Lecture 5 Data Visualisation

Data Visualisation

Data Visualisation

Psychology of Data Visualisation

Charts

Bits and Pieces

@Victoria University BCO6007 Business Analytics


Audience Considerations


Common Data Visualisation Issues


Why Data Visualisation?

A picture is worth 1,000 words


Why Data Visualisation


This process consists of the following six fundamental stages:

1. Determine your message and identify the data necessary to

communicate it.

2. Determine if a table, graph, image or combination is needed to

communicate your message.

3. Determine the best means to encode the values.

4. Determine where to display each variable.

5. Determine the best design for the remaining objects.

6. Determine if particular data should be featured above the rest, and if so,

how.

Few


In the Beginning


William Playfair 1786

Trade balance


John Snow 1854


The Value of Data Visualisation


Exploration or Explanation?


Member Engagement (Harvard Business Review)

https://hbr.org/2013/04/when-visualizing-data-you-have




Why the visualisation

What am I looking at?

Why are you showing it to me?

Examples





Youth Unemployment Rates in Europe


Why are you showing me?

Youth Unemployment Rates in Europe

Historically high and dividing Europe

Psychology of

Visualisation


Memory


Hick’s Law


Working Memory


Chunking


More details


Pre-attentive Processing

Pre-attentive processing is the unconscious accumulation of

information from the environment. All available information is pre-

attentively processed. Then, the brain filters and processes what is

important. (Wikipedia)


Preattentive Processing

Attentive Processing

Preattentive Processing

Goal: How many 5’s



Compare and contrast objects to make a point


Face of Meth Campaign

2001 2004 2006


Group objects so that relationships between the elements becomes

clear




Relationships



Break elements down so that the individual parts become clear


CO2 per capita emissions per country



Pictorial Representations


Raleigh Bicycle – exploded


Storyboard a narrative so that it unfolds in a logical way



Create cutaways to see inside an object



Measure or compare



Scale is relative


Scale is relative


Label Parts



Location


Colour


Interference Effects


Colour

Use Colour sparingly

Use 2 colours at most where possible

Avoid colour gradients


Smartphones Dominating Sales


Smartphone Dominating sales


Avoid Colour Gradients


Colour Blind – 8% Males


Deuteranopia Tritanopia

Protanopia



Charts


Tables Vs Charts

Tables

Need to looked up individual

values

Data needs to be precise

Charts

The message you wish to

communicate resides in the

shape of the data



Business Data Relationships

Time-Series Relationships

When quantitative values are expressed as a series

of measures taken across equal intervals of time,

this relationship is called a time series.

Ranking Relationships

When quantitative values are sequenced by size,

from large to small or vice versa, this relationship is

called a ranking.

Part-to-Whole Relationships

When quantitative values are displayed to reveal the

portion that each value represents to some whole,

this is called a part-to-whole relationship.


Business Data Relationships

Deviation Relationships

When quantitative values are displayed to feature

how one or more sets of values differ from some

reference set of values, this is called a deviation

relationship.

Distribution Relationships

When we show how a set of quantitative values are

spread across their entire range, this relationship is

called a distribution.

Correlation Relationships

When pairs of quantitative values, each measuring

something different about an entity are displayed to

reveal if there is significant relationship between

them.


Encoding Quantitative Data in Charts

Recommended



Not Recommended – 2 dimension



Use line charts to show time series

Notes on Matters Affecting the Health, Efficiency, and Hospital Administration of the British Army and sent to Queen Victoria in 1858


A better visualisation



Enrolments

2011, 45

2012, 48 2013, 55

2014, 50

Enrolments By Year

ERP, 50

MBA, 45

Accounting, 43

Marketing, 20

2014 Enrolments By Course


What is the problem?



NO 3D!



Reporting Guidelines - Charts




Label axis


Always Scale to Zero


Direct labelling


Andrew Ablea

Bits and Pieces


Reduce Data Ink / Non data Ink Ratio


Maps


Maps

Mercator map accuracy?



Resources

http://labs.juiceanalytics.com/chartchooser/index.html

Google Charts

Stephen Few

SlideShare

http://labs.juiceanalytics.com/chartchooser/index.html

Lecture 5 Data Visualisation

Education

value of data visualisation

data necessary

particular data

contrast objects

remaining objects

youth unemployment rates

best design

available information