, often, occasionally, and rarely or never, respectively.
Go to the web site for the General Social Survey:
http://sda.berkeley.edu/sdaweb/analysis/?dataset=gss14
Click on the Tables tab. In the Row box, type: hapmar(r: 1; 2-3) In the Column box below that, type: socfrend(r: 1-2; 3; 4-5; 6-7)
To select respondents ages 18 to 44 for the appropriate years:
Under Selection Filter(s), type: year(2004-2014) age(18-44)
Click on the bar that says Output Options. Check Unweighted N, and uncheck Weighted N.
Under Other Options, check Summary Statistics.
Under Chart Options, check Show percents.
Click on Run the Table.
A window with the table will pop up. (You may want to print it for convenience, but you won’t need to turn it in.)
Look at the distribution of responses on marital happiness by how often people socialize with friends. Focus on the “very happy” category. Does it seem to differ? How? Look at the summary statistics. Is the p-value for the Rao-Scott tests less than .05? If so, we’d say the two variables are related.
Remember that the Rao-Scott tests tell us whether two variables are probably related to each other, or strictly speaking, the likelihood that they aren’t. So if the p-value is less than .05, it means that there is less than a 5% chance that the two variables are unrelated.
Research Report
Answer the following questions in a 1–2 page typed, double-spaced paper. Your paper will be most readable if you structure your responses to the questions below as separate paragraphs.
Begin your paper with a general one-paragraph introduction of what you will do in your paper. Be sure to note the kinds of information that you will present in your analysis. (10 points)
Your second paragraph will describe the data analysis you performed. (50 points)
• Start by saying that the data came from the General Social Survey.
• Give the years of the survey you used and state how you defined the sample for analysis (that is, which respondents were included in the data analysis).
• State what the total UNWEIGHTED sample size was for each crosstabulation you performed.
• State what variables you crosstabulated. NOTE: In your report, label the variables in terms of what concepts they measure rather than the GSS variable names you typed in.
• For the two crosstabulations, also state which variable was the dependent variable, and which variable was the independent variable.
In a third paragraph, report the results of your first crosstabulation, in which you compared the percent very happily married across categories of the frequency of spending time with friends. Report the general pattern the table showed, as well as specific percentages. (You don’t have to report the percent who were not very happily married.) Also indicate whether the relationship between the two variables was statistically significant, and give the p-value. (15 points)
In a fourth paragraph, report the results of your second crosstabulation, in which you compared the percent very happily married across categories of the frequency of spending time with relatives. Report the general pattern the table showed, as well as specific percentages. Also indicate whether the relationship between the two variables was statistically significant, and give the p-value. (15 points)
