Infographic

• Manually created
• High degree of customization
• Pixels manually placed by someone

Data Visualization (artefact)

• Algorithmically created
• Output from software
• Easy to replicate, modify, scale
• Low degree of customization

Nussbaumer

1. Understand the context
2. Choose an appropriate display
3. Eliminate clutter
4. Draw attention where you want it
5. Think like a designer
6. Tell a story

Phase 1: What to vizualize

1. Consider your inputs
2. Consider your (Designer) goals
3. Consider their needs (Reader)
4. Consider the data properties
5. Write a spec for the vizualization; e.g. "Show the relationship between A and B [and C...] across X [and Y...] from m to n in order to determine our best and worst performing widgets.”
6. Select data to include

Phase 2: How to vizualize

1. Select axes for most important data/relationship
2. Consider defaults and best encodings for all other data
3. Experiment/iterate

Choosing an Effective Visual

Simple Text 

• Consider simple text if you have a number or two
• If you have a number - does not mean you need a graph

Choosing an Effective Visual

More numbers

Consider to use table or a graph

Focus on data not borders, font, etc. 

Choosing an Effective Visual

Simple text

Tables

Heatmap

Graphs

Line graph

Slope graph

Bar charts

Pie charts - including donut chart

3D

Secondary y axis

Simple text - when you have few numbers. 

Table - interact with our verbal system. more numerical - focus on data not look (borders)

Heatmap - table that shows intensity using colour. 

Graphs - interact with visual system

• Type of graph - points, lines, bars, area

Points - Scatter plot - relationship between two things - shows what relationship between two is. 

Points - Lines - plot continuous data. We can use standard line or slope graph. 

Slopegraph - use to show two point in time period. and show if relationship increased or decreased. 

Bars - common and therefore reduce cognitive load. 

Try to use data labels and remove axis. Bars shoudl be wider than a white space between bars. 

Vertical bar - single, two, multiple series. But if you add to many, hard to understand. 

Stacked vertical - more limited in use. 

Waterfall chart - pulls a part pieces of vertical stacked to focus one at the time. 

Horizontal bar - good choice for categorical data. 

Stacked horizontal - can show absolute numbers or sum to 100%

Area - ok to use for visualization of magnitude only.  

Cognitive load

Clutter

When visual slides look too busy/complicated. 

Clutter - visual elements that take space but not add value or increase understanding. 

1. Proximity
2. Similarity
3. Enclosure
4. Closure
5. Continuity
6. Connection