DataScience Training



Data Journalism and Storytelling
Feedback form    |       Play Audio    |   Download:    |   


Data Journalism and Storytelling

Data Journalism

The new role of data Click to read  

Numerical data are playing an increasingly important role in the production and distribution of information.

Data can be used to transform something abstract into something understandable and connectable to everyone. All sectors are adapting to this transformation, including journalism: more and more news organisations (New York Times, Sky News, The Guardian...) are relying on data analysis and visualisation to produce and publish informative and engaging stories.

More than half of all news organizations in the US and Europe now have at least one dedicated data journalist working in their newsrooms.

For the average global citizen, more and more of our everyday lives are impacted by computers and data. In order to hold power to account, journalists should be empowered with the skills and tools to make sense of this data

 

Definition Click to read  

Data journalism is a type of journalism in which the use of data is used to tell complex stories. It is the entire process of deriving meaning from data to develop a story - not only the visual output. A data journalist is someone who uses statistics to facilitate the writing and reporting of news stories to provide insights based on relevant data. A written story that relies on data analysis and interpretation. The key ingredient is asking questions to our data just as if we were interviewing it. Data can be the source of data journalism or the tool used to tell the story, or both. This should not be seen as replacing, but rather as complementing traditional journalism.

Good data journalism helps readers understand and draw conclusions about their world. For example, it can bring scientific discoveries to the forefront of a narrative and make them accessible to readers.

           

 

The new ways of journalism Click to read  

Today, news comes as it happens, from a variety of sources (eyewitnesses, blogs, etc.), so the main focus of journalists shifts from being the first to report to explaining what a certain development might actually mean. Reporters create stories using massive datasets. This reflects the growing importance of numerical data in the production and dissemination of information, the growing connection between journalists and professions like design, computer science, and statistics.

.

Journalists data skills Click to read  

The main skills that a data journalist should have are:

  • Search for news articles from an unlimited number of sources
  • Interpret data
  • The ability to read graphical representations of information from which to derive stories to tell.

Finding a dataset to examine is the first step in every data narrative. The findings section of any publication that you feel delivers an engaging tale is a natural source for scientific writers. Consider if the data offers a compelling story when evaluating a possible dataset for use in a project.

 It takes seasoned journalists with the stamina to look at frequently complicated or uninteresting raw data and uncover the hidden story inside.

 

Good data-driven news’ elements Click to read  

1) Good data: It is necessary to search for high quality data, ensuring that the collection methods, research subjects and any analysis are valid and free from bias.

2) Story from the data: Identify the central narrative line and make sure it is solid and consistent.

3) Storyboard and structure: For the story to be coherent and engaging, it is important to organise the content and determine which to include and which to leave out. It can be helpful to sketch out a map of the structure of the data story, as well as the data, visualisations and written content.

4) Clear narrative line: Care must be taken not to present the data as a set of numbers without sufficiently contextualising it. The role of narrative is crucial in helping the audience make sense of the information and data.

5) Interactivity: "Any sense of animation that you can bring into a piece really helps: you can have the feeling of interacting with the content." (Ronan Hughes, Output Editor of Sky News)

Case study: Goalkeepers Click to read  

Goalkeepers is an annual report on the progress of the 17 goals of the 2030 Agenda for Sustainable Development by Bill and Melinda Gates. See, for example, the 2019 report: "Examining equality: How geography and gender stack the deck for (or against) you".

This report is an example of an excellently presented data story. The solid data and information is embedded within a story featuring a little girl living in southern Chad, a Saharan country. Seeing the child's face helps the reader to better understand what the data presented in the report represents. 

The narrative line is clear and coherent, the story structure is solid, and the reading is made smooth and light thanks to the visualisation techniques and interactivity of the report: moving down while reading the page, images and diagrams appear on their own, zooming in or out according to the paragraph we are reading. 

Data Storytelling

Communicating data Click to read  

Discovering key insights is one skill that requires a set of hard skills related to data analysis. Communicating these insights in a clear and compelling way requires other kinds of soft skills. Both are equally fundamental to deriving value from data.

The ability to effectively communicate insights from a dataset using narratives and visualisations is called Storytelling. It can be used to contextualise data and inspire audience action

From data analysis to data storytelling Click to read  

The purpose of any data collection is to extract value from it. Once analysed, however, it is essential to be able to disseminate this value in order for it to be meaningful.

When narrative is combined with data, it helps explain to the general public what is happening in the data and why a particular insight is important.

In this sense, data storytelling is the tool that enables data analysts to translate information from the 'language of numbers' into a story and narrative that is accessible to users unfamiliar with data science.

