Research Impact Indicators & Metrics

What it is:

Times cited for datasets or data papers is the same metric as times cited for articles or books.  Scholarly publications have cited your data in their references.  

What it is used for:

It is the primary quantitative metric for the scholarly impact of your shared data.  How often are others standing on your shoulders? 

Nota Bene:

• The practice of citing datasets as you would any other source is not yet standardized in many fields.  Many journals do not provide instructions to authors on how to cite datasets (Mooney, 2011).  Thus, the times cited count is likely to be lower than your data's actual use. 
• As with most metrics, citation rates and frequency for data vary widely by field.  
• The availability and discoverability of the data are significant factors in times the data are cited.  The times an openly deposited dataset with thorough metadata in a well-known disciplinary repository is cited should not be compared to a paid-access or "provided upon request" dataset in the same topic.

How to find it:

Different indexing databases count whether your dataset has been cited, vs. whether your data paper has been cited.  Data papers, or data publications, are published in journals just like any other paper in a journal.  They are often easier for others to cite than a dataset in a repository.  

Some of the tools for citations below can give metrics for both your dataset and your data paper: some will only show one or the other.  

Web of Science
Data papers are indexed in the Web of Science Core Collection like any other journal article. Search for your data paper to get the Cited By count and explore the list of citing articles.

Web of Science's Data Citation Index attempts to create a federated, indexed search of datasets across repositories. It is a useful tool for counting citations of your dataset and your data paper, and exploring ways your data have been used by others.

• Switch from the 'Web of Science Core Collection' to the 'Data Citation Index'
• Search by any of the citation information or identifiers associated with your dataset.
• Click the 'Times Cited' number to explore the publications citing your dataset.

Google Scholar

• Search for your dataset or data paper by DOI or title, as it would be written in a citation to it.
• 'Cited By' displays the count and links to the list of items citing it.

Dimensions
A comprehensive collection of linked data in a single platform; from grants, publications, data papers, and clinical trials to patents and policy documents.  It tracks traditional citations to your dataset and data paper, and also provides the Altmetrics Attention Score for many items.

DataCite - *Not usually applicable to data papers*
Global non-profit organization that provides persistent identifiers (DOIs) for research data and other research outputs.  They not only mint DOIs, but track resolutions of those DOIs.  Search their index to find your dataset and get the number of times it has been cited, viewed, and downloaded on the platforms DataCite tracks.


 

SciELO Data
SciELO, a "Scientific Electronic Library Online" is a publicly funded initiative set up to promote cooperative, free electronic publishing of scientific journals from developing countries, with a focus on Latin American, Spanish, and Portuguese journals.  "SciELO Data is a multidisciplinary repository for deposition, preservation and dissemination of research data from articles submitted and approved for publication, already published in SciELO Network journals or deposited in SciELO Preprints."  Their analytics aggregate downloads of the data.  Note: both the Data repository and analytics functions are currently in beta.
 

What it is:

A download is defined as a user clicking the link to save the dataset to their own device.  It is not a count of number of users who have visited the page with the download link.

What it is used for:

This metric is commonly used as a proxy for usage, or number of users who have consumed the information in your data.  It would be more accurate to view this metric as an indicator of interest in your data, enough to devote hard drive space to.

Nota Bene:

Saving a file does not equal a thorough read of the data, use in further research, or even that the file was ever opened by the downloader.  In the other direction, once the dataset is downloaded, a researcher may choose to share it with colleagues directly which would  short you those download stats download stats.  

Either way, it is a useful metric to track interest in your data, and downloads are a prerequisite for future citations.

How to find it:

If your dataset is posted to a repository, your author dashboard can give you the download count.  Depending on which repository you have posted in, you can explore where your data downloaded to, when, and potentially who is using it.  A few of the biggest repositories have additional features that can enrich your impact narrative.

If you have not made your data available in a repository but posted it to your research website or make it "available upon request" it's up to you to track the downloads directly.  Your website may provide usage statistics for the download link.  You can search your email for the number of times you've sent someone your data.  

GitHub
Primarily for software and code.
GitHub metrics can be used in your narrative of your work's impact, but it is important to contextualize the use.  Forks, pulls, and commits are counted by the Git software running on the Github.com servers, but the raw numbers don't tell you much about who is using it or to what extent.  Explore derivative projects as much as possible to flesh out your narrative.

OSF (Open Science Framework)
OSF Projects have their own built-in analytics to track popular pages and number of visits over time. You can view the impact of any public project by clicking its Analytics tab.

• View forks, links, and templates
• View page visit analytics
• View download counts on files

Data Repository @ ScholarWorks
UMass Amherst's data repository.  Free of charge, it accepts nearly all types of data, though with some limits on larger data sets or data sets with confidential or non-public information.  Deposit your data in the UMass Amherst data repository at ScholarWorks and see how often, where, and when your work is downloaded.