Marie Segger, Data Journalist from The Economist London

Article by Chen Zihao and Raja Altaf

On 28th September 2018, The Centre for Urban Science and Progress London (CUSP London)
welcomed Marie Segger, Data Journalist from The Economist London. Marie, who undertakes a bridge
role at The Economist’s data team, was invited to give a talk on data storytelling for the students from
the Urban Informatics MSc Programme in which she explained data storytelling in three dimensions,
namely the what, why and how. Discussions followed the speech where CUSP London staff members
and students posed questions to the speaker and further discussed the issues covered.

1. What is data storytelling?

Marie started by expounding the ubiquity of data – essentially all human activities produce data. At the
same time, the amount of data they produce grows rapidly each year, doubling between the year 2012
and 2015. Therefore, it is important to tell stories with these data. Huge amount of data and the fact
that it is ever growing creates potential for generating stories. However, as Marie pointed out, data
journalism in the news industry is still quite new. Pioneers such as The Financial Times has had its data
team for only ten years, but it has now become a trend for newsrooms to have such an establishment.
Even so, a lot of stories to date are still left untold because of the limited amount of resource.
Marie went on to make examples of the use of data storytelling. People who make political decisions
are being told stories by their advisors to facilitate policy making. Consumers are being told stories by
advertisers to help make financial decisions on purchases. There could be a lot more examples that are
yet to be discovered.
Data storytelling, as Marie defined, is simply about finding a narrative for data and conveying it to an
audience. She pointed out a common mistake in this process – people tend to pay too much attention
on numbers that stick out, but doing that does not always make a story, not to mention that often
numbers that look odd are actually errors. Therefore, it is always important to double check data and its
sources. Student Jennie Williams posed a question to the speaker: what comes first, data or story? Marie
personally believes that generally it is data that comes first. However, she did not deny the existence of
other cases, in which people notice things in real life and form hypotheses and then went on looking for
data for support or proof although this sometimes does not work.
There are some typical topics for stories that are data driven. Some of the examples Marie gave include
election coverage, diversity, and crime. A major part of telling stories with data is to analyse data and
potentially visualise it. Data-driven stories can come in many different formats, in which Marie used
several examples from The Economist discussed later in the article. This includes news articles and
long-form journalism, which often come with static or interactive visualisation as seen from figure 1
and figure 2.

2. Why do we tell stories with data?

Data visualisation comes with obvious advantages than traditional plain text alone. People are more
likely to remember visualised data than plain figures. Visualisation also helps people understand
complex issues better. More importantly, readers tend to associate numbers with truth and facts.
Although figures can be manipulated, sources that rely heavily on data tend to be more trustworthy.
Marie noticed that even in such hostile social media environment, people still prefer using figures and
numbers from trusted news sources to win their arguments against another. She pointed out that this is
something that newsrooms should use to their advantage, especially in today’s world where fake news
prevails, making it hard to gain trust of readers and social media audience.
Moreover, in the business world, data is often used to substantiate argument whilst trying to convince
stakeholders.

3. What skills do we need to tell a story with data?

Marie displayed the importance of having various sub-teams within an organisation to collaborate and
deliver projects effectively. The skills required can range from journalism, social media, mapping,
statistics, and even computing such as developing. Thus, from this one must understand that one does
not necessarily require technical skills in order to tell a story with data, one can interpret data from
various viewpoints. As an example, Marie states that in the data team at The Economist, the team
consists of 12 members of which 2 members are developers and the rest divided into visual and data
teams journalists. The team works on ‘producing static graphs and maps and data-driven articles’. Such
interactives include the following graph shown on figure 3.

 

Figure 3: Data showing wage earnings from Bachelor’s Degrees in the United States (The Economist
Presentation CUSP).

The Big Mac Index is another example of some of the work produced by The Economist. It is an
interactive tool which allows users to visually understand the purchasing power parity of several
countries which have a McDonald’s food chain. More interesting, is how the visual produced first
begins on paper through illustrations as can be seen in figure 4 below.

Figure 4: Figure showing the translation of a visualisation plan from paper to digital (The Economist,
2018).

4. How does one begin telling a story with data?

Marie explains that one does not require complex data visualisation software when starting out initially.
Beginning with software such as Microsoft Excel or google sheets is simply what one requires to begin
collecting data and interpreting the data. Through making use of tools on Microsoft Excel such as pivot
tables and using platforms such as GitHub to share the data and gather more data sets from other users
one can hone their data science skills. For data visualisation, Tableau is a very good tool in allowing
for a story to be told with the data as it allows for powerful insights to be made. For those who are not
very technical, websites such as Data Wrapper allow for a variety of pre-set options to be selected that
users can choose depending on the relevance to the data being analysed.

Marie stated that in her department at The Economist, the team mainly works with the software D3. D3
is essentially a JavaScript library which allows for the production of interactive and dynamic data
visualisations in web browsers. This software is very sophisticated and allows for some very exciting
visualisations to be produced making articles accessible to a wider range of audiences.
The future of big data is an exciting one and one can only expect more exciting and innovative methods
to be used by The Economist over the course of the next few months as Marie Segger states that her
team is working on producing a major overhaul to The Economist platform.

Bibliography:
1. Anon, (2018). [online] Available at: https://www.economist.com/graphicdetail/
2018/06/18/every-world-cup-goal-ever-scored) [Accessed 11 Oct. 2018].
2. The Economist. (2018). Fixing the flaws in today’s capitalism. [online] Available at:
https://www.economist.com/open-future/2018/04/16/fixing-the-flaws-in-todays-capitalism
[Accessed 11 Oct. 2018].
3. The Economist Presentation. (2018). Marie Segger.
4. The Economist. (2018). The Big Mac index. [online] Available at:
https://www.economist.com/news/2018/07/11/the-big-mac-index [Accessed 11 Oct. 2018].

 

Figs Marie Segger