Credit where credit’s due – citing datasets and data papers
24/08/2016
Why would you cite datasets?
Citing datasets ensures credit is given to their creators and that data impact is tracked, as data is an important research output in its own right. If you use other people’s datasets, you should cite them as you would cite their papers. If you publish your own datasets and they underpin an article, you should cite the dataset in its own right, with an in-text pointer to an entry in the reference list. Whilst you may already be including data access statements to meet funder requirements, full citations are valuable because data statements don’t treat data as a first-class record of research and don’t give due credit to the creators, especially where they’re different from the article authors.
How do you cite datasets?
As with other types of citation, advice on the exact elements to include can vary by style. The key elements are the identifier (see our post on DOIs), creator(s), title, date, and publisher. An example reference in Harvard-Cranfield would look like this:
Partridge, M.C. (2014) Spectra from LPG sensor repeat testing with 100ppm toluene. Cranfield University. 10.6084/m9.figshare.1004753.v1
In systems such as CORD, our data repository, there are various citation options on CORD item pages, with links to export the record directly to Mendeley or other reference managers, and a ‘cite’ tab that offers a cut-and-paste citation in a generic format. You may find the Library’s Quick guide to the author-date referencing style (internal pdf) useful, including tips for referencing with Mendeley.
How do data papers fit in?
A data paper is a paper that describes a dataset rather than drawing conclusions from it, and is usually published in a dedicated data journal, such as Nature’s Scientific Data. (This list of data journals may be helpful.)
Some people like to write data papers to promote their dataset, and to make it easier for others to credit you, as academics are more familiar with citing papers than datasets, although this is likely to change over time. If you write a data paper, a citation of the data paper counts as a citation of the dataset, so all uses of your data are acknowledged.
For more information, see this great UK Data Service pdf on data citation.
Image: thanks merci by infrogmation at https://www.flickr.com/photos/infrogmation/4136308490/, CC-BY 2.0.
Categories & Tags:
Leave a comment on this post:
You might also like…
Systematic Literature Review – Identifying your search terms and constructing your search strings
Our previous posts on the systematic review have looked at getting started and selecting your sources. In this post we will look at the next fundamental stage: Identifying your search terms and constructing your search ...
Navigating the World of Robotics: My Journey at Cranfield University
Hey there, I'm Manideep, and I'm thrilled to share my experience pursuing an MSc in Robotics at Cranfield University. Let me take you through my journey and how Cranfield became the ...
Exploring safer and smarter airports with the Applied Artificial Intelligence MSc group design project
Artificial intelligence (AI) technologies have experienced rapid development in recent years, spanning from large language models (LLMs) to generative artificial intelligence (GAI). These cutting-edge advancements have significantly impacted various aspects of ...
My aerospace manufacturing journey at Cranfield University
Hey there, I'm Abhishek and I wanted to share my journey into aerospace manufacturing, guided by my experiences at Cranfield University. Let's dive into how this remarkable institution shaped my career ...
Changes to Library Services over Easter, 29 March – 1 April
Libraries on the Cranfield site Both Kings Norton Library and the School of Management Library (Building 111, first floor) will be open 24/7 over the Easter weekend. You will be able to use the study ...
How to present well as a group
You will have put a lot of work into your research or project and want to show everyone what you have achieved or discovered, so you need to impart this knowledge as clearly as possible. ...