University College, London’s Centre for Information Behaviour and the Evaluation of Research (CIBER), recently released a commissioned report on Social Media and Research Workflow. The study is based around a survey of active users of social media and provides an informative picture of the technologies they are using. Though the study includes a wide definition of “social media,” the results are much in line with previous analyses: technologies that create new efficiencies are being picked up, while those without immediate and obvious benefit are being ignored.
As Phil Davis recently pointed out, researchers performing studies like this need to be keenly aware of the biases introduced by the methods used. The researchers at CIBER do a nice job of being very upfront about their methods, and what they purport to show and what they clearly do not show. By using an online survey, essentially a convenience sample, they openly admit that the results are subject to sampling bias. As such, the study does not provide an accurate picture of the research community as a whole, but instead looks at the behaviors of a self-identified group of social media enthusiasts willing to participate in such a survey.
While there’s certainly useful information to be gleaned here, there’s no way to discern whether those surveyed are early adopters with behaviors predictive of future trends by the mainstream, or just outliers who enjoy experimenting with technology. Nor is there any way to gauge what percentage of the overall community they represent. The actual tools included under the umbrella of “social media” further muddies the waters.
Not very many of these self-identified social media users are actively employing the sorts of tools that we usually think about as “social media”:
- 27% use social networking
- 23.2% image or video sharing
- 14.6% blogging
- 9.2% microblogging
- 8.9% social tagging or bookmarking
Even among the enthusiasts, there doesn’t seem to be much enthusiasm for the types of user-generated content prevalent in Web 2.0. Most surveyed only take advantage of one or two types of tools, very few using more than that.
Larger percentages of users start to turn up when one includes things like Google Docs for collaborative authoring (62.7%), Google Calendar for setting up meetings (41%), and Skype for communicating (48.3%).
I suppose each of these tools can be considered “social” in its own way, but none really fit the sort of “open” and “networked” or “crowdsourced” definitions that we often think of when discussing the current information revolution. Each is, in it’s own way, merely a new technology enhancing the functionality of a previous tool rather than creating a new functionality. Google Docs is a more efficient system than emailing drafts back and forth, Google Calendar is faster than having your secretary call my secretary, and Skype is a cheaper way to make conference calls than using traditional phone companies.
If closed means of communicating between defined individuals like this are “social media,” then shouldn’t we consider e-mail to be “social media” as well? What about the shared calendars in Microsoft Outlook? Is the telephone a “social media” tool?
In a particularly confusing section, CIBER divides respondents up by field of study. They contrast behaviors seen between researchers in the “Biological Sciences” and researchers in the “Life Sciences.” What exactly is the difference? Then they throw in “Neuroscience” as a separate field. Isn’t that a form of studying biology or life? “Health Sciences” is also separated out on its own, as is “Pharmacology and Toxicology”. Shouldn’t the latter fall under the former? No definitions for these fields are given so behaviors by field are difficult to interpret.
Those issues aside, the results fall fairly well in line with previous studies. Even among those immersed in using social media, only a small minority are using newly developed tools, and traditional channels for research dissemination are greatly favored over informal channels.
Researchers continue to back dissemination routes that they know and trust. It is clear that social media users see informal tools as a complement to the existing system of scholarly publishing, not as a replacement.
The majority of “social media” use shown here is not leading to new ways of communicating and interacting, it is instead being used to increase efficiency and cut costs involved in accomplishing the things scientists already needed to accomplish. For those paying attention, that’s long been an important message — create tools to fit the needs of your community rather than expecting the community to change its behavior to benefit your shiny new online business.
Researchers are a time-pressed community, and tools that are seeing adoption are tools that create efficiencies, rather than timesinks. That’s the simple answer to questions like the one posed by this science blogger, who wants to know why scientists are so slow in picking up things like social networking or microbloggging. Scientists rapidly adopt new technologies like e-mail when they show immediate and obvious benefits that result in increased efficiency. No such benefits are obvious for the time investment required by tweeting or blogging or online social networking.
Sadly, the actual results of the study seem to be largely ignored in the press coverage in the Chronicle of Higher Education, which presents the findings as representing mainstream researchers rather than a small biased subset and lumps in tools seeing wider use (Skype) with those largely unused (Twitter).
Other results of the CIBER study worth mentioning:
- Mainstream social tools are favored over specialized tools developed specifically for researchers. “Facebook for scientists” once again proves itself to be plain old Facebook.
- Age is a poor predictor of social media use in a research context (adding further data dispelling the “digital native” myth). Younger researchers in the study do favor the use of microblogging, social tagging and bookmarking over older researchers (“over 35’s”). The study suggests this may be due to the different roles played by more senior academics, which is a good point. I’d also add in the increasing time pressure as one’s career advances–a first year graduate student has a lot more time to waste on Twitter than a professor actively seeking tenure.
8 Thoughts on "Researchers And Social Media: Uptake Increases When Obvious Benefits Result"
Regarding email, the listserve is widely used in science and is the principal competitor of the blog. The listsrve has the advantage of a kind of privacy. It is this privacy aspect of science that the Web 2.0 people have tended to ignore. We clearly need to understand this feature better. Is it a matter of control, or of community?
I’m still on the e-mailing list of the lab where I did my postdoc more than a decade ago, and it’s still the main method of communication between lab members (other than face to face). This, despite the efforts of a young go-getter to set up a lab wiki and network which was a non-starter.
While the question of privacy is of extreme importance (researchers need what they consider a trusted and safe environment to speak freely for a variety of reasons), it goes beyond that. The mailing list is a very simple, low effort tool that efficiently does exactly what the researcher needs. Contributing to a wiki is a lot more work, more setup time, more training time, etc., all to achieve essentially the same goal.
Within the context of science, I think the phrase “social media” is both distracting and misleading.
Science communication is inherently social as the validation of new truth claims requires results to be disseminated widely and open to the possibility for dialog and debate.
New forms of “social media” are little more than the horseless carriages of traditional science communication. As you rightfully argue, David, we should view these technologies from a functional standpoint (i.e. how do they help scientists communicate) rather than from a novelty standpoint.