I am a senior data journalist at the BBC’s Visual Journalism data team.Over the past year, I have been involved in a wide range of projects published on the BBC News website where data has been central to the story. I have worked on projects where data analysis produced the topline as well as where data visualisation played a key part in telling that story to our audiences. Both individually and with colleagues, I have worked on all elements of the storytelling process, from identifying the idea, analysing the data to find the story, to writing the copy and producing the visualisations for the article page.Some of the projects I have produced over the past 12 months year include:1. A detailed analysis of house price change in real terms since the 2007 financial crisis in England and Wales that challenged the traditional narrative on property prices and exposed huge regional disparities. Worked with the ODI Leeds on this project. 2. Uncovering the extent of overseas ownership of properties in England and Wales. 3. Visualising the destruction of Mosul resulting from the devastating occupation by ISIS, highlighting the challenges that residents might find on their return to the city.4. Analysing the results of the June 2017 UK general elections using maps and graphics, providing crucial context to help people understand the story of the election on a day where millions of readers relied on the BBC News website to inform them about the vote.5. A visually rich research-led piece on the potential impact of climate change on the quality of coffee, making a difficult topic like climate change more relevant and relatable to daily life. This project was possible thanks to a close collaboration with academics researching the topic, translating their data into visualisations and case studies.6. In a visually engaging way, illustrating the obstacles that President Trump faces in building a wall between the United States and Mexico.7. A new iteration of the ‘How long would it take you to earn a top footballer’s salary?’ project, an idea that I came up with and produced in 2015. The new version included updated data, visualisations and the addition of female players. The stories I have worked on had as their main aim helping the BBC’s audience, a general audience, to better understand the news, doing complex data analysis but ensuring that what we are presenting the public is understandable and relatable to them. I was also part of the winning team at the GEN Summit Editors Lab Final, with ‘Appy Helper’ (final portfolio link), a prototype for an in-story tool that aims to make it easier for readers to follow complex ongoing news stories. The prototype formed the basis for the in-story conversational ‘chatbots’ in BBC pages over the last few weeks, built by the Visual Journalism team and BBC News Labs, helping readers better understand complicated and long-running issues in a more conversational tone and personalised format. Here is an example of a story where one is included: http://www.bbc.co.uk/news/health-42737326
What makes this project innovative?
My projects demonstrate a diversity of new ideas, skills, technical ability and thinking outside the box, however with the user’s experience at their core.What they have in common is thorough data analysis with solid statistical foundations, told in an engaging way with a strong public interest theme. Where possible, the stories I worked on include ‘personal relevance calculators’, so that users can easily identify what is significant to them, for example how house prices have changed in their neighbourhoods as opposed to national or regional overview figures. In this way we are able to turn a single story into potentially hundreds - or even thousands - of stories more relevant to each person based on where they live, their salary, or other characteristics specific to them. The deep data dive into house price recovery involved analysing 8.5 million transactions, matching them to detailed local geographies, adjusting for inflation based on the specific month the transaction took place and removing areas with too few transactions to be statistically significant. The original topline that came out of the investigation went against the established media narrative on house prices, which is often one of inevitable rise. Our analysis revealed that the reality for the vast majority of homeowners, particularly in the north of England, is just the opposite, once inflation is taken into account. The ‘6 things that could topple Donald Trump’s wall’ project uses animations on scroll to tell the story in a visually engaging and creative way. The animations help those lacking detailed knowledge of the geography of the US and Mexico border position themselves and understand the issues in a more effective way, providing crucial context.The ‘Appy Helper’ project demonstrates that as well as data analysis, visualisation, developing and writing stories, I can identify new and innovative solutions for telling stories online to audiences put off by traditional news coverage.
What was the impact of your project? How did you measure it?
A number of the pieces I have submitted were among the most read stories on the BBC website on the day they were published. They received a higher proportion of their traffic from social media than traditional BBC news stories do, as well as having high engagement.The analyses of overseas property ownership, flight delays and house prices generated several stories for different regional BBC teams, both online, on local radio, local television channels and the national news bulletins. They were also subsequently picked up by other major UK news organisations. Given that most of my analysis is done in R, I also produce filterable R Markdown tables that can then disseminated to regional UK teams (and World Service language services) to find their own stories relevant to their own local audiences.Other stories, for example the longform piece on the obstacles to Trump’s wall, have been used as evergreen explainers for the site. The story has been re-promoted on the BBC home page since its initial publication and also linked to within related articles when the topic has been in the headlines, increasing engagement and recirculation. A number of the stories included in my portfolio were translated into dozens of other languages including, among others, Arabic, Persian, Spanish, Russian, Urdu, Uzbek, Pashto and Hindi. They have been published and promoted on the various BBC World Service language sites and consumed by audiences across the world.
Source and methodology
The sources used in my work vary. Some use open data from sources like the UK’s Office of National Statistics, Land Registry and Civil Aviation Authority. Other stories use data from organisations like the UN and its agencies. Combined with other datasets or analysed in an original and creative way, open data can provide unique insights and angles to tell interesting stories. Other projects I have worked on draw upon peer-reviewed scientific data, election data and census data. In some cases we have collated data ourselves. Even when data is from official sources, it is still reviewed for accuracy. In terms of methodology, I have played a key role in establishing the framework for our team to structure data analysis in a reproducible way. Over the last few months, the code each team member writes is uploaded to our organisation’s Github account, adhering to a specific template. I work predominantly using R projects, with the data stored on a team server so that my analysis can be re-run and re-produced, publishing my results in R Markdown pages. Additionally, each element of the results from our data analysis is reviewed by at least one other team member. Where relevant, my projects include a detailed section on the methodology at the foot of the page (see house prices, footballers wages, flight delays, overseas ownership). Where an explanation relating to the methodology is essential for the audience to understand the story, it is also included within the copy. The methodology behind each project uses different statistical techniques to ensure accuracy. One example is from the footballers wages project, where salary data is adjusted using the World Bank's Purchasing Power Parity (PPP) conversion rate based on country before comparison the footballer’s data to the user in order to reflect variations in the cost of living from one country to another. This is an indication of the level of detail that regularly goes into developing the methodology for each project.
BBC News Visual Journalism Data Team