2

Occasionally I need to present some figures that I calculated with custom scripts against a dataset, such as a dictionary, text corpus, or some form of lexical database. The calculations are my own, but the dataset was created by other researchers. When presenting my figures, how can I write an APA-style in-text citation in a manner which makes it clear that I relied on the dataset, but that I did not take the figures from somewhere explicitly stated in the dataset? I do not wish readers to blame that source I cited if my figures are wrong.

Here is a simple example:

'E' is the most frequently occurring vowel in the Spanish language and appears most often after the letter 's' (Brown, E. K., 2014).

6

If you are computing something from a dataset, then you need to explain what you have computed and how. This should resolve the question of citation fairly naturally. For example, with the example you have given, you might say something like:

Using the Spanish-language corpus in (Brown, E. K., 2014), n-gram frequency computation indicates that 'E' is the most frequently occurring vowel in the Spanish language and appears most often after the letter 's'.

Because the computation is yours, you need to explain and attribute it correctly, as you may have made a mistake in your interpretation, or the dataset itself may be biased. Thus, it is best to use a few more words and given the reader a clear explanation of the relationships.

1
  • +1. You need to use one phrase just to identify the dataset, and cite that identification, then use a separate phrase to state your result.
    – BrenBarn
    Nov 22 '14 at 7:26

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.