A bit about Data Visualisation

A few years ago I came across the term “Data Visualisation” which I supposed was something very complex that turned data into those wonderful pictures of aeroplane flight paths in many colours on a map of the world.
Related image

Or, I marvelled at huge pictures of digital art, like the one below, Courtesy of Oak Ridge National Laboratory, U.S. Dept. of Energy.

ORNL EVEREST visualization

I longed to be able to learn to do something just as bold and beautiful, and started finding out about the process. I didn’t realise that although these impressive images are data visualisation, there is a much simpler and far more prosaic method of visualisation data.

Quite simply it is charts and graphs – things that I have known how to do since secondary school and actually taught to quite young children in my days of being an infant teacher. This may be simpler than the digital art, but it still takes data (the height of children, for example) and communicates the information of the tallest or shortest person easily to anyone who cannot see the children standing together.

chart

The example above is done in Google Sheets – other tools are available. In my teaching days it would have been made more interesting by doing a pictograph – little pictures of people standing in a line, something like this one:

pictograph
Historically, information and data that has been communicated through a picture, or visualisation, has changed the world. For example, Florence Nightingale was actually a very astute statistician who gathered data about army deaths and their causes. She compiled the diagram below to show how many of these deaths were preventable. I particularly like the detailed explanation of how to read the visualisation

Diagram of the causes of mortality in the army Wellcome L0041105

I also recently come across a rather wonderful visualisation of Napoleon’s March across Russia which is a very poignant visualisation of the human losses incurred at war. The thick beige line dwindles in size as soldiers die on the journey. The returning troops are shown by the black line which also gets thinner and thinner. In all around 410,000 men were lost. I believe that the visualisation demonstrates the enormity of that loss in a more powerful way that looking at the figures.

Minard's Map (vectorized)

Comparing these simpler pictures with the digital art that I showed at the beginning of this blog, I think that the simplicity tells the story and communicates the meaning of the data in a much better way. So there is really no need to use very complex algorithms or expensive data visualisation tools to produce a picture of your data. Google Sheets and Excel have the wherewithal to produce a clear chart or graph.

At this point, I have to give a plug to JUSP and the technical team behind the resource. They have produced visualisations of JUSP data using “Tableau” so that e-resource librarians can download them and use them for analysis or presentations. So far, these visualisations are proving popular with JUSP users.  We at Evidence Base will be monitoring the use of these visualisations over the forthcoming weeks and will find out how useful and powerful it will be to put journal usage data into a visual form.

Advertisements

Working with IRUS-UK

Evidence Base works with IRUS-UK, which is a Jisc resource for Higher Education Institutions (HEIs) such as Universities, Colleges and other Institutions that offer higher qualifications, such as HNC, HND, and graduate and post graduate degrees.

Flag Iris in Victorian garden Quex House Birchington Kent England.jpg
By AcabashiOwn work, CC BY-SA 4.0, Link

I said IRUS, not iris! Most of the UK Universities keep an Institutional Repository: an electronic archive of the work produced in the university, such as theses, research reports, data, presentations, images and articles written by staff and students. Many of these Institutional Repositories are freely accessible to anyone who wants to read the work.

As you can imagine, the Institutional Repository managers are keen to find out what work has been accessed. We are in age of Metrics where counting everything is important because use and usefulness has to be justified and value for money demonstrated. But then, as a taxpayer, you will certainly want your money to be spent wisely. IRUS – UK  can be used by Institutional Repository managers to find out what items have been downloaded from their repository. IRUS-UK sorts the downloads into lists: including type of item, title and author. Sometimes it is not so easy to identify a specific author, people use different variations of their names. For  example I could be M. Bamkin, M.R. Bamkin, Marianne Bamkin, Marianne Ruth Bamkin, or even decide to use my pre-marital surname – King, so that makes another four variations. I could be listed as eight different people, and therefore downloads of my work could indicate that it comes from eight individuals, not one. With a name like Bamkin, is it not so hard to work out that M. Bamkin is probably the same author as Marianne Bamkin, but just think about J. Smith. Is that John, Jane, James, Jamina, and if so, which one?

Now, I expect you are wondering why I am going on about different author names when I started talking about Institutional Repositories and IRUS-UK. Well, there is a link – ORCID identifiers.

I said ORCID, not orchid! ORCID is one of those names that started as an acronym but which everyone forgets, so is now a word for the electronic identifier that can be used on all authored work to show that you are one person, whatever you decide to call yourself on that day for that piece of work.

IRUS-UK is working hard to incorporate ORCIDs into the system and has written a blog post about what Institutional Repositories can do to expose ORCIDs. This means that IRUS-UK can have a searchable list of authors with their individual identifiers, so you can be sure that you have found the right J. Smith.