Assignment #2 (Nov-9)
– Work with challenge data
– Divide and conquer your data
– Submit individually (one unique submission per
student. No plagiarism!)
– Explore your data using – Simple statistics
– Simple charts
– Report observations (What do you see?)
– Formulate hypotheses (Why could that be?)
Assignment #2 (Nov-9)
– Describe data
– General type of data
– How large and complex
– Fields, links,
– What’s your challenge question
– Explore
– 4-5 exploratory graphs – Explanations
– Simple stats
– Reflect and hypothesis – Reflect on data
– 3-5 hypotheses
Python programming
Outline for today
What are visualizations and what do we need to learn about them?
1. Visualization 101
2. Visualization Literacy
3. Basic Charts
4. Visualization design process
Visual representation of data:
Visual representation of data:
Pre-attentiveness: –
– Parallelity:
Anscombe’s Quartet
Anscombe’s Quartet
Statistics
– Numbers + math
– Comparison
– Small differences
– Answers questions
(hypothesis testing)
– Detail
– Single information
– …
Visualization
– Overview
– Creates questions
(hypothesis generation)
– Serendipity
– Multiple information
– …
Communication
– Interpretation
– Critiquing
– Contextualization – Presentation
– …
Sense making process
Data
Numbers, relations, records, text, analysis, …
Action
Decisions Emotions, Knowledge Insight
Sense making process
Data
Numbers, relations, records, text, analysis, …
Visualization
visual representation
Action
Decisions Emotions, Knowledge Insight
Sense making process
Data
Numbers, relations, records, text, analysis, …
Visualization
visual representation
Information
Insights, Facts
Action
Decisions Emotions, Knowledge Insight
Exploration
Exploration Insights
Exploration Insights
Explanation
Exploration
Data centered Domain experts
Generating Insights
Lab Setting
Insights
Explanation
Exploration
Data centered Domain experts
Generating Insights
Lab Setting
Insights
Explanation
Human centered Non-experts
Conveying messages
In-the-wild / diverse
Visualization Literacy
– confidently use
–
visual queries
– interpreting visual patterns
Visualization Literacy Reading:
– Correctly decode (simple & complex) visual representations
– Know pitfalls and deceptions
– Think critically ‘beyond’ and see ‘through’ the visualization
Design:
– Create efficient and effective visualizations
– Design efficient and effective visualizations
Explore:
– perform tasks: ask and answer questions
– Interact with with visualizations
What are the differences?
What are the differences?
What are the differences?
What are the differences?
Exploratory Data Analysis
– Observe your data from many perspectives
– Observe as many facets as possible
– It’s called the Grand Tour
– Obtain as much observations as possible
– Generate insight and information
– Take notes
– Collaborate
– Communicate
Distributions
Descriptive Statistics with Numpy
(lat.: medius), middle (think: median => middle)
(think: mode => most)
Mean == average
https://www.wolfram.com/mathematica/new-in-8/statistical-visualization/specify-bin-sizes-for-histograms.html
Violin plots
Violin plot
Violin plots
Violin plots for comparison
Violin plots vs. box plots
http://www.stat.cmu.edu/~rnugent/PCMI2016/ papers/ViolinPlots.pdf
More plots!
Dot plot Error bars Box plots Violin plots
Confidence Intervals (error bars)
Error Bars in Line Charts
Alternatives
Bar plot
Mean +
CIs (Uncertainty)
Box plot
Median + IQR (spread)
Violin plot
Median + Distribution
Cheat sheets
https://visualizationcheatsheets.github.io/
Pie charts
Pie Charts
+ Few values
+ Huge differences
+ Easy communication + Order by size!
Alternatives to pie charts
Alternatives to pie charts
Multi-dimensional data
Parallel Coordinates Plot
+ Scalable
+ Consice
+ Good overview
– Depending on ordering
– Can suffer from clutter
– Visual path following can be hard
Multi Dimensional Scaling (MDS)
+ Dimension reduction
+ Visualization in 2D or 3D
+ Visual clustering
– Information lost – Creates artifacts:
– false neighbors and – Tears
Which visualization?
