altmetrics.org

This is version 0 of an abstract to be presented at altmetrics11.

Hadas Shema
Judit Bar-Ilan
Department of Information Science, Bar-Ilan University, Israel

Introduction

Researchblogging.org (2008) is an aggregator of science blogs. Bloggers discussing peer-reviewed research can register with the aggregator, and when they mark relevant posts in their blog, these posts appear on the aggregator’s site, allowing one-stop access to research reviews to interested readers. The site’s editors ensure that posts follow the guidelines and are of appropriate quality. In this study, our objective was to learn about these bloggers and the type of research they choose to review in order to get insights into scientific blogging in general.

Method

Following Groth and Gurney (2010) we chose to use an aggregator of science blogs called ResearchBlogging.org (RB for short) for characterizing science blogs and bloggers. Blogs chosen for the study were non-commercial blogs written by 1-2 individuals and had a minimum of twenty entries posted at the RB aggregator between January 1, 2010 and January 15, 2011. There were 133 bloggers in 126 blogs satisfying our criteria (two bloggers had two blogs each and 9 blogs had two authors each). We collected the data from the blogs and bloggers’ RB pages as well as the “About” and “Profile” parts of the blogs themselves. In cases the “About” or “Profile” parts were unclear we searched the Internet for mentions of the blogger’s name in different contexts. The publicly available parts of profiles from LinkedIn, Facebook, Twitter and other social networks were used as additional sources of information on the bloggers. The blog post characterization was based on the bloggers’ last five posts appearing on RB at the time of the data collection in March, 2011.

Gender

In the academic year 2008-2009 women earned 50.4% of the doctorates in the U.S. Council of Graduate Schools, 2010). Despite the large percentage of doctorates earned by women, men dominate the science blogging arena (Fig. 1). About two-third of the blogs have one male authors, 18.25% have one female author, 4.76% have two male authors and almost 4% have one female and one male author. The gender of 5.56% of the writers of blogs is unknown. The gender disparities in science blogs authorship seem similar to those of Wikipedia contributors: 12.64% of the contributors to Wikipedia are women (Glott, Schmidt & Ghosh, 2010). Our findings are in line with those of Munger (2010) who studied the general gender ratio of RB and found that “male bloggers outnumber female bloggers by over three to one.”

Figure 1: Gender distribution of sampled RB bloggers:

Anonymity

Bloggers who do not supply a name or only supply a nickname/first name are referred to as anonymous. It must be noted that we have not made inquiries about the authenticity of each name, so it is possible that names which appear to be authentic may be pseudonyms. Most bloggers choose to blog under their full name. Out of the 133 bloggers in our sample only 24 (18.05%) blog anonymously.

Blog Networks

A scientific blog can be an independent venture, or be a part of a larger group of science blogs. There have been several launches of science blog networks by traditional media outlets. For example, the British newspaper the Guardian launched its own science blogs network in August 2010 Sneyd, 2010) and Wired magazine followed suit (Mason, 2010). In our sample 87 (69.05%) were independent blogs and 39 (30.95%) were part of a bigger group of blogs. Out of the 39 blogs, 15 (38.46%) belong to one of the three networks ran by Seed Magazine (in English, German and Portuguese).

Language

English is the dominant language of scientific blogs. Out of the 126 blogs in the Sample 108 (85.71%) were written in English (Fig. 2)

Figure 2: Language distribution of the sampled blogs:

Journals reviewed The references appearing in the last five blog posts in each of the 126 blogs were extracted. This resulted in 913 references to articles appearing in 429 journals, 9 references to articles uploaded to arxiv.org, 3 references to conference proceedings and 2 references to books. The distribution of the number of times journals were referenced appears in Table 1. The journals that were referenced ten times or more, were Science (61 times), Nature (53 times), PNAS (42 times), PLoS ONE (37 times), Psychological Science (16 times), Proceedings of the Royal Society B (15 times), BMJ, JAMA and Physical Review Letters (10 times each).

Table 1: Number of times journals were referenced in the sample:

Table 2: Subject categories:

Subject categories

For each of the journals that was referenced twice or more we identified the JCR subject category/categories they belong to (601 articles). Only 7 journals were not in ISI’s JCR for 2009. Based on the JCR journal categorization, the articles were classified into seven major classes (see Table 2). In a few cases the journal was categorized into more than one major category.

Conclusions

Science blogs have some of the characteristics of traditional scientific discourse. Science bloggers show a preference for high-impact, multidisciplinary journals (Science, Nature, PNAS and PLoS ONE). This preference is similar to that found by Groth and Gurney (2010) for RB chemistry blogs. Nevertheless, a wide variety of journals were discussed, with almost three quarters of journals only being referenced once in the time period. About 85% of the blogs are written in English, the language of most scientific discourse. Although bloggers can choose to post anonymously, more than four-fifths of them apparently use their own name. We have confirmed Munger’s (2010) findings about the gender gap in science blogging. Our study was limited by our sample (blogs from RB that had over 20 posts recorded in the aggregator) and by the number of references from these blogs that were analyzed.

References

Council of Graduate Schools (2010). Graduate Enrollment Continues Strong Growth in 2009. www.cgsnet.org/portals/0/pdf/N_pr_ED2009.pdf

Glott, R., Schmidt, P. & Ghosh, R. (2010). Wikipedia survey – Overview of results www.wikipediasurvey.org/docs/Wikipedia_ Overview_15March2010-FINAL.pdf

Groth, P. & Gurney, T. (2010) Studying Scientific Discourse on the Web using Bibliometrics: A Chemistry Blogging Case Study. In: Proceedings of the WebSci10., Raleigh, NC,US. journal.webscience.org/308/

Mason, B. (September 14th, 2010). Meet the New Wired Science All-Star Bloggers. www.wired.com/wiredscience/2010/09/new-wired-science-blogs-network/

Munger, D. (September 22, 2010). Blogging out of balance. Seed Magazine. seedmagazine.com/content/article/blogging_out_of_balance/

Researchblogging.org (2008). About us. www.researchblogging.org/static/index/page/about

Sneyd, E. (August 31, 2010). The Guardian launches science blogs network blogs.journalism.co.uk/editors/2010/08/31/the- guardian-launches-science-blogs-network/

• 

3 Comments

1. Brian Kelly
   Posted May 12, 2011 at 11:33 am | Permalink

   Broken link in reference to Mason, B. (September 14th, 2010). Meet the New Wired Science All-Star Bloggers.(Spurious space in URL),

2. dassysh
   Posted May 20, 2011 at 8:37 pm | Permalink

   Fixed, thank you!

3. parra
   Posted June 14, 2011 at 2:09 pm | Permalink

   What about an study on how much these blogs influence researchers in their work? I for once, get a lot of my ideas from blogs that later try to find in peer reviewed publications.

   Furthermore, I think one particular use of research blogs is to keep track of fields that are not part of your work, but include topics that are interested to know at least superficially. Again, in my case, even though I’m not an astrophysics, i like it and follow it mainly through blogs.

   It would be interesting to extend this work and study not only characteristics of the academic blogs, but also their impact, influence, usage and audience.

Post a Comment