NORC General Social Survey

This page contains general instructions on how to create a dataset from the General Social Survey, which is "the only full-probability, personal-interview survey designed to monitor changes in both social characteristics and attitudes currently being conducted in the United States.” 

If you have questions, please feel free to email us at ilrref@cornell.edu

Notes on locating data in the GSS

Here are some basic instructions on utilizing the GSS Data Explorer interface to pull General Social Survey data for analysis. Please note that NORC recently introduced an entirely new interface for the GSS Data Explorer, so even if you have used this tool in the past, the new navigation is slightly different from the old version.

  • Create an account if you have not done so already.
  • Navigate to Search GSS Variables (the gray box in the top right corner)
  • Select your year or years of interest. (For the purpose of this assignment, you will want to choose a single survey year.)
    • Note: Not every question is asked every year, or on every survey ballot. Some years have up to three separate sets of questions that are asked to various participants, so you can filter by year and then by a specific ballot to ensure that you’re getting a set of variables/questions that were asked to every participant in your sample – otherwise, you could put together a list of great variables to support your research question but get a lot of NAs or blank cells in your sample. 
  • Select your variables of interest. Note that respondent year and ID will automatically be selected for you.
    • There are 6000+ variables total (and 1100 just for the 2018 survey!) so there are a lot to choose from. 
    • You can search for variables in the following ways: you can filter by year, can search by keyword, and you can filter by “tags” or modules under the Filter By: tab, or you can type a keyword to search. If you work with the GSS a lot, you can assign your own tags to variables, or you can click over to the GSS-assigned tags or the modules, which are the question categories from the survey.
    • Note: When you click on a variable, it gives you a year-by-year summary of the answer distributions. This is a great way to quickly check whether that variable will have enough data for you to sample from for this assignment. 
  • Creating your data extract:
    • Once you have selected all of your variables of interest, click on your username in the top right-hand corner, and then My GSS from the drop-down menu. (Note: if you have used the previous version of the GSS, this step replaces the previous data shopping cart.) You will see all of the variables you selected in the previous step.
    • Click the Extracts tab, and then the "Create Extract" button. 
    • Name your extract, and then click "Build Extract." Select your variables of choice (or click "add all") and then click Next.
    • Under "Choose Output Options," select your chosen year, and then your file format, which for this assignment will be "Excel Workbook (data + metadata)." In the Excel output options dropdown menu that will pop up at the bottom of the screen, select "Label & Question Text."
    • Click the "Create Extract" button.
  • Your extract will be added to the queue. Processing can take some time, but since you have created an account, you can return to this page later to download your extract.
  • Once your extract is completed, you can select and download your data extract. This will download your extract as a .tar.gz compressed file. Use your program of choice to open the archive or extract the files, and you will see a file called GSS.xlsx with your data. The Catherwood Library computers have the 7-Zip file manager software that you can use to do this extraction if needed.