Data-driven journalism guidelines

Jacopo Ottaviani

@JacopoOttaviani

www.aljazeera.com/ewasterepublic

Guidelines

To tell data stories and live peacefully

1. Truth

Data can be used to approach the truth (supposing it exists),

but a lot depends on how it's used and visualised

"Every map is a lie."

2. Completeness

Watch out for cherry-picking: always represent phenomena in their entirety and complexity

Correlation vs. causation

Correlation does not imply causation

• Methodology should be open
• Transparency is a duty and a right
• Sources should be linked
• Beware of privacy when publishing new data

3. Transparency

4. Accuracy

Information design

Content

Fox News

Truth

Which very often is a matter of honesty

The Unbearable Lightness of Y axis

http://www.vox.com/2015/11/19/9758062/y-axis-zero-chart

5. Lucidity

"One of the biggest enemies to data reporting is speed."

Knight Lab, Northwestern University

http://verificationhandbook.com

6. Humanity

Behind rows and columns, data and statistics,

there are human stories

Numbers are a means of storytelling, not the story itself.

Columbia Journalism Review

Work in team, like scientists

themigrantsfiles.com

Collect new data, but be transparent

www.generatione.eu

Thanks!

@JacopoOttaviani