 

Key elements Click to read  

According to Brent Dykes, Author of Effective Data Storytelling: How to Drive Change with Data, Narrative, and Visuals—Data Storytelling is a combination of data, visuals, and narrative:

  • Data: Analysis of data using descriptive, diagnostic, predictive and prescriptive analyses can enable understanding the full picture and extract knowledge and insights from the data.
  • Narrative: Storyline is used to effectively communicate the insights gained from the data, the context surrounding them and the recommended actions; it is a key vehicle for conveying information because it enhances our ability to understand.
  • Visualisation: Transforming data into graphs, charts, images or videos allows us to see the data more clearly; they provide snapshots at a glance of the data, but without the context needed to explain why something happened.

When you combine the right images and narrative with the right data, you get a data story that can influence and drive change. Data, when used and analyzed properly, can help clarify rumors and reveal facts.

Here are some guidelines for inserting the key elements:

Crafting effective visuals:

1. Choose the best visualization for your story: Line plot, bar pilot, scatter plot, histogram. 

You can obtain more information in these links:

https://towardsdatascience.com/data-visualization-how-to-choose-the-right-chart-part-2-fb32ed14c7be

https://www.youtube.com/watch?v=csXmVBw8cdo

2. Keep visualizations minimal and avoid clutter

  • Use just enough white space to keep the visualization from looking busy 
  • Remove chart borders when applicable 
  • Remove or minimize grid lines or axes when applicable 
  • Clean up axis labels when applicable 
  • Label data directly (as opposed to using a legend) 
  • Remove data markers when applicable 

Use special effects (bold, underline, italics, shadows) sparingly

3. Use text appropriately:

  • When applicable, label axes and titles for clarity 
  • Label important data points when necessary 
  • Provide useful context around insights within the title or subtitle 
  • Adjust font size when highlighting specific messages within your labels 
  • When applicable, try to answer common audience questions with labels

4. Use colors effectively:

Color is one of the most powerful tools available for emphasizing different aspects of your data visualization.

  • Hue represents the range of possible colors, from red, through orange, green and blue, to purple and back to red. 
  • Chroma is the intensity of the color, from gray to a bright color. 
  • Luminance is the brightness of the color, from black to white.

You can get aditional information in these links:

https://academy.datawrapper.de/article/140-what-to-consider-when-choosing-colors-for-data-visualization

https://www.y42.com/blog/color-rules-data-visualization/

5. Do not mislead with data stories:

  • If you are visualizing times series data, make sure your time horizons are large enough to truly represent the data 
  • If the relative size of each value is important, then ensure that your axes start with zero 
  • Ensure that axes scales are appropriate given the data you re treating 
  • If you are sampling data for descriptive purposes, make sure the sample is representative of the particular case. A representative sample is a subset of a population that seeks to accurately reflect the characteristics of the larger group. For example, a classroom of 30 students with 15 males and 15 females could generate a representative sample that might include six students: three males and three females. Samples are useful in statistical analysis when population sizes are large because they contain smaller, manageable versions of the larger group.
  • Use centrality measures such as mean or median to provide context around your data 

Crafting effective narratives with data:

  1. Know the audience: To communicate effectively, you need to know who your audience is, and what their priorities are. There is a range of possible audiences you may encounter when presenting, and crafting an audience-specific message will be important.
  •  Executive: Basic Data literacy skills. Prioritizes outcomes & decisions. Cares much more about business impact than a 1% incremental gain in a machine learning model accuracy or a new technique you are using. (Here is where the general public is located) 
  • Data leader: Data expert. Prioritizes rigor & insights. Cares much more about how your arrived at your insights and to battle test them for rigor 
  • Business Partner: Advanced data literacy skills. Prioritizes tactical next steps. Cares much more about how your analysis impacts their workflow, and what should be their main takeaway from the data story. 

Considerations when crafting audience specific messaging

Aspect

What do you need to consider?

Prior knowledge

  • What context do they have about the problem?
  • What is their level of data literacy?

Priorities

  • What does the audience care about?
  • How does your message relate to their goals?
  • Who is driving decision-making within your audience?

Constraints

  • What is the audiences preferred format?
  • How much time does an audience have to consume a data story?

 

  1. Choose the best medium to share your story

 

Presentations

  • Ensure the length of your presentation is appropriate
  • Leave any highly technical details to the appendix
  • Ensure there is a narrative arc to your presentation
 

Long-form report

  • Be extra diligent about providing useful context around data visualizations and insights
  • Leave any highly technical details to the appendix 
 

Notebook

  • Ensure that you provide useful context on how you arrived at a certain conclusion 
 

Dashboard

 

  • Make use of the dashboard grid layout
  • Organize data insights from left to right, top to bottom
  • Provide useful summary text of key visualization in your dashboard
 

 

Data story

Why is story more effective? Click to read  

The brain's preference for stories over pure data stems from the fact that the brain receives a large amount of information every day and has to determine what is important to process and remember and what can be discarded.

When listening to a story, several parts of the brain are involved, including:

• Wernicke's area, which controls language comprehension;
• Amygdala, which processes emotional response;
• Mirror neurons, which play a role in empathy with others.

When more brain areas are involved, the hippocampus, which stores short-term memories, is more likely to convert the experience of hearing a story into a long-term memory.

