Live Chat

# Conduct a multivariate regression using given data

Instructions: Your assignment is to conduct a multivariate regression using this data. Explore these variables in the GSS 2012 codebook and data, and think about the possible relationships that might exist between them, specifically between the independent and dependent variables. Come up with a question to test about the possible relationships between the dependent and independent variables, and then conduct a regression to test those hypotheses. Include a literature review where you explain the sociological grounding of your project and research questions using established academic research. (Using STATA 13) Paper is supposed to have tables, graphs, charts interspersed throughout the writing in a way that flows. Should follow format: Introduction, Literature Review, Variables and Measurement, (Measurement of independent variables, w descriptive table, measurement of dependent variables), Results, Discussion)

Guidelines for Project -

1) Come up with a specific research question and hypothesis:

First you will want to check your assigned variables. You will have 6 total variables, one dependent, and 5 independent. These variables are thematic, selected because they may have a relationship to each other.

You'll want to check these variables and construct a hypothesis around them. For example, "I predict that "Independent Variable X" will have an effect on "Dependent Variable," even when accounting for other variables that may confound the relationship." Your hypothesis can involve prioritizing multiple independent variables, if you'd like. Make sure to describe what your prediction is as to a possible relationship. Specify your null and alternative hypotheses.

2) Use the specific research question to guide your literature review:

Once you have this research focus, you can do a literature review that checks for prior academic research on the issue. If I'm interested in the relationship between education and political views, for example, I'd want to know what people's political views generally are,what factors tend to affect those views, and if there was any research specifically on the relationship between education and political views. I'd also want to include a brief review that checks if other independent variables have a relationship to political views.

Next you'll want to describe the variables that you are using in your project. What dataset did you use? Where did the data come from? What variables are you using? Did you recode any of them? What do they measure? Assume the reader isn't familiar with the GSS at all, and walk them through your data sources clearly and thoroughly.

Then, present a descriptive table for your data (We'll be going over this in more detail), and describe it in some detail.

4) Describe your regression set up and results:

Then, after describing the data, you can walk the reader through how you set up your regression, then present your regression table (again, we'll show you how to do this), and discuss the results. What do the regression results say about your data and the relationships you were testing? Which hypothesis is supported by the results?

5) Discuss your project, its findings, its limitations, and future implications:

Finally, you can use the discussion section to overview your project. Summarize your project, what the literature described, what you tested, what your findings came out to be. Discuss the limitations of the project in terms of time, data, etc, and what you think your results say about society. What else can be done in the future by other researchers interested in this topic?

(Class site offers a download of 2012 GSS data to pick variables from and to use as dataset.)