For help or advice with any Measuring Research Impact / Bibliometrics query please email eprints@soton.ac.uk.
This is an emerging movement to use a wider range of metrics in addition to traditional citations. This can include discussion (e.g. Twitter), saving (e.g. Mendeley, Delicious) and sometimes viewing or recommending an item.
This can give a better picture of impact, but until this is better understood probably needs to be used with caution.
We have a lot more information on our guide to altmetrics.
Altmetrics are alternative metrics used to measure the impact of research.
The term altmetrics was first proposed in a tweet by Jason Priem in 2010, and further detailed in a manifesto.
The term is not clearly defined and can be used to mean:
Impact measured based on online activity, mined or gathered from online tools and social media for example:
OR
Metrics for alternative research outputs, for example citations to datasets.
OR
Other alternative ways of measuring research impact.
Altmetrics can be used as an alternative, or in addition, to traditional metrics such as citation counts and impact factors.
Image: Altmetric bookmarklet result for the article: Piwowar, H. (2013). Altmetrics: value all research products. Nature, 493(7431), 159-159. Screenshot taken Nov 4 2014.
| Speed | Altmetrics can accumulate more quickly than traditional metrics such as citations. |
| Range | Altmetrics can be gathered for many types of research output, not just scholarly articles. |
| Granularity | Altmetrics can provide metrics at the article, rather than journal level, such as journal impact factors. |
| Detail | Altmetrics can give a fuller picture of research impact using many indicators, not just citations. |
| Non-academic | Altmetrics can measure impact outside the academic word, where people may use but not cite research. |
| Sharing | If researchers get credit for a wider range of research outputs, such as data, it could motivate further sharing. |
| Standards | There are a lack of standards for altmetrics. |
| Unregulated | Altmetrics could be manipulated or gamed. |
| Reliability | Altmetrics may indicate popularity with the general public rather quality research. |
| Time | There is no single widely used rating or score and altmetrics can be time consuming to gather. |
| Difficulty | Altmetrics can be difficult to collect, for example bloggers or tweeters may not use unique identifiers for articles. |
| Overload | There are many different metrics and providers to choose from and it can be hard to determine which are relevant. |
| Acceptance | Many funders and institutions use traditional metrics to measure research impact. |
| Context | Use of online tools may differ by discipline, geographic region, and over time, making altmetrics difficult to interpret. |
An article by Linda Aiken in 2014 is a good example of these features.
Items with a DOI have an Altmetric 'badge' if statistics are available. This works better for items published after July 2011 - it is likely to underestimate activity of older items.
There are also download statistics for any item with available files. This only includes the version in ePrints Soton, and not other digital versions e.g. publishers websites and other repositories. These can be particularly useful where items are mostly accessed via ePrints Soton (e.g. theses, working papers and datasets).
