Worth 1,000 Words

1
Worth 1,000 Words

• How to use information graphics to make data
  meaningful
• National Association for Career and Technical
  Education InformationMay 17, 2012
• Rhonda Welfarerhonda.welfare_at_dpi.nc.gov

2
Visualization

• Using graphic images to tell a story with data

3
Goal of information graphics

• Readable
• Understandable
• Usable
• Every artistic decision made with the needs of
  readers in mind.

4

• Graphical excellence is that which gives to the
  viewer the greatest number of ideas in the
  shortest time, with the least amount of ink in
  the smallest amount of space.
• Edward TufteThe Visual Display of Quantitative
  Information

5
Why use information graphics?

• Helps the reader see patterns and relationships
• Reinforces information
• Efficient use of reader time
• Communicates complex information clearly

6
Getting started

• Ask what story you want to tell
• Think character development

7
Design considerations

• Explain encodings (legend)
• Label axes if needed
• Get the math right
• Edit unnecessary words and digits
• Graphic should be able to stand alone
• Include the source
• Consider the audience

NC CTE Analysis Reporting System, NCDPI 2012
8

• PROOFRED
• PROOFREED
• PROOF

X
X
9
Organize data

• Spreadsheet
• Assign record number
• Use unique identifier

• Find problems with data
• Relevant
• Accurate
• Realistically reported
• Check for missing data
• Not misleading
• Identify projections vs. real data
• Identify contributing factors

10
What software to use?

• WHATEVER WORKS FOR YOU

• Examples
• Out of the box
• Excel
• Google Spreadsheets
• Some assembly required
• Java
• Apex
• R
• Illustration tools
• Adobe Illustrator
• Freehand

11
Types of graphics

• Text driven
• Timeline
• Can be tied to a series of historic events
• Tables
• When numbers themselves are important
• Fact boxes
• To emphasize

12
Categorical data (proportional)

• Pie charts
• Small number of values
• Sort from greatest to least
• Must add up to 100
• Indicate what the whole represents
• Proportions and design must be accurate
• Use of color

NC CTE Analysis Reporting System, NCDPI 2012
13
Categorical data (proportional)

• Donut chart
• Very few values
• Stacked bar
• Instead of multiple pie charts
• Combination
• Call out area of interest

Total 804,850
NC CTE Analysis Reporting System, NCDPI
2012 Breakdown is approximate
14
Categorical data (not proportional)

• Bar graph
• Either horizontal or vertical
• X is category
• Y is number or percent

NC CTE Analysis Reporting System, NCDPI 2012
15
Time series (cross section)

• What are you looking for?
• Trends over time
• Bar graph
• Y scale
• X time
• Use of color

NC CTE Analysis Reporting System, NCDPI 2012
16
Time series (cross section)

• Stacked bar
• Only if values are meaningful
• Scatterplot
• Y scale
• X time
• Use grid to make it easier to read

NC CTE Analysis Reporting System, NCDPI 2012
17
Time series (continuous measure)

• What are you looking for?
• Changes over time
• Line graph (fever chart)
• Shows how a single value changes across time

NC CTE Analysis Reporting System, NCDPI 2012
18
(No Transcript)
19
Multiple variables
NC CTE Enrollment Trends

• What are you looking for?
• Correlation
• Relationships
• Scatterplot
• Bubble chart
• Distribution/histogram

NC CTE Analysis Reporting System, NCDPI 2012
20
(No Transcript)
21
Design considerations

• Label each data point
• Sans serif font
• Legend and labels 10 pt (print), 14 pt
  (presentation)
• Headlines 18 pt (print), 24 pt bold
  (presentation)
• When should you hire someone to do it for you?
• When you dont have time to do it right
• When your ideas are greater than your skills
• When you dont have any ideas

22

• PROOFRED
• PROOFREED
• PROOF

X
X
23
Data can be fun!

• http//graphjam.memebase.com/Using charts and
  graphs to explore popular culture
• http//blog.okcupid.comQuantitative
  representation of tweets
• http//www.jir.comThe Journal of Irreproducible
  Results

24
Where to find data

• Your reports
• Performance indicators
• Enrollment
• Financial data
• Other sources of data
• www.census.govThe Census Bureau
• Data.gov and numbrary.comCatalogs of government
  data
• Nces.ed.govNational Center for Education
  Statistics

25
Where to learn more

• Wong, Dona M. (2010). Wall Street Journal Guide
  to Information Graphics The Dos and Donts of
  Presenting Data, Facts and Figures.
• Harris, Robert L. (2000) Information Graphics A
  comprehensive Illustrated Reference
• Klanten, R., Ehmann, S., and Schulze, F. (2011).
  Visual storytelling Inspiring a New Visual
  Language.
• Yau, Nathan. (2011). Visualize This The
  FlowingData Guide to Design, Visualization, and
  Statistics. Flowingdata.com
• George-Palilonis, Jennifer (2008). A Practical
  Guide to Graphics Reporting Information Graphics
  for Print, Web Broadcast.
• Tufte, Edward (2011). The Visual Display of
  Quantitative Information.

26

• PROOFRED
• PROOFREED
• PROOF

X
X