They are an attempt to disaggregate the impact of the article from that of the journal and thus provide a more granular and immediate measure of impact.
These impact measures recognize the increasingly diverse ways in which scholarly information is being communicated to audiences both within and beyond the scholarly community.
As outlined in Article Level Metrics: a SPARC Primer, these measures examine five broad elements:
Usage
How many times has an article been viewed on the publisher's site? How many times has the full text been accessed and/or downloaded? How many times has its supplemental data been accessed? These counts give a quantifiable sense of the extent to which an article is actually being read.
Captures
How often has an article been bookmarked on CiteULike? How frequently has it been shared within Mendeley? These and other "recommender systems" provide a unique window into what research scholars seek to highlight to their peers.
Mentions
What is the discussion around an article? How many times has it been blogged about? How many news stories have been written about it? Is it mentioned in Wikipedia? This element seeks to quantify the extent to which the article has seeped beyond a narrow academic audience and generated a broader conversation.
Social Media
How many Facebook likes does the article have? How many times has it been shared on LinkedIn and other networks? How many tweets about the article have been generated? These data points can contribute to a sense of an article's immediacy.
Citations
What articles are citing this article? This is a traditional metric that complements those above by demonstrating the long-term contribution an article makes to the scientific literature.
Tools you can use
Traditional citation counts featured in most databases such as "times cited", "citation counts" and "citation reports" are being enhanced with new tools. The links below showcase some of these emerging article level tools.
PLoS
Article level metrics are available for all PLoS articles upon publication. Metrics include HTML page views; PDF downloads; XML downloads; PubMed Central usage; citations computed by third party databases and search engines (including Google Scholar and Scopus); mentions on social networks such as Facebook and Twitter; mentions in blogs, as reported by aggregators such as Researchblogging.org and Nature Blogs; and feedback within the comments section of the PLOS website.
Scopus
In addition to its traditional citation index and citation reports, Scopus had included an application in partnership with Altmetric.com to track hundreds of thousands of links to scholarly articles on Twitter, Google+, Facebook, the scientific blogosphere, newspapers and magazines then matches that data to the search results or full text articles that you are viewing on SciVerse.
Nature Publishing Group
Like Scopus, NPG utilizes Altmetric tools to display quantitative and qualitative data regarding an article's social and scholarly reach. To view the metrics, choose any article and click on the article metrics button.
HighWire
HighWire is a publishing platform that delivers scholarly content on behalf of learned societies, university presses, and other publishers. In collaboration with ImpactStory, HighWire is making article level metrics an optional feature for publishers using their platform. The first publisher to implement these is eLife; click on an article then use the metrics tab to see the ImpactStory features.