Welcome to the Social Science Statistics and Data Guide. For further assistance, please contact one of the Guide editors or the subject librarian for your major or program.

After you've located your datasets or statistics, you may wish to consult a specialist in our Research Data Services unit for assistance with services ranging from data visualization to text mining.

Data or Statistics?

Do you need statistics or data? Often, "how many" or "how much" questions may be answered with a few statistics.

Questions which seek to understand relationships between variables - how and why questions - may require work with one or several datasets. For additional background, you may find this brief slide presentation - Data or Statistics - helpful.

Key questions

To search for data effectively, you must define a topic and devise a research strategy. This page provides a few questions to help you plan your work.

What:  What kind of topic or phenomenon are you studying?

  • Are you searching for demographic data, organizations, commodities, or industries? Something else?

Where:  What geographic area(s) are you studying?  

  • Are you searching for information on a broad scale (e.g. country or state) or a more granular level (e.g. voting districts or neighborhoods)?

When:  What time frame(s) are you interested in? 

  • Remember that there's often a lag between collection and publication of data and statistics; subsequently, the most current information may be several years old.
  • What frequency do you require - monthly, quarterly, annual, something different?
  • Remember...more data are available each year!  Information collected in recent years may not have been collected 10, 20, or 50 years ago.  If you plan to do historical work, investigate the datasets available in earlier years before proceeding.

Who:  Who would be likely to collect data and generate statistics on your topic?  Some candidates might be:

  • Government agencies
  • Non-governmental organizations (NGOs)
  • University research centers
  • Private sector organizations (note: these proprietary datasets may be difficult/expensive to retrieve)

How:  How were/are the data collected/organized?

  • Time series
  • Longitudinal (panel)
  • Survey
  • Cross-sectional survey
  • Administrative
Your topic may need to be defined flexibly enough to fit the information available unless you're building a dataset from your own survey results or other materials.