1. Data
1. Data
Item
1. Data
Item
Attribute
1. Data
Value
Attribute
Item
Visual Marks
Visual Variables
Bertin, Jacques. “Semiology of Graphics: Diagrams.” Networks, Maps 10.00690805.1987 (1983): 10438353.
Visual Variables
Bertin, Jacques. “Semiology of Graphics: Diagrams.” Networks, Maps 10.00690805.1987 (1983): 10438353.
Visual Variables
Bertin, Jacques. “Semiology of Graphics: Diagrams.” Networks, Maps 10.00690805.1987 (1983): 10438353.
Visual Variables
Bertin, Jacques. “Semiology of Graphics: Diagrams.” Networks, Maps 10.00690805.1987 (1983): 10438353.
Visual Variables
Bertin, Jacques. “Semiology of Graphics: Diagrams.” Networks, Maps 10.00690805.1987 (1983): 10438353.
Visual Variables
Bertin, Jacques. “Semiology of Graphics: Diagrams.” Networks, Maps 10.00690805.1987 (1983): 10438353.
Visual Variables
Bertin, Jacques. “Semiology of Graphics: Diagrams.” Networks, Maps 10.00690805.1987 (1983): 10438353.
Visual Variables
Visual Variables
Double Diamond
https://uxdesign.cc/how-to-solve-problems-applying-a-uxdesign-designthinking-hcd-or- any-design-process-from-scratch-v2-aa16e2dd550b
Design Thinking
Design thinking is a human-centered approach to creative problem solving.
¡ñ is about people
¡ð empathy, problems, context, problem
http://www.theagileelephant.com/what-is-design-thinking/ https://www.ted.com/talks/david_kelley_human_centered_design Rowe, Peter G. Design thinking. MIT press, 1987.
Design Thinking
Design thinking is a human-centered approach to creative problem solving.
¡ñ is about people
¡ð empathy, problems, context, problem
¡ñ Highly creative
¡ð ideas, discussion, iteration
http://www.theagileelephant.com/what-is-design-thinking/ https://www.ted.com/talks/david_kelley_human_centered_design Rowe, Peter G. Design thinking. MIT press, 1987.
Design Thinking
Design thinking is a human-centered approach to creative problem solving.
¡ñ is about people
¡ð empathy, problems, context, problem
¡ñ Highly creative
¡ð ideas, discussion, iteration
¡ñ hands-on
¡ð develop, prototype, test, try, …
¡ñ
http://www.theagileelephant.com/what-is-design-thinking/ https://www.ted.com/talks/david_kelley_human_centered_design Rowe, Peter G. Design thinking. MIT press, 1987.
Design Thinking
Design thinking is a human-centered approach to creative problem solving.
¡ñ is about people
¡ð empathy, problems, context, problem
¡ñ Highly creative
¡ð ideas, discussion, iteration
¡ñ hands-on
¡ð develop, prototype, test, try, …
¡ñ Show, don’t tell
http://www.theagileelephant.com/what-is-design-thinking/ https://www.ted.com/talks/david_kelley_human_centered_design Rowe, Peter G. Design thinking. MIT press, 1987.
Design Thinking
Design thinking is a human-centered approach to creative problem solving.
¡ñ is about people
¡ð empathy, problems, context, problem
¡ñ Highly creative
¡ð ideas, discussion, iteration
¡ñ hands-on
¡ð develop, prototype, test, try, …
¡ñ Show, don’t tell
¡ñ iterative
¡ð failure, progress, iterate, feedback,…
http://www.theagileelephant.com/what-is-design-thinking/ https://www.ted.com/talks/david_kelley_human_centered_design Rowe, Peter G. Design thinking. MIT press, 1987.
Wrap-up
1. Many forms of visualization
2. Visualization works through pre-attentiveness and parallel
processing
3. Visualization is a process
4. Visualization literacy is the ability to correctly understand and
use visual representations of data
5. Many basic chart types
6. Visual variables for design
7. Design thinking process