Can apply to: Primarily journal articles, but also other types of research outputs, such as web pages, books and conference papers, which users add to their Mendeley reference libraries
Metric definition: The number of Mendeley users that have added a particular document to a Mendeley library. Aggregated demographic information, such as geographic location and discipline for Mendeley readers, are also available.
Metric calculation: Mendeley readership is calculated as the number of unique Mendeley users that have added copies of a particular document to their personal library. Demographic data is user supplied, aggregated, and anonymized.
Data sources: Mendeley personal libraries and user-supplied demographic data.
Appropriate use cases: Mendeley readership counts are best used to understand early scholarly attention to individual research documents. Studies have found positive correlations between Mendeley readership counts and later citation activity (Li & Thelwall, 2012; Zahedi, Costas, and Wouters, 2014; Thelwall, 2018).
Limitations: Mendeley’s coverage of science and medicine research is stronger than its coverage of the social sciences and humanities (Mohammadi & Thelwall, 2014), especially for books. While several investigations have demonstrated moderate to strong correlations between Mendeley readership counts and later citation activity, this relationship is less significant for articles in the humanities and the social sciences (Mohammadi & Thelwall, 2014).
Mendeley has developed a method for clustering different copies of the same document that are stored in different locations around the Web, so the number of readers is calculated across all of the different versions of the same document. Searching Mendeley via a persistent identifier, such as a DOI, can be useful and efficient strategy for retrieving readership count across duplicate files, but it is not 100% accurate.
Inappropriate use cases: Mendeley readership data should not be used as a direct measure of research quality or impact.
Transparency: Mendeley readership data is freely available via the Mendeley API, but it only includes readership counts for the three most frequently occurring user types for each document (Haustein & Lariviere, 2014). Additionally, Mendeley has started registering its data with the Crossref’s Event Data service. However, it is not possible to see exactly who has bookmarked a reference in Mendeley due to user privacy restrictions. Bookmarks are reported anonymously and in aggregate only.
Timeframe: Mendeley was founded in 2007, but in theory research of any age can be added to a Mendeley library.