New Capture Metrics Launched

image

PlumX tracks 5 categories of metrics:  usage, mentions, social media, citations, and captures.  Metrics that we categorize as captures are when an individual is interacting with research online and “captures” it for later.  Examples of capture metrics that are already a part of PlumX are:

  • Bookmarking the page on Delicious
  • Adding it to Elsevier’s Mendeley as a reader
  • Favoriting a slide presentation on Slideshare
  • Favoriting a video on YouTube
  • Subscribing to a channel on Vimeo or YouTube
  • Followers, forks, or watchers of source code on GitHub

All of these actions indicate that a user might want to later come back to this research.  For example, a researcher finds an article and adds it to their Mendeley reader or bookmarks it in a bookmark manager such as Delicious so they can go back and read it later.

In the scholarly space, studies have shown that capture metrics are a good leading indicator of future citations.  For example, In the research article When are readership counts as useful as citation counts? Scopus versus Mendeley for LIS journals  (Open access version here), the researchers Maflahi and Thelwall introduce their research study as:

This article investigates the influence of time on the number of Mendeley readers of an article through a theoretical discussion and an investigation into the relationship between counts of readers of, and citations to, four general Library and Information Science (LIS) journals. For this discipline, it takes about seven years for articles to attract as many Scopus citations as Mendeley readers and after this the Spearman correlation between readers and citers is stable at about 0.6 for all years. This suggests that Mendeley readership counts may be useful impact indicators for both newer and older articles.

This week, we added new capture metrics to PlumX that continue to expand this important category to include data from EBSCO’s EDS discovery, EBSCOhost, and eBook platforms.  These new captures are calculated per artifact, and measure when a user emails, exports, prints, or saves an artifact’s citations.  More specially these actions are defined below:

  • Emails: The number of times an artifact’s citation, abstract and full text (if available) have been emailed
  • Exports: The number of times an artifact’s citation has been exported direct to bibliographic management tools or as file downloads
  • Print-outs: The number of times an artifact’s citation/abstract and HTML full text (if available) have been printed
  • Saves: The number of times an artifact’s citation/abstract and HTML full text (if available) have been saved

Some of these metrics comprise a part of the “hidden web.”  Being able to calculate the number of times users on a platform email or print out a citation, is a glimpse into what work is being saved for later more than others.

The metrics we can collect are of course only a proxy for ALL places and times that research has been saved for later.  We’ve expanded these new capture types using data from EBSCO, and will expand our sources over time as others become available to us.

Given the scale of this data we have just exposed, some PlumX customers will see dramatic increases in capture counts in their profiles. This is one reason why it’s so challenging to distill alternative research metrics down to a single number: what happens when new metrics sources start to get collected?  Our approach is now, as it always has been, to collect as comprehensive a collection of indicators as possible.  Using the data should always be done in context and through benchmarking against other items of interest.