Confidence in the banking system. The General Social Survey (GSS) is a sociological survey used to collect data on demographic characteristics and attitudes of residents of the United States. The survey is conducted by the National Opinion Research Center of the University of Chicago, which interviews face-to-face a randomly selected sample of adults (18 and older). SDA (Survey Documentation and Analysis) is a set of programs that allows you to analyze survey data and includes the GSS as part of its archive. In Exercises 25.43 and 25.44 (text page 654) you used data from the GSS to study the relationship between the average respondent age and confidence in the banking system. A one-way ANOVA showed a statistically significant difference between the age of the respondent and his or her confidence in the banking system. Download the data file following the directions in Exercise 25.43, and do Tukey's multiple comparisons at the 5% overall significance level to compare the mean age for the three levels of confidence in the banking system. Which pairs of means are different?
Exercises 25.43:
Confidence in the banking system. The General Social Survey (GSS) is a sociological survey used to collect data on demographic characteristics and attitudes of residents of the United States. The survey is conducted by the National Opinion Research Center of the University of Chicago, which interviews face-to-face a randomly selected sample of adults (18 and older). SDA (Survey Documentation and Analysis) is a set of programs that allows you to analyze survey data and includes the GSS survey as part of its archive. Go to the Web site sda.berkeley.edu/ and click on Archive. Unless there is a more recent file, open the 1972-2010 cumulative data file (without the quick tables option).
(a) In the "Analysis" tab at the top of the page, click on "Comparison of means". Do an ANOVA that examines how the mean age of the respondents varies with their confidence in the banking and financial systems. To do this, type in the dependent variable as "Age" and the row (treatment) variable as "Confinan." For the selection filter, type in "Year (2010)" or the most recent year available. For weight, change it to "noweight." Finally, in the table options, the only boxes that should be checked are "Std dev," "N," and "ANOVA stats." Make sure that the checks are removed from the other boxes. Now click on "Run the table."
(b) How many respondents are included in the analysis? What are the three means and standard deviations? Explain how the degrees of freedom were obtained. What are the F- and P-values? Write a brief report explaining the relationship between the average respondent age and confidence in the banking system.
Exercises 25.44:
Confidence in the banking system, continued. This exercise is a continuation of the previous Web exercise. You will download the data file and reproduce the analysis, as well as provide some additional plots. First, open the 1972-2010 cumulative data file following the instructions in the previous exercise.
(a) In the "Download" tab at the top of the page, click on "Customized Subset." For the data file, if you highlight the CSV bubble, an Excel spreadsheet will be downloaded (unclick "Codebook"). For the selection filter, again type in "year (2010)" or the year used in the previous exercise. In the box for entering the names of individual variables, enter "Age" and "Confinan." Click "continue" at the bottom of the page. In the new window, click on "Create the Files," and in the next window click on "data files." You can now either open or save the data file to your computer.
(b) Import the data into your statistical software package. You first need to "clean" the data a little because there are observations for which either the "Age" or the "Confinan" variable is missing. For the "Confinan" variable, any value other than a 1, 2, or 3 is a missing-value code. Delete these observations. For the "Age" variable, the missing-value codes are 0, 98, and 99. Eliminate any observations with these values for "Age." You should now have the same number of observations as in the previous exercise.
(c) Draw comparative boxplots of the age distribution for the three values of "Confinan." Describe the shapes of the three distributions. What information can you obtain from the boxplots that was not included in the output for the previous exercise?
(d) Reproduce the one-way ANOVA table using your software. Your results should agree with those of the previous exercise.