Fall, 2005 (Roos, Soc. 501)

Assignment 2: Developing Research Hypotheses and Proposal Presentations
   (presentations due Wednesday, October 5th; email to class by 5 p.m. Monday, October 3rd)

This assignment is designed to get you started on your final project. Unless you have your own data set, you will all use the same data set, the General Social Survey (GSS). This data set has a large number of variables available for a random sample of the U.S. population in 2002. The 2002 GSS codebook is available online [click here]. Use "Analyze" button to get access to the complete 1972-2002 codebook.

Here's the plan of action:

1) Skim through the codebook to get a sense of what variables are included. Search through it more carefully to select 5 to 10 variables you would be interested in spending the rest of the semester investigating. Think final project. For your final paper, I will ask that you write a research paper investigating a set of research hypotheses you develop.

What you're doing here is starting with your indicators (operationalization of your variables) and inductively asking what substantive questions you can address with these data. Make sure you have at least one dependent variable (you may want to choose a few to investigate--i.e., several measures of the same concept), and a set of independent variables you believe affect that dependent variable. Be thinking of possible control variables. Don't forget to include basic demographic variables that may be relevant (e.g., age, sex, race, ethnicity, education). Feel free to select interval level variables, although for this assignment, please recode them into categorical variables.

2) Decide what your sample will be. For example, if you are investigating the differing occupational distributions of male and female workers, you may want to limit yourself to those members of the sample who currently work.

3) Use the GSS web site to generate a simple frequencies distribution on each variable you've chosen. In addition, generate a bivariate (2-way) crosstabulation of two variables--your major dependent variable and your major independent variable. You can play around with generating 3-way tables (i.e., adding in test variables), but since that will be a future assignment don't write about the trivariate tables yet. I'll go through an example in class, but here's a hint about how to get started: from GSS site, click "Analyze," then "Frequencies or Crosstabs."

4) To write this all up: Again, I want you to think sociologically. Think back to the actual model of science that we talked about in class. Use this to help structure your writeup. What I want is a proposal that outlines what you will be investigating over the semester. What is your theory, what is (are) the hypothesis(es) you will investigate, how will you conceptualize and operationalize your variables, what data and what analytic techniques will you use to investigate your hypotheses, and what additional variables would you consider as control (test) variables to reformulate your theory? Some of this is straightforward--i.e., you will be doing secondary data analysis, using the GSS data, and your major analytic technique will be crosstabulation (or perhaps t-tests of means).

As part of your proposal, interpret the 2-way crosstabulation you ran. Include a table (properly titled and labelled). Do not just xerox the computer output; use the computer- generated table to create your own table.

Keep your writeup to 2-3 double-spaced pages. Your assignment must be typed. Use Word or Excel to type tables.

Proposals must be emailed to class members by 5 p.m., Monday, October 3rd. Please come to class prepared to present your work, and to provide comments for your classmates.