Carnegie Mellon Institute for Strategy & Technology: 84-266 Research Design for Political Science: Data and Statistics

Data and statistics are gathered and disseminated by many organizations, federal agencies, individual researchers, companies and more, often making it difficult to find just the data or statistic you need. Here are a few tips to keep in mind when you embark on a search for data and statistics:

Data vs. Statistics:

While the terms data and statistics are often used interchangeably, they actually aren't the same! Data refers to the raw information that is collected and if often in the form of a spreadsheet, where each row represents one case. Statistics are summaries of data, often provided as a percentage, proportion, or average value. Do you need raw data or a summary statistic to answer your question?

Time frame and geographic scope

Think about what time frame and geography you're interested in. Do you need historical information, or information spanning several years? Or just the most current information? Do you want information at a country, state or local level?  Do you need international, non-US data or statistics?

Who collects and disseminates the data?

Think about who would likely collect the data and disseminate the data or statistics for you topic. Would this data be collected by a large federal agency in a nationwide survey? If so, which agency? Or would it be more likely collected by a non-profit organization at a local scale?

Consider different sources

There are many places to look for data. Here is a series of search strategies to try. You'll find links to many of these sources in this research guide.

1. Data aggregators through the Libraries: These are searchable databases that bring together data and statistics from many different sources. This can be a good place to start. Examples are DataPlanet Statistical Datasets, Statista, Insider Intelligence and Roper iPoll.
2. Data producers: If the above aggregators don't work, look for data and statistics directly from the agencies and organizations that are likely to collect and disseminate that data. You'll find some of these in the boxes below.
3. Published literature: Researchers regularly use publicly available datasets, or their own datasets that they share, or report useful statistics in their publications. Try searching databases like those on the Finding Research and Industry News tab of this guide.
4. Targeted internet searches: If you still haven't found what you need, try searching Google and Google Dataset Search. Add terms like 'data' or 'statistics' to your search terms. You can also add filetype:xls to limit your search to Excel files.
5. Ask the Library! Contact Sarah Young, your liaison librarian, for more help finding data and statistics.

Try Google's dataset search. Search the web for datasets on any topic.

The library offers access to a number of commercial products that aim to facilitate the discovery and use of data from public and private sources.

The Military Balance+ is the online defense database from The International Institute for Strategic Studies, that allows researchers and users to plan, analyze and enable delivery of policy and commercial objectives for the private sector, governments, armed forces and media. The database includes equipment data, defense budget data, defense procurement data, military deployment data and policy and capability analysis. To request a live demo or tutorial, please contact Sarah Young, sarahy@andrew.cmu.edu.

Microdata are data for which the statistical units are individual people or objects. These data have not yet been aggregated into statistics about groups, and thus can be downloaded, manipulated, used in statistical models, etc. Below are a few sources for large population and economic microdata sets.