Advanced Methods of Social Research Assignment -
This assignment is based on the ideas for research projects included in Appendix E of the course textbook (Healey, Joseph F. 2015. Statistics: A Tool for Social Research. Stamford: Cengage Learning. 10th edition).
Exercise - Select variables that are available in the 2004, 2010, and 2016 General Social Survey (GSS) to estimate characteristics of the U.S. adult population. You will use Stata to estimate and analyze sample statistics. These variables can be the same as those used in the classroom or in the previous assignment. See the GSS codebook or the GSS Data Explorer website for a list of variables available in GSS.
A. Dependent variable at the ordinal level of measurement
Include at least 3 bivariate tables (1 for each independent variable).
1. Utilize respondent's income at the ordinal level of measurement (rincome) as the dependent variable.
2. Utilize respondent's sex (sex) and choose two other independent variables from the nominal- or ordinal-level variables. These two extra variables should not be related to the respondent's age (e.g., age, agegr1, agegr2) or respondent's education (e.g., educ, degree).
3. Develop research questions and hypotheses about the associations between the dependent variable and independent variables. Make sure the causal links you suggest are sensible and logical.
4. Generate bivariate tables for each year. Place your dependent variable in the rows and independent variables in the columns of the cross tabulations. Show column percentages and column absolute totals for every table. Estimate chi square, Cramer's V, Lambda, Gamma, and Spearman's rho for every table you request.
5. Report and explain the results of these correlations. For each combination of variables, report the appropriate measures of associations and tests of significance. Analyze the statistical significance, strength (i.e., importance, magnitude), and pattern/direction of the associations between the dependent variable and independent variables.
B. Dependent variable at the interval-ratio level of measurement
Include at least 6 scatterplots (1 for each independent variable in each year) and at least 1 table with correlation measures. The scatterplots can be organized as six graphs in one page.
1. Utilize respondent's income at the interval-ratio level of measurement (conrinc) as the dependent variable.
2. Utilize respondent's age (age) and highest year of school completed (educ) as independent variables.
3. Develop research questions and hypotheses about the associations between the dependent variable and independent variables. Make sure the causal links you suggest are sensible and logical.
4. Generate scatterplots for each year. Place your dependent variable in the vertical axis and independent variables in the horizontal axis. Plot a regression line in all scatterplots. Show the least-squares regression equation for each scatterplot (it can be added to the footnote of the scatterplot).
5. Estimate Pearson's r, test of significance for Pearson's r, and coefficient of determination (r2) for each association of the dependent variable and independent variables for each year.
6. Report and explain the results of these correlations. Answer the following questions with scatterplots: (1) is there an association between the variables? (2) How strong is the association (i.e., importance, magnitude)? (3) What is the pattern/direction of the association? Furthermore, use scatterplots to check for linearity. Interpret the estimated Pearson's r, tests of significance for Pearson's r, and coefficients of determination (r2).
C. Regression models
Include at least 1 table.
1. Utilize respondent's income at the interval-ratio level of measurement (conrinc) as the dependent variable.
2. Utilize as independent variables: respondent's sex (sex), respondent's age group, highest educational degree completed (degree), and the other two independent variables selected above from the nominal- or ordinal-level variables (Part A).
3. Generate dummy variables for all independent variables, which will be used in the regression models.
4. Estimate one least-squares multiple regression model for each year, which should include all independent variables. Report estimated coefficients, robust standard errors (between parentheses), statistical significance (with asterisks), coefficient of multiple determination (R2), and sample size for each model. Include the standardized regression coefficients (i.e., standardized partial slopes, beta-weights) in a separate column for each model.
5. Report and explain the results of these models. Analyze the statistical significance, strength (i.e., importance, magnitude), and pattern/direction of the associations between the dependent variable and independent variables.
Explain the overall fit of each model (based on R2). Explain which independent variable has the strongest impact on the dependent variable (based on standardized regression coefficients).
Attachment:- Assignment Files.rar