You don't need to be signed in to read BMJ Blogs, but you can register here to receive updates about other BMJ products and services via our site.

User behaviour

How to Get More Likes, Comments and Shares on Facebook

20 Jun, 12 | by BMJ

For those looking improve engagement with their Facebook posts, it seems that adding more pictures and speaking in the first person is a good place to start. Social media ‘scientist’ Dan Zarrella has tracked and analysed more than 1.3 million posts from the 10,000 most-liked Facebook pages. He has created an infographic (see below) showing what kind of posts perform best on Facebook in terms of likes, shares and comments.

Photos perform best across the board, followed by text and video, according to the data. News links bring in the lowest number of likes, shares and comments. In opposition to Twitter, posts with a high number of self-referential words such as “I” and “me” get more likes .  “It’s also important to be passionate, not neutral,” which means that both positive and negative posts tend to do well. more…

How does content “go viral” through social networks?

28 May, 12 | by BMJ

You may (or may not) have noticed that having blogged every Friday for over 2 years, there was a distinct lack of activity on the BMJ Web Development blog last week. Was I out enjoying the sunshine? Absolutely not. I was in the office researching the best time to disseminate blog posts on social media networks, of course.


Key trends in the information-seeking behaviour of researchers

14 Jan, 11 | by BMJ

At the HighWire Publisher meeting in California this week, a number of key trends in research behaviour were revealed. The Libraries and Academic Research division of Stanford University has carried out 45 interviews to ascertain the information-seeking behaviours of its own researchers, with some enlightening results.

General patterns of behaviour
Many respondents described using “more automated” alerts to stay up-to-date with new material in their field of study. Some reported reading 5-10 journals per week. However, what is interesting is that users defined ‘reading a journal’ as simply browsing through an electronic table of contents or email alert, rather than scanning it from cover to cover. On these eTOCs, researchers would like to see greater annotation. Ideally, they pointed to the inclusion of ‘take home messages’ for each article, so that they might ascertain the suitability of the article to their needs before clicking through to the abstract.

“I don’t review journals, I search databases”. Researchers admitted to only swatting up on the current literature in their field when writing a specific paper or applying for a grant. Research is therefore often driven by necessity and not habit. “I read for a purpose, always with a goal”. This means that users are often missing out on discovery, browsing and serendipitous findings when working in online environments. Interviewees also felt that online journals were missing thematic connections. They reported that whilst these used to be present in print journals through editorial curation, the same context was not usually available through online journal sites.

Context is key and was enthusiastically advocated as one of the major benefits of researching within an online environment. During a panel session at the conference, one researcher suggested that articles should be viewed as portals to greater information rather than the end product. The respondents would like to see more ‘similar articles’ and recommendations on particular topics, instead of browsing archives by issue or year, which would be an obvious benefit of semantic technology.

Annotation tools
Another panel member at the conference who participated in the study, argued that highlighting and annotating tools are imperative for researchers online and need to be as easy to use as in print. A great deal of time and effort is wasted on keeping track of what a researcher has read rather than exploring the subject matter. This process is multi-level, piecemeal and often involves individual systems with no interoperability. Kindle was offered as a solution to this problem. Not only can users add comments and highlighting to articles and ebooks, but they can then log-in to their account and pull off all of these notes into one central document. The device can also be used to see which sections of text other users are focusing on.

Discovery tools
PubMed was listed as the top discovery tool of choice. The second most popular was Web of Science, followed by Google Scholar (mentioned by a third of respondents), Wikipedia (to obtain initial overviews of new topics) and Google. One interviewee confided that they, “use Google to vacuum around the edges of the carpet”. Most databases only contain formal articles but researchers are sometimes eager to find additional context, and that is where Google comes into play.

Reading patterns
Unsurprisingly, there is a common tendency to print the PDF of an article when reading in-depth. On-screen reading is associated with retrieving snippets of information and following links, but many prefer to read significant chunks of text offline. Abstracts, figures, introductions, conclusions and subheadings were all identified as skimmers’ touchpoints. Supplemental data was regarded as important but inconvenient due to it’s fragmented and inconsistent nature. Some publishers, such as The Journal of Neuroscience, are getting rid of supplemental material completely. Their thinking is that if a figure is essential to the article body, it should embedded within the text and not tagged onto the side as an afterthought.

The researchers in this study were conservative towards change. They were only interested in new concepts that could significantly increase their productivity. HighWire highlighted a number of key areas for journal publishers to focus on:

  • Facilitate selection of what to read
  • Information extraction and skimming
  • Supplemental data management
  • Portable annotation
  • Integrated literature management

Publishers should be working with the tools researchers are using externally. To be in a position to fully achieve this, we should be investing more time in semantic technologies and initiatives such as data mining and linked data. The study is not yet complete so watch this space for updates in the coming year.

BMJ Journals Development blog homepage

BMJ Web Development Blog

Keep abreast of the technological developments being implemented on the BMJ journal websites.

Creative Comms logo