By Blair Granville, INSIGHT ANALYST
It’s World Statistics Day, a five-yearly celebration of data, stats and numbers that mean something, and a day where a number of different statistics-related activities and initiatives commence (eg see those of the Royal Statistical Society calendar).
Of course these days statistics hardly needs a special day, as it is more present than ever, what with the rise of so-called “Big Data”, and an increasing urge for business and organisations to take advantage of the wealth of digital information now available. Anyone can appreciate the importance of stats in science, and the Biochemical Society has previously run successful courses such as June’s “R for Biochemists” Training Day. Statistics is a key focus for our new training programs, such as our “Quantitative Proteomics and Data Analysis” course being run in April 2016.
But there are other ways the Society uses data and stats. The publishing wing of the Biochemical Society, Portland Press, regularly relies on such info to help track the performance of our journals including the Biochemical Journal, Clinical Science and Bioscience Reports – for example, to track our progress at decreasing the time taken between the acceptance and publication of an academic article in one of the journals.
The above chart shows how the Acceptance to Publication time for the Portland Press journals has decreased dramatically in 2015 due to new initiatives to speed up turnaround time.
We also use some tools that can allow anyone to get involved in data analysis. The below network diagrams show relationships between terms used in the titles and abstracts of articles we publish, mapping out the domain of biochemistry and molecular biology covered in our journals.
Density map of terms used in the abstracts and titles of articles across Portland Press journals January 2014 to June 2015, created using VOSviewer. The redder colours are the more commonly occurring terms. Words are clustered according to how closely they occur together in the text.
Network map of terms used in the abstracts and titles of articles across Portland Press journals January 2014 to June 2015. Words are clustered according to how closely they occur together in the text, and the colours indicate clusters which commonly occur together.
The data in the above network maps was extracted from PubMed, a freely accessible database of academic journals, and was processed with the excellent and also free to use VOSviewer software developed by researchers Nees Jan van Eck and Ludo Waltman at Leiden University’s Centre for Science and Technology Studies (CWTS). The presence of these kinds of free tools and databases means anyone can start to teach themselves how to use statistics and data to solve real-world problems – for example solving protein structures via the foldit website or perhaps analysing your own genome at home.
There is no doubt data will continue to change the opportunities we have in science, publishing and the world at large at an ever more rapid pace. I wonder if anyone is putting together some statistics on the success of this year’s World Statistics Day; with the growing impact of data on our everyday lives, maybe it’s time to make this event an annual feature!