Displaying Data in Graphs and charts Challenger Disaster

  • Slides: 30
Download presentation
Displaying Data in Graphs and charts

Displaying Data in Graphs and charts

Challenger Disaster

Challenger Disaster

Report of the Presidential Commission on the Space Shuttle Challenger Accident, 1986 (vol 1,

Report of the Presidential Commission on the Space Shuttle Challenger Accident, 1986 (vol 1, p. 145)

How to Display Data (Badly) l Showing the data (choosing the data) l Showing

How to Display Data (Badly) l Showing the data (choosing the data) l Showing data accurately l Showing data clearly H Wainer(1984). “How to display data badly. ” American Statistician 38(2): 137 -147. http: //www. jstor. org/view/00031305/di 020581/02 p 0048 x/0

Showing Data

Showing Data

Showing data l Edward Tufte l Data Density Index l Chart Junk l It

Showing data l Edward Tufte l Data Density Index l Chart Junk l It does not make sense to use graphs to display very small amounts of data. l The human brain is quite capable of grasping up to three values.

Showing Data

Showing Data

Showing Data l Graphs are only as good as the data they display. l

Showing Data l Graphs are only as good as the data they display. l No amount of creativity can produce a good graph from dubious data.

Showing Data

Showing Data

Showing Data Accurately l Graphs should not provide a distorted picture of the values

Showing Data Accurately l Graphs should not provide a distorted picture of the values they portray. l Distortion can be either deliberate or accidental. l Common l Use Distortions of 3 dimensional “effects” l Use of linear scaling when using area or volume to represent values.

3 dimensional effects

3 dimensional effects

No 3 D Effect

No 3 D Effect

Linear Distortion Fact: 1973 dollar : $0. 44 1958 dollar: $1 Dimensions 1958 dollar

Linear Distortion Fact: 1973 dollar : $0. 44 1958 dollar: $1 Dimensions 1958 dollar is less than twice the height and width of the 1973 dollar Area: 1958 dollar has 3 x the area

No Distortion

No Distortion

Hiding Information

Hiding Information

Not so hidden

Not so hidden

Showing Data Clearly l Graphs should be no more complex than the data which

Showing Data Clearly l Graphs should be no more complex than the data which they portray l Unnecessary complexity can be introduced by chartjunk l irrelevant l color decoration

Showing Data Clearly Age Structure of College Enrolment (19721976) • 4 variables • 6

Showing Data Clearly Age Structure of College Enrolment (19721976) • 4 variables • 6 colors • Weird perspective • SPLIT AXIS American Education Magazine.

Showing Data Clearly

Showing Data Clearly

Tips for “Good” Graphs, Tables and Charts l Tell the truth – don’t distort

Tips for “Good” Graphs, Tables and Charts l Tell the truth – don’t distort the data l If the data is very simple, just use text l Simplify If the “story” is simple, keep it simple. l If the “story” is complex, make it look simpler l “Keep it Simple” (for presentations) l

A complex graph is sometimes the best way to tell a story

A complex graph is sometimes the best way to tell a story

The Cognitive Style of Power. Point (Edward Tufte) l How Not to Use Power.

The Cognitive Style of Power. Point (Edward Tufte) l How Not to Use Power. Point l Bullet Points are bad l Style no substitute for content l 40 words = 8 seconds of silent reading = very little actual content

Analytics According to Captain Kirk l Power. Point does not provide enough detail for

Analytics According to Captain Kirk l Power. Point does not provide enough detail for analysis l One needs to look at all of the data to do a proper analysis

Clarification Simple

Clarification Simple