The power of a story Click to read  

Information obtained through data analysis, although logical and clearly reported, does not have the power to influence decisions and push the public to act. 

For example, business decisions are thought to be based solely on logic and reason, but neuroscientists have confirmed that emotions play a decisive role in decision-making. Narrative seems to be more effective in changing beliefs than writings specifically designed to persuade through arguments and evidence.

Building a story from data insights dignifies creating a bridge from data to the emotional and influential side of the brain. People are moved by emotions, so attitudes, fears, hopes and values are strongly influenced by stories.

The story is a tool that enables the transmission of information, ideas and insights in an extremely effective way mainly for three reasons:

♦ Memorability: Stanford Chip Heath (author of Made to Stick) found that when students are asked to recall speeches, 63% of them are able to remember stories, but only 5% are able to remember a single statistic. 

♦ Persuasiveness: In a study conducted to test two variants of a Save the Children organisation brochure, it was shown that sharing life stories of African children is much more persuasive than reporting statistics on their living conditions.

♦ Engagement: Green and Brock's study (2020) show that the more readers are absorbed by a story, the more the story has an effect on them and their beliefs: when listening to a story, we tend to lower our intellectual guard and be less critical and sceptical. The story has the power to move us emotionally and make us lose sight of rational considerations.

Key story elements Click to read  

Data storytelling uses the same narrative elements as any story you’ve read or heard before: characters, setting, conflict, and resolution. The following is an example of a data story proposed by Harvard Business School.

Imagine you’re a data analyst and just discovered your company’s recent decline in sales has been driven by customers of all genders between the ages of 14 and 23. You find that the drop was caused by a viral social media post highlighting your company’s negative impact on the environment, and craft a narrative using the four key story elements:

♦ Characters - The protagonists of our story are customers aged between 14 and 23, environmentally conscious consumers and your internal team. This doesn’t need to be part of your presentation, but you should define the key players for yourself beforehand.

♦ Setting - Set the scene by explaining there’s been a recent drop in sales driven by customers of all genders ages 14 to 23. Use a data visualization to show the decline across audience types and highlight the largest drop in young users.

♦ Conflict - Describe the root issue: A viral social media post highlighted your company’s negative impact on the environment and caused tens of thousands of young customers to stop using your product. Incorporate research (such as this article in the Harvard Business Review) about how consumers are more environmentally conscious than ever and how sustainably-marketed products can potentially drive more revenue than their unsustainable counterparts. Remind the team of your company’s current unsustainable manufacturing practices to clarify why customers stopped purchasing your product. Use visualizations here, too.

♦ Resolution - Propose your solution. Based on this data, you present a long-term goal to pivot to sustainable manufacturing practices. You also center marketing and public relations efforts on making this pivot visible across all audience segments. Use visualizations that show the investment required for sustainable manufacturing practices can pay off in the form of earning customers from the growing environmentally conscious market segment.

If there isn’t a conflict in your data story—for instance, if the data showed your current marketing campaign was driving traffic and exceeding your goal—you can skip that element and go straight to recommending that the current course of action be maintained.

Whatever story the data tells, you can communicate it effectively by formatting your narrative with these elements and walking your audience through each piece with the help of visualizations.

 



Keywords

Data; Journalism; Storytelling; Narrative; Data Visualization

Objectives/goals:

-    Learn about data journalism and the skills required to be a data journalist

-    Understand what is data storytelling and how to build an effective data story

-    Learning how to exploit narrative techniques to disseminate insights from data


Description:

This course presents the concepts of data journalism and data storytelling. These concepts are described and explained in relation to the world of data. It is explained how to merge data science, a field of study characterised by hard skills, with soft skills and what the advantages of this combination are.

Bibliography

https://insightsoftware.com/encyclopedia/data-journalism/

https://www.analyticsvidhya.com/blog/2022/01/what-is-data-journalism-all-about/

https://shorthand.com/the-craft/how-to-create-great-data-journalism/index.html

https://online.hbs.edu/blog/post/data-storytelling#:~:text=What%20Is%20Data%20Storytelling%3F,inspire%20action%20from%20your%20audience.

https://www.forbes.com/sites/brentdykes/2016/03/31/data-storytelling-the-essential-data-science-skill-everyone-needs/?sh=3823af6152ad

https://www.nugit.co/what-is-data-storytelling/

Green, M. C. & Brock, T. C. (2000). The Role of Transportation in the Persuasiveness of Public Narrative. Journal of Personality and Social Psychology. 79, 701-721.

Heath, C. & Heath, Dan. (2007). Made to Stick: Why Some Ideas Survive and Others Die. New York: Random House.

Guber, P. (2011). Tell to Win: Connect, Persuade, and Triumph with the Hidden Power of Story. Sidney: Currency.


Related training material

Partners

Università del Salento
Demostene Centro Studi
IHF
IWS
Universidad de Oviedo
ASE
WAI