Skip to Main Content

APA – Citing Sources

Tags: English



A data set is a document that contains the collected raw data, or treated data, that forms the empirical basis for a study. This data can be reused for new studies by others who want to explore that data from other aspects, or for making new analyses and interpretations. A data set may consist of statistical data or interviews etc.

In many cases there is an organization behind the collected data, like a statistical agency for instance, in other cases, there might be one or several authors who have compiled the data for an article or a report. 

If a data set has a so-called doi (digital object identifier), which is a unique and consistent link, it should be used in the reference.

Write the data provider and year and italicize the title of the data set. You also write a suitable document type, such as data file and/or codebook (a codebook is used as a guide on how to interpret the raw data), in square brackets.  

If no year can be verified, write (n.d) instead (n.d. = no date).

Basic format:

Organization or author. (year). Title of data set [Data set]. https://xxxx

If there is a doi:

Organization or author. (year). Title of data set [Data set].


​Data set on a website (for more specific examples on different websites, check the chapter Websites:

Pew Hispanic Center. (2004). Changing channels and crisscrossing cultures: A survey of Latinos on the news media [Data set].

Unpublished data set obtained from an author or organization:

Bordi, F., & LeDoux, J. E. (1993). [Auditory response latencies in rat auditory cortex]. Unpublished raw data.

If the data set doesn't have any title, as in the example above, you write a description or use a name from some other place in the dataset (like Project data etc.) in square brackets, without italics.

In-text citations:

  • Based on statistics on media consumption of latin americans (Pew Hispanic Center, 2004), it becomes clear that . . .
  • (Bordi & LeDoux, 1993)
  • Bordi and LeDoux (1993) investigated how . . .

If an organization has an established acronym (abbreviation), you write the full name the first time you cite the source, and include the acronym within parentheses, and use only the acronym the subsequent times you cite the same source. Note that if you write the full name within parenthesis, you write the acronym within square brackets, but if you write the full name in the running text you state the acronym within ordinary parentheses.

First time:

  • Data on this is compiled yearly (World Health Organization [WHO], 2017), so a longitudinal analysis can be made . . .
  • The World Health Organization (WHO, 2017) compiles data on this . . .

Subsequent times:

  • Analyzing the data (WHO, 2017) it becomes clear that . . .
  • WHO (2017) continues to collect this data . . .