Skip to Main Content

Cite Data and Statistics

How to cite numeric data and statistics

General Guidelines

When you use numeric datasets or a prepared statistical table you must cite where you retrieved the information.  Data and statistical tables contain unique elements not specifically addressed by most citation styles.  Citations for data or statistical tables should include at least the following pieces of information, which you will need to arrange according to the citation style you use.  

  • Author or creator - the person(s), organization, issuing agency or agencies responsible for creating the dataset
  • Date of publication - the year the dataset was published, posted or otherwise released to the public (not the date of the subject matter).
  • Title or description - complete title or if no title exists, you must create a brief description of the data, including time period covered in the data if applicable
  • Publisher  - entity (organization, database, archive, journal) responsible for hosting the data 
  • URL or DOI  - the unique identifier if the data set is online

Certain styles may also ask for additional information such as:

  • Edition or version
  • Date accessed online (Note: APA does not require this)
  • Format description e.g. data file, database, CD-ROM, computer software

Tips for finding additional citation guidance:

  • Check to see if the publisher or distributor of your dataset provides suggestions for citing their data.  For example data providers like OECD and repositories like ICPSR and Dryad offer guidance for formatting citations to the hundreds of datafiles they host or produce.
  • Look through your style manual for instructions on using a similar format such as citation styles for electronic resources, electronic references, web pages, or tables.

This guide provides information for citing data and tables to include in your bibliography.  Consult the Purdue OWL for guidance on incorporating data and statistics in the body of your paper.

This guide is intended as a guideline only, check your citation manual, ask a librarian, or confer with your professor if your specific data set does not contain the elements needed to draft a useful citation. In general, it’s better to include more information than called for than to leave out information that could help the reader locate data you cite.

Examples - APA Style

Unless otherwise noted, the basic elements and guidelines described here are from the Publication Manual of the American Psychological Association, 6th edition (McHenry Reference Desk BF 76.7 .P83 2010).  You may also wish to consult the Purdue OWL or How to Cite Data from Michigan State University for MLA examples and explanations.

Notes:

1. Include format type in brackets [ ] to describe format, not title information (e.g. data set, data file and codebook).  [See APA guidelines for "Nonroutine information in titles" (pp. 186)]

2. Use “Available from” if the URL or DOI points you to a website or information on how to obtain or download data at a general site that houses data sets. Use “Retrieved from” if the URL or DOI takes you directly to the data table or database. (APA Style Manual, 2001 ed., pp.281 or Purdue OWL Electronic Sources: Data Sets)

I. Data sets:

 

Basic Elements: [Follow APA guidelines for "Data set" (pp. 210-211) or online from MSU] 

Author/Rightsholder, A. A. (Year). Title of publication or data set (Version number if available) [Data File]. Retrieved from (or available from) http://xxxx

The title of the data set should be italicized unless the data set is included as part of a larger work or volume

Example:

The World Bank, World Development Indicators (2012). GNI per capita, Atlas method  [Data file]. Retrieved from http://data.worldbank.org/indicator/NY.GNP.PCAP.CD

Example of Table generated from an interactive data set:

Bureau of Economic Analysis, U.S. Department of Commerce (2013). U.S. Direct Investment Abroad, All U.S. Parent Companies 2009-2010. [Data file].  Available from BEA.gov/iTable 

II. Table from a publication 

Basic Elements: [Follow APA guidelines for "entry in a reference work" (p. 205)] 

Author. (Year). Title of entry. In Editor (Edition), Title of publication (pp. xxx-xxx). Retrieved from http:// OR Location: Publisher OR doi:xxxx.

Example: (Note: Editor & Edition elements are not applicable in this example)

World Trade Organization. (2012). Table I.3: World merchandise trade and trade in commercial services by region and selected economy, 2005-2011.  In International Trade Statistics, 2012 (p. 22).  Retrieved from: http://www.wto.org/english/res_e/statis_e/its2012_e/its12_toc_e.htm

The title of the data set should be italicized unless the data set is included as part of a larger work or volume, as in the example above.  

Quick Guides to Citing Data

What is a DOI?

DOI stands for Digital Object Identifier and is a unique number used to precisely locate electronic items like webpages, articles, files, etc.  A DOI is persistent, which means it does not "break" the way a URL can when a website is updated.