Data Visualisation

Software is making it easier to create data visualisations that are interactive and can be shared via the web. For example, we can create maps with ArcGIS and QGIS, representations of networks with Gephi, interactive models of complex systems with NetLogo, 3D renderings of large datasets with Blender, statistical insights with RStudio and Shiny, and script libraries such as D3.js.

In this category of OxTALENT we were looking for visualisations that tell a story, provide an insight, make the complex simple or illustrate a beautiful pattern in a data set.

Winner: Will Allen for ‘Seeing Data: Are Good Big Data Visualisations Possible?’

As a key intermediary in informing public debate about migration in the UK, The Migration Observatory (MigObs) wanted to understand how it could improve its visualisation practices for the benefit of non-experts. Although it already has published some visualisations (such as maps and charts) on its website, MigObs wanted to find out how the public would react to a more comprehensive approach to visualisation: namely, one that was more interactive, immersive, and customisable. To do this, it contributed to the Seeing Data project funded by the AHRC, using international migration as a topical case study for exploring how visualisations of complex data are produced, received, and understood.

The judges picked Will’s entry as the winner because it makes a large dataset available to anyone with an internet connection. The team made an extra effort to work with non-experts to understand how to improve the design of the data visualisations and arrived at five heuristics for engaging the public with data visualizations of large data sets:

  1. consider the aims, rationale, and potential political context of a visualisation in the first place, before decisions about design are made;
  2. link choices of features to visualise with available design options;
  3. build time into the work for developing clear communication among academics and professional visualisers who may come from different backgrounds and speak different ‘languages’;
  4. consider what the intended audiences of the visualisation should gain; and
  5. acknowledge how the topics chosen as well as their visual presentation can lead to judgments of whether to believe the visualisation at all from the perspectives of viewers.

Runner-up: ‘Bodleian Library in Numbers’

BodNumbers - WhileYouWereHere_Readers‘Bodleian in Numbers’ gives visitors an insight into the complexities of the Bodleian Libraries institution in a simple, entertaining and engaging interactive. It provides bite-size visualisations of the huge variety and quantity of tasks undertaken by library staff. Much of this takes place behind closed doors, so this feature uses the Bodleian’s own data to reveal the extensive work of the libraries to the public in a novel way. This forms a part of the Bodleian’s new public engagement strategy, aiming to open access to its collections and work in new ways. The facts and figures range from the number of staff employed by the libraries to the number of printed items they hold, the number of visitors to the Bodleian’s exhibitions in a year and the number of items being scanned for readers.

The judges praised the visualisations for the simple but effective use of pictures and numbers that remind us how fundamental these the Bodleain Libraries – and indeed libraries in general – are to the fabric of academic and public life.

‘Bodleian in Numbers’ team:

  • Bodleian staff:
    • Frankie Wilson, Head of Assessment
    • Matthew Kimberley, Digital Strategy and Portfolio Manager and Samsung Project Manager
    • Jennifer Townshend, Web Content Editor and Samsung Project Officer
  • PAD Video Production Team:
    • Tom Wilkinson
    • Tom Fuller
  • External:
    • Samsung SDS team (design and coding of the feature including animation)
    • One Ltd (advising on design and gamefication of the data)
Posted in OxTALENT | Tagged , | Comments Off on Data Visualisation

Comments are closed.