The Right Metrics for Generation Open: a guide to getting credit for Open Science

You’re not getting all the credit you should be for your research.

As an early career researcher, you’re likely publishing open access journal articles, sharing your research data and software code on GitHub, posting slides and figures on Slideshare and Figshare, and “opening up” your research in many other ways.

Yet these Open Science products and their impacts (on other scholars, the public, policymakers, and other stakeholders) are rarely mentioned when applying for jobs, tenure and promotion, and grants.

The traditional means of sharing your impact–citation counts–don’t meet the needs of today’s researchers. What you and the rest of Generation Open need is altmetrics.

In this post, I’ll describe what altmetrics are and the types of altmetrics you can expect to receive as someone who practices Open Science. We’ll also cover real life examples of scientists who used altmetrics to get grants and tenure–and how you can do the same.

Altmetrics 101

Altmetrics measure the attention your scholarly work receives online, from a variety of audiences.

As a scientist, you create research data, analyses, research narratives, and scholarly conversations on a daily basis. Altmetrics–measures of use sourced from the social web– can account for the uses of all of these varied output types.

Nearly everything that can be measured online has the potential to be an altmetric indicator. Here are just a few examples of the types of information that can be tracked for research articles alone:




faculty of 1000

popular press


traditional  citation



scholarly blogs

blogs, twitter


mendeley, citeulike



pdf views

html views


When you add research software, data, slides, posters, and other scholarly outputs to the equation, the list of metrics you can use to understand the reception to your work grows exponentially.

And altmetrics can also help you understand the interest in your work from those both inside and outside of the Ivory Tower. For example, what are members of the public saying about your climate change research? How has it affected the decisions and debates among policy makers? Has it led to the adoption of new technologies in the private sector?

The days when your research only mattered to other academics are gone. And with them also goes the idea that there’s only one type of impact.

Flavors of impact

There are many flavors of impact that altmetrics can illuminate for you, beyond the traditional scholarly impact that’s measured by citations.

This 2012 study was the first to showcase the concept of flavors of impact via altmetrics. These flavors are found by examining the correlations between different altmetric indicators; how does a Mendeley bookmark correlate to a citation, or to a Facebook share? (And so on.) What can groups of correlations tell us about the uses of scholarship?

Among the flavors the researchers identified were a “popular hit” flavor (where scholarship is highly tweeted and shared on Facebook, but not seen much on scholarly sites like Mendeley or in citations) and an “expert pick” flavor (evidenced by F1000 Prime ratings and later citations, but few social shares or mentions). Lutz Bornmann’s 2014 study built upon that work, documenting that articles that are tagged on F1000 Prime as being “good for teaching” had more shares on Twitter–uncovering possible uses among educational audiences.

The correlation that’s on everyone’s mind? How do social media (and other indicators) correlate with citations? Mendeley bookmarks are found to have the most correlations with citations; this points to Mendeley’s use as a leading indicator (that is, if something is bookmarked on Mendeley today, it’s got better chance of being cited down the road than something that’s not bookmarked).

Correlations with citations aren’t the only correlations we should pay attention to, though. They only tell one part of an impact story–an important part, to be sure, but not the only part.

Altmetrics data includes qualitative data, too

Many don’t realize that altmetrics data isn’t only about the numbers. An important function of altmetrics aggregators like and Impactstory (which we describe in more detail below) is to gather qualitative data from across the web into a single place, making it easy to read exactly what others are saying about your scholarship. does this by including snippets of the blogs, tweets, and other mentions your work receives online. Impactstory links out to the data providers themselves, allowing you to more easily find and read the full-length mentions from across the web.

Altmetrics for Open Science

Now that you have an understanding of how altmetrics work in general, let’s talk about how they work for you as an Open Scientist. Below, we’ve listed some of the basic metrics you can expect to see on the scholarship that you make Open Access. We’ll discuss how to find these metrics in the next section.

Metrics for all products

Any scholarly object that’s got a URL or other permanent identifier like a DOI–which, if you’re practicing Open Science, would be all of them–can be shared and discussed online.

So, for any of your scholarly outputs that have been discussed online, you can expect to find Twitter mentions, blog posts and blog comments, Facebook and Google+ shares and comments, mainstream media mentions, and Wikipedia mentions.

