Visual Analytics Detect the Expected Discover the Unexpected A Tutorial for Middle School and High School Teachers - PowerPoint PPT Presentation

Loading...

PPT – Visual Analytics Detect the Expected Discover the Unexpected A Tutorial for Middle School and High School Teachers PowerPoint presentation | free to download - id: 560368-NDUwO



Loading


The Adobe Flash plugin is needed to view this content

Get the plugin now

View by Category
About This Presentation
Title:

Visual Analytics Detect the Expected Discover the Unexpected A Tutorial for Middle School and High School Teachers

Description:

Visual AnalyticsDetect the ExpectedDiscover the UnexpectedA Tutorial for Middle School and High School Teachers. Module 3- Misinformation and Lying with Graphics – PowerPoint PPT presentation

Number of Views:71
Avg rating:3.0/5.0
Slides: 26
Provided by: purdueEdu1
Learn more at: http://www.purdue.edu
Category:

less

Write a Comment
User Comments (0)
Transcript and Presenter's Notes

Title: Visual Analytics Detect the Expected Discover the Unexpected A Tutorial for Middle School and High School Teachers


1
Visual Analytics Detect the Expected Discover the
Unexpected A Tutorial for Middle School and High
School Teachers
  • Module 3- Misinformation and Lying with Graphics

2
Tutorial Outline
  • Introduction
  • Module 1 What is Visual Analytics?
  • Module 2 Proper Visualization Construction and
    Use
  • Module 3 Misinformation and Lying with Graphics
  • Module 4 Department of Homeland Security (DHS)
    Why and How They Use Visualizations
  • Module 5 Understandable Applications
  • Module 6 Exercises and Resources for the
    Classroom
  • Conclusion

3
Module 3 Misinformation and Lying with Graphics
  • Statistics
  • Statistical Sampling
  • Historical Examples
  • Problems with Data Collection
  • Using a Critical Eye

4
Lies, Damned Lies, and Statistics
  • Statistics Add Legitimacy
  • Statistics Crunch Numbers
  • Statistics Dont Discriminate
  • Statistics Run on Bad Datagt Statistics of Bad
    Data

5
Bad Data
  • What is Bad Data?
  • Why Does Bad Data Abound?
  • Why Dont We Get Good Data?
  • Where Does Bad Data Come From?

6
People Collect Data
  • Bad Data Comes From Us!
  • Data Collection is a Social Activity
  • Data is a Product of Choices, Compromises,
    Expertise, Attitude, Proximity

7
Algebra vs. Statistics
  • Algebra
  • Statistics
  • Used to find answer
  • 7x3
  • X-4
  • One variable /one data point
  • One answer
  • Used to analyze data
  • Corpus of data
  • Characteristics of data

8
The Random Sample
  • Random Selection from Entire Universe
  • Set of Samples Representative of Universe
  • Expensive Difficult

9
Stratified Random Sample
  • Universe Broken Down to Proportional Groups
  • How Do You Know the Right Size of the Groups?
  • Which Came First, Chicken or the Egg?
  • Sampling Bias

10
Sampling Bias
  • Bane of Polling Organizations
  • Techniques Evolved Over Time
  • More accurate
  • The Poll That Changed Polling

11
1936 Presidential Election - Sampling Bias
  • American Literary Digest
  • Poll Based Off of Subscribers, Car Owners, and
    Phone Users
  • Predicted Alf Landon Would Become the Next
    President of the U.S.A.

12
...But Roosevelt Won Re-election!
  • What Happened? Sampling Bias!
  • Over Represented Rich Prosperous Therefore
    Most Likely to Vote Republican
  • Collecting (Good) Data
    is Hard and Expensive

13
President Dewey Poor Data Collection
  • Close Race
  • Gallup Stopped Polling 2 Weeks Prior
  • Prejudice

14
Data Collection Problems
  • Sampling Bias
  • Changing Opinions
  • Bias Toward Own Opinions

15
Data Collection Problems
  • Who is Asking?
  • How Can I Answer?
  • What Are They Asking?
  • Public or Private Discourse?

16
Data Collection Problems Who is Asking?
  • Post WWII Study
  • Would Blacks Be Better Off if Japan/ Germany Won
    the War?
  • Response to White Questioner
  • Response to Black Questioner

17
Data Collection Problems How Can I Answer?
  • Essay
  • Multiple Choice
  • Verbal Response

18
Data Collection Problems What Are They Asking?
  • Sensitive Issue
  • Controversial Issue
  • Unknown Issue

19
Data Collecting Problems Public or Private
Discourse?
  • Public Ordering at the Restaurant
  • Same item
  • Satisfaction
  • Private Secret Ballot
  • Privacy
  • Prevent bribery, intimidation

20
Data Collection - Why Do We Care?
  • Poor/Bad Collection Affects Results
  • Methodologies Being Developed To Remove Outside
    Influences
  • Human - Psychological
  • Machine - Calibration

21
Blind Experiments
  • Single Blind
  • Subject blind experimenter not blind
  • Pepsi Challenge
  • Double Blind
  • Subject experimenter blind
  • Placebo Group

22
Conclusion
  • Results Often Missing How Data Was Obtained
  • Data Collection Statistics Not Trivial