¶ 1 Leave a comment on paragraph 1 2 Altmetrics are an emerging set of techniques, and an area being intensively researched and written about. The need for altmetrics is founded on the assumption that the vast volume of scholarly literature available, means that having effective filters is crucial to making sense of it. The altmetrics manifesto refers to three filters that are used in academia: peer-review, citation-counting and journal impact factor (Priem et al. 2010), and reasons that none of these filters are any longer fit for purpose. This is in part due to the scale of scholarly publishing (Jinha 2010; C Neylon & Wu 2009) and also because a notable volume of scholarly interaction now takes place outside of the traditional venues like journals or conferences, largely due to the advent of the web.
- ¶ 3 Leave a comment on paragraph 3 0
- Online reference managers like Mendeley
- Microblogging services like Twitter
- Recommendation systems like Digg, Slashdot or Stumbleupon
- Code-sharing services like GitHub
- View/download statistics
- Conversations on social media sites and blogs
- Feedback from an open peer-review like Faculty of 1000
- Social bookmarking services like CiteULike or Reddit
¶ 4 Leave a comment on paragraph 4 1 Altmetrics differ from tradmetrics in that they can “measure impact at the journal article level as evidenced through social media activity as well as impact measured by examining other significant research output. The new metrics offer the possibility to discover new insights into impact that have been previously impossible to obtain, and they are fast compared to traditional metrics” (Galligan & Dyas-Correia 2013). Speed is in fact one of the key differentiating factors from tradmetrics and creates the opportunity for “real-time recommendation and collaborative filtering systems” (Priem et al. 2010). These kinds of systems can potentially help build connections between researchers and authors in a similar way that citing related work does, however when the connections are made through social web-based systems they become evident immediately and aren’t tied to review and publication schedules.
¶ 5 Leave a comment on paragraph 5 1 Since the release of the 2010 altmetrics manifesto numerous papers have been published on the subject exploring the potential applications, limitations, and what the knock-on effects of altmetrics may be. Galligan & Dyas-Correia (2013) raise a number of pertinent questions that give a good feeling for the landscapes that altmetrics are changing. The answers aren’t static and are constantly evolving as the field develops:
- ¶ 6 Leave a comment on paragraph 6 0
- What tools are available?
- What are the limitations of current tools?
- Why are altmetrics of interest?
- How do altmetrics relate to predicted/actual impact?
- What will the impact on peer-review be?
- What’s the relationship between altmetrics and open access journals?
¶ 7 Leave a comment on paragraph 7 0 Specifically coming from a scientometric perspective, Priem & Hemminger (2010) conduct a review, describing why altmetrics are useful (in terms of with tradmetrics’ shortcomings) and also provide a curated list of potential data sources for altmetric analysis. In further work studying altmetrics ‘in the wild’ Priem et al. (2012) demonstrate and showcase the value of altmetrics in practice. One conclusion of this work is the suggestion that having “flavours” of impact based upon observations such as “some articles may be heavily read and saved by scholars but seldom cited”. The significant problem of gaming altmetrics is also discussed.
¶ 8 Leave a comment on paragraph 8 0 Consensus of those writing on the subject coalesces around the sentiments that “Altmetrics give a fuller picture of how research products have influenced conversation, thought and behaviour” (Piwowar 2013). Clay Shirkey’s words (2008, cited in C Neylon & Wu 2009) “you can complain about information overload but the only way to deal with it is to build and use better filters” is also strongly resonant.
¶ 9 Leave a comment on paragraph 9 1 The corpus of literature that discusses altmetrics gives a compelling account of why an alternative view of metrics is necessary based on problems (with tradmetrics) and opportunities (the potential to exploit newly available data). Scholars have demonstrated how altmetrics work in practice. Altmetrics are inherently based upon socially relevant data, and indeed data that is usually generated through social interactions, but thus far implementations of altmetric compasses have failed to account for the value of stimulating or nurturing those interactions. Altmetrics demonstrate that social interactions can be used as a measure of impact or influence, stemming from a particular piece of research. It does this through tracking the signatures of these social interactions. Whereas tradmetrics are structurally impotent in this regard, altmetrics fall short in terms of CoP in that they simply don’t attempt to account for the tacit value of the social interactions that they are measuring the signature of. Is this an oversight?
¶ 10 Leave a comment on paragraph 10 1 The signatures are undoubtedly useful tools for measurement, as are the tradmetrics derived from citation counts, however CoP theory suggests that the actual social connections – crucial to social learning – have value too. The characteristics that describe CoP can be abstracted to describe how the interactions that generate altmetric datasets come about.