Introduction to Data Visualisation


Tarek Amr


@gr33ndata


What is Data-driven Journalism


What is Data-driven Journalism


Data can be:


Source for journalist.

Tool with which the story is told.

Both.

Why Journalists Should Use Data


In the past: Only source of news.

Today: Twitter, Facebook, Blogs, etc.

Why Journalists Should Use Data


Gathering, filtering and visualizing what is happening beyond what the eye can see has a growing value.

Why Journalists Should Use Data


The Guardian


Wikileaks (Afghanistan): 92,201 rows of data.

Wikileaks (Iraq): 391,000 rows of data.

Why Use Data


Customise to reader

DDJ Examples


Parliament


The Telegraph: MP Expenses*

The Times MP Speeches.


What is Visualisation


What is Data visualization?


"Graphical display of abstract information"

Stephen Few


Mental Model


2 + 2 = ?

Mental Model


Mental Model

Communication

Mental Model

Sense-making (Analysis)

Visualising GDP


Egypt - Emirates - Saudia

Gross Domestic Product, 2011

Intro. Examples


Charles Joseph Minard (1781-1870)

Intro. Examples


Which metro map is more useful?

Intro. Examples


Which metro map is more useful?

Intro. Examples


Is this really a visualisation?

Exercise


Visualis the following:

$ 1 = L.E 6.89 (2013)

$ 1 = L.E 5.95 (2011)

Visualisation vs. Infographic


Visualisation: Let readers Explore Data.

Infograph: Edits data to Tell a Story.

Boundaries are Fuzzy!

Form Follows Function


Before editing data into an infograph,

think of the questions readers may have.

and the Journalistic stories you want to tell.

Your infograph should then answer the readers' questions.

Exercise


What are the questions and stories?

GINI Index, Education and GDP*.

Favourite Music.

What data at the moment?


Design Principles


What can you see?



Don Norman's Design Principles


Visibility


Web links are blue and underlined

Black text doesn't invite users to click on them.

Feedback and Constraints


What is your shirt size?

Small

Mapping


Go Down!

Go Up!

Consistency




Global: Visual metaphors.

Local: Keep my style consistent.



Design Elements


Design Elements

Line

A line defines boundaries and edges.

A line shows directions and progess.

Design Elements

Shapes

Areas defined by lines or colours.

Positive & Negative Shapes.

Compare lengths vs. areas.

Design Elements

Colours

Colours can aid in organization.

Colours can give emphasis.

Composition


Rule of Thirds.


Rule of Odds.


Rule of Space.

Clustering


Clustering


Clustering


Clustering



Data Types


Numerical


Price: EGP 1230

Height: 187 cm.

Age: 26 years

Categorical


Fruits: Apples, Oranges, Pears

Cities: Cairo, Alexandria, Aswan

Ordered


T-Shirt Sizes: S, M, L, XL

Weekdays: Sun, Mon, Tues, Wednes, etc.

Univariate


Ahmed: 19 years

Ali: 22 years

Bivariate


Ahmed: 22 years, 91 Kg.

Ali: 26 years, 62 kg.


How to Represent Information


Comparison


Barcharts


Axis, numbers, start at zero

Horizontal vs Vertical

How about Age + Weight?

Relationships


(+ve /  -ve)  (2-D /  3-D) 

Histograms


They look like barcharts

Are they the same thing?

Connections


Network vs. Matrix

Connections


The network representation is more useful in representing the topographical structure of the elements and how each pair of them are connected on a micro level.

Connections


The matrix representation on the other hand is more flexible in reordering the element in order to show how the relationships between elements on a macro level.

Geographic Data



Most popular girl names by state

Maps


Driving is why you are fat.

Is map the best option for this?

Can you redisign it?

Heavy Metal



Heavy Metal Bands per 100.000 people


Appendix


Useful Resources


Which chart to use?

Selecting Colors for Data Visualization

Data visualization guidelines – By Gregor Aisch

Data Ink