Visualising Data Narratives

Guardian Narrative Data Visualization
Example of narrative data visualization from The Guardian

To say that we are living in The Information Age and that we suffer from information overload has become something of a truism. However, it is true to say that for many of us there is a growing requirement that we make sense and use of complex data. This is partly due to advances in personal and mobile computing power, allied to faster broadband speeds, which mean we are collecting, processing and distributing datasets that are far larger than was previously practical. In addition, there are also new factors such as social networking, social media, the successes of the open data movement and the growth of companies such as Google whose whole business is data. Furthermore, these trends look like they will only continue as personal health monitors, smart energy meters and such like come into common use.

One response to this has been the growth in data journalism in which interactive data visualization techniques are incorporated into the story telling process. These narrative visualizations offer the reader the opportunity to gain insight into the story’s subject through guided exploration of the data, which may be enhanced by enabling the reader to personalise the data to varying degrees, thus making the story more relevant to their own perspective. Many excellent examples of this can be found on the websites of The Guardian and New York Times newspapers. Another response has been the idea of social data analysis using online resources such as ManyEyes, where data is visualised publicly and the resulting insights discussed and shared.

Such developments point to ways in which people can engage with complex data in a creative manner and make positive use of it. They also highlight the influence of the different perspectives that groups or individuals may bring to the original data and how their points of view and expertise can be shared productively. This shows that the purpose of visualising data is not simply to provide an accurate representation of it, but also to act as a platform for argument and discussion, for storytelling and scenario testing. This is creative analysis in which individual narratives can be constructed and compared, and it is likely to play an increasingly important role in making important decisions.

Take for instance the process of designing and launching a new car. This is not simply a matter of engineering improvements in aerodynamics and power to weight ratio followed by a clever advertising campaign, important though these are. It also involves appreciating trends in taxation and petrol prices or the availability of electricity charging points; understanding population migration patterns between country, city and suburbs; tracking changes in family size, demographics and the number of cars per family; and following social attitudes to factors such as safety and status. This is all before the arguably more subjective and emotive elements of styling come into play, reactions to which are also collected, collated and visualised. Each of these is a data story that represents the perspective of one or more domain expert and the success of the project is likely to depend on how well they can be combined.

My work as a PhD research student in the Centre for Creativity in Professional Practice involves developing and studying tools that will help support people as they construct and share data stories that reflect their personal or professional perspective. The aim here is not to promote a situation where all decisions are made by compromise or committee, but to give people the tools with which they can effectively argue their point of view and also to better take on board the implications of issues raised by experts from other domains. This, I hope, will eventually lead to better, more informed decisions and a more effective design process for both products and services.


Published by

Graham Dove

PhD research student in the Centre for Creativity in Professional Practice at City University London

3 thoughts on “Visualising Data Narratives”

  1. I find the Guardian graphics often confusing, especially the use of circles which are very hard to compare visually. I am really quite sceptical about designers getting involved with statistics; it is not clear what expertise they bring to the accurate communication of facts – I can see they might have more to offer in stimulating discussion, but that can also be done in words rather than graphics. Someone (perhaps in a PhD) should apply Tufte’s principles of visual design to common methods used in the media for visualising numbers. I suspect they would not score highly.

    1. I think I would respond by saying that the purpose of visualization is not necessarily the ‘accurate communication of facts’. Facts, as they say, are very much in the eye of the beholder and accuracy is a somewhat loaded term. The idea that one is presenting opinions, telling stories and offering perspectives is part of what interests me. Yes this can be done effectively with words, but to exclude visualization on these grounds is perhaps a similar argument to suggesting that there is no need for Picasso because James Joyce is able to express similar ideas. That said, however, I would also agree that there are many examples of less successful visualization and that trying to get users to do detailed comparison through bubble charts is probably not a good idea.

  2. Thanks for the nice post, Graham! As you know, my master thesis was in the field of InfoVis, and this story of yours again reminded me of the quote by Edward Tufte, whom Clive mentioned in the comment above: “Information consists of differences that make a difference”. 🙂

    I agree that InfoVis should be more and more interactive, aiming not only to present, but also creatively explore large data sets. Recently I see people use infographics to tell their personal professional histories. Some lovely examples of such creative resumes at:

    Good luck with your work! 😉

Comments are closed.