advice on tracking article level metrics (UNCLASSIFIED)

[David Wojick]

Nancy,

You might try looking at the spread of the language. Articles frequently 
introduce new language, sometimes coining words but more often coining 
descriptive phrases. This is because every research article (as opposed to 
review articles) is talking about something new in the world, or it would 
not be published. New language is often required to do this. "Quantum dot" 
for example.

The spread of this language may be an important marker for the spread of 
the idea introduced in the article, which is what we sometimes mean by 
impact. I call it natural citation (thus creating new language). Note that 
since impact tends to be a diffusion process we do not try to track all the 
paths. Hence the idea of a marker.

David

At 07:28 AM 10/30/2013, you wrote:
>Adminstrative info for SIGMETRICS (for example unsubscribe):
>http://web.utk.edu/~gwhitney/sigmetrics.html
>
>Classification: UNCLASSIFIED
>Caveats: NONE
>
>I'm trying to demonstrate the complexity of impact on a few articles 
>published
>OA, and there are a few ways I plan to track impact.  But I would appreciate
>advice on what I'm missing and how to capture it.
>
>We'll look at Mendeley, altmetricit, web analytics, related tweets/impact.
>Additionally, I thought a blog post through an academic library that offers
>doctoral degrees in that subject might draw attention and allow us to cross
>promote our jobs.
>
>What are the other thousands of things we're missing in the impact?
>
>Nancy
>
>
>
>Classification: UNCLASSIFIED
>Caveats: NONE
>