Soc709, Class paper and everything else page
Guides for using
Stata
This class assumes you used Stata in the first semester statistics class. In general, I will provide you with the Stata syntax you need as we go. However, I strongly recommend looking at the following websites that will go through the basics of using Stata.
UCLA resources for learning Stata, particularly the data management module.
CPC Stata tutorial, gives you an overview of important Stata commands and concepts.
Data for the class
paper
If you don’t know what data set to use for your class paper, I can help you with the following data sets.
National Longitudinal Study of Youth
Current Population Survey via NBER
American Community Survey and Public Use Micro Data (1860-2000) via IPUMS
I have Stata versions of the 2001 WVS (and codebook), the 2005 CPS, and the NLSY that I can give you if you want to jump right in.
IPUMs is very user friendly, so you can make extracts from there. A word of advice: don’t take too many variables, you will wind up with a very big data set. You may want to restrict your analysis to certain states or sample from the cases.
You are welcome to use any data set of your own that you have.
Class Paper
We are going to work on the class paper in steps in order to integrate the data management and analysis directly into class lecture.
The goal of the class paper is to write a 12-13 page paper with the following sections:
1) abstract
2) introduction (2 pages)
3) brief literature review (2 pages)
4) data section (2 pages)
5) descriptive analysis (3 pages + tables)
6) multivariate analysis (3 pages + tables)
Step 1:
a) Identify the data set you want to use.
b) Write a brief abstract of what you intend to do. (You will modify this later as you work on your paper).
c) identify what your dependent variable and key independent variables will be.
Step 2
a) get the data into Stata.
Options: DBMScopy is available on the Oasis citrix server, oasis.unc.edu. See the class reserve page for information on logging into citrix.
See help infix, infile, insheet for importing data into Stata.
IPUMs (www.ipums.org) will create a Stata do file to help you read the data into Stata. Here is an example of what it looks like.
See the UCLA webpage for converting between SPSS, SAS, and Stata, or use DBMScopy.
b) when you get your data into Stata, type “des” to see what you have.
c) summarize your variables. Note: if you have lots of variables, just do this for the ones you intend to use in your analysis.
d) Do a descriptive analysis on your data. Use this do-file, basic_descriptive.do (and this data, basic.dta, if you want to run it) to provide an illustration of the syntax you can use. Can you show the basic relationship between your dependent variable and the independent variables?
Step 3
a) rough draft of the introduction and literature review of your paper (1 page on each)
key issues to include
later:
Recoding your data and creating new variables.
Missing data
sampling weights
getting tables into Stata