Open Access Publications

Your open access publications will likely accrue citations same as your publications that appear in subscription journals, with two key differences: you can track citations to work that isn’t formally published (but has instead been shared on a preprint server like ArXiv or other such repository) and you can track citations to work that appear in non-peer reviewed literature. Citation indices like Scopus and Web of Science can help you track the former. Google Scholar is a good way to find citations in the non-peer reviewed literature.

Views and downloads can be found on some journal websites, and often on repositories–whether your university’s institutional repository, a subject repository like BioRXiv, or a general purpose repository like Figshare.

Screen Shot 2014-10-22 at 4.16.36 PM.png

Bookmarks on reference management services like Mendeley and CiteULike can give you a sense of how widely your work is being read, and by what audiences. Mendeley, in particular, offers excellent demographic information for publications bookmarked in the service.

Software & code

Software & code, like other non-paper scholarly products, are often shared on specialized platforms. On these platforms, the type of metrics your work receives is often linked to the platform itself.

SourceForge blazed the trail for data metrics by allowing others to review and rate code–useful, crowd-sourced quality indicators.

On GitHub, you can expect for your work to receive forks (which signal adaptations of your code), stars (a bookmark or virtual fistbump that lets others tell you, “I like this”), pull requests (which can get at others’ engagement with your work, as well as the degree to which you tend to collaborate), and downloads (which may signal software installations or code use). One big advantage to using GitHub to share your code is that it allows you to mint DOIs–making it much easier to track mentions and shares of your code in the scholarly literature and across general purpose platforms, like those outlined above.


Data is often cited in one of two ways: citations to data packages (the dataset itself, stored on a website or repository) and citations to data papers (publications that describe the dataset in detail, and that link out to the dataset). You can often track the former using an altmetrics aggregator (more on that in a moment) or the Data Citation Index, a database that’s similar to Web of Science which searches for mentions of your dataset in the scholarly literature. Citations to data papers can sometimes be found in traditional citation indices like Scopus and Web of Science.

Interest in datasets can also be measured by tracking views and downloads. Often, these metrics are shared on repositories where datasets are stored.

Where data is shared on GitHub, forks and stars (described above) can give an indication of that data’s reuse.

More info on metrics for data can be found on my post for the e-Science Portal Blog, “Tracking the Impacts of Data–Beyond Citations”.


Videos are created by many researchers to summarize a study for generalist audiences. Other times, videos are a type of data.

YouTube tracks the most varied metrics: views, likes, dislikes, and comments are all reported. On Vimeo and other video sharing sites, likes and views are the most often reported metrics.

Slide decks & posters

Slide decks and posters are among the scholarly outputs that get the least amount of love. Once you’ve returned from your conference, you tend to shelve and forget about the poster that you (or your grad students) have put hours worth of work into–and the same goes for the slide decks you use when presenting.

If you make these “forgotten” products available online, on the other hand, you can expect to see some of the following indicators of interest in your work: views, favorites (sometimes used as a bookmark, other times as a way of saying “job well done!”), downloads, comments, and embeds (which can show you how often–and by whom–your work is being shared and in some cases blogged about).

How to collect your metrics from across the Web

We just covered a heck of a lot of metrics, huh? Luckily, altmetrics aggregators are designed to collect these far-flung data points from across the web and deliver them to you in a single report.

There are three main independent altmetrics aggregators:, PlumX, and Here’s the scoop:

  • we’re a non-profit altmetrics service that collects metrics for all scholarly outputs. Impactstory profiles are designed to meet the needs of individual scientists. We regularly introduce new features based on user demand. You can sign up for a 30-day free trial on our website; after that, subscriptions are $10/month or $60/year.

  • PlumX: a commercial service that is designed to meet the needs of administrators and funding agencies. Like Impactstory, PlumX also collects metrics for all scholarly outputs. PlumX boasts the largest data coverage of all altmetrics aggregators.

  • a commercial service that collects metrics primarily for publishers and institutions. Altmetric can track any scholarly output with a DOI, PubMed ID, ArXiv ID, or Handle, but it does publications the best. Uniquely, they can find mentions to your scholarship in the mainstream media mentions and policy documents–two notoriously hard to mine locations.

Once you’ve collected your metrics from across the web, what do you do with them? We suggest experimenting with using them in your CV, year-end reporting, grant applications, and even tenure & promotion dossiers.

Skeptical? You needn’t be. An increasing number of scientists are using altmetrics for these purposes.

Researchers who have used altmetrics for tenure & grants

Each of the following researchers used altmetrics, alongside traditional metrics like citation counts and journal impact factors, to document the impact of their work.

Tenure: Dr. Steven Roberts, University of Washington

Steven-Roberts1-528x528.jpgSteven is an Associate Professor in the School of Aquatic & Fishery Sciences at the University of Washington. He decided to use altmetrics data in his tenure dossier to two ends: to showcase his public engagement and to document interest in his work.

To showcase public engagement, Steven included this table in the Education and Outreach section of his dossier, illustrating the effects his various outreach channels (blog, Facebook, Flickr, etc) have had to date:

Screen Shot 2014-10-20 at 2.19.52 PM.png

For evidence of the impact of specific products, he incorporated metrics into his CV like this:

Screen Shot 2014-10-20 at 2.24.04 PM.png

Screen Shot 2014-10-20 at 2.25.35 PM.png

Steven’s bid for tenure was successful.

Want to see more? You can download Steven’s full tenure dossier here.

Tenure: Dr. Ahmed Moustafa, American University in Cairo

ahmed.jpgAhmed’s an Associate Professor in the Department of Biology at American University in Cairo, Egypt.

He used altmetrics data in his tenure dossier in two interesting ways. First, he included a screenshot of his most important scholarly products, as they appear on his Impactstory profile, to summarize the overall impacts of his work:

Screen Shot 2014-10-20 at 2.52.15 PM.png

Note the badges that summarize in a glance the relative impacts of his work among both the public and other scholars. Ahmed also includes a link to his full profile, so his reviewers can drill down into the impact details of all his works, and also review them for themselves.

Ahmed also showcased the impact of a particular software package he created, JAligner, by including a link to a Google Scholar search that showcases all the scholarship that cites his software:

As of August 2013, JAligner has been cited in more than 150 publications, including journal articles, books, and patents, ( covering a wide range of topics in biomedical and computational research areas and downloaded almost 20,000 times (Figure 6). It is somehow noteworthy that JAligner has claimed its own Wikipedia entry (!

Ahmed received tenure with AUC in 2013.

Grant Reporting: Dr. Holly Bik, University of Birmingham

0167.pngHolly was awarded a major grant from the Alfred P. Sloan Foundation to develop a bioinformatics data visualization tool called Phinch.

When reporting back to Sloan on the success of her project, she included metrics like the Figshare views that related posters and talks received, Github statistics for the Phinch software, and other altmetrics related to the varied outputs that the project created over the last few years.

Holly’s hopeful that these metrics, in addition to the traditional metrics she’s reported to Sloan, will make a great case for renewal funding, so they can continue their work on Phinch.

Will altmetrics work for you?

The remarkable thing about each of these researchers is that their circumstances aren’t extraordinary. The organizations they work for and receive funding from are fairly traditional ones. It follows that you, too, may be able to use altmetrics to document the impacts of your Open Science, no matter where you work or are applying for funding. After all, more and more institutions are starting to incorporate recognition of non-traditional scholarship into their tenure & promotion guidelines. You’ll need non-traditional ways like altmetrics to showcase the impacts of that scholarship.

7 thoughts on “The Right Metrics for Generation Open: a guide to getting credit for Open Science

  1. I see that ‘Twitter’ has been categorized under Public and I’m afraid that 75% of the tweets are from academic persons. Twitter data from very well confirms this, i suppose. Henceforth, I personally believe that Altmetrics tracks communication among informal channels and Altmetric data can not be categorized as Scholarly Vs Public, as the involvement of public in scholarly content is negligible.

    • Agreed! The breakdown that we tend to use between what’s a “public” data source vs an “academic” one can be very blurry, since so many academics use Twitter and blogs to discuss their work. The table might be better imagined as a series of Venn diagrams, since there’s such overlap. 🙂

      I’m unfamiliar with the 75% figure you give. I’d love to read the study it’s from, if you have the link handy!


  2. Well, it was a mistake. I think the phrase ‘most of the tweets’ is what I meant to convey. There is no specific study published yet to back up this statement.

    But, to my observation, the rate of tweets by public seem always lesser compared to the tweets by other academicians. The data in the below webpage link will give you a clear view of what I’m talking about.

Leave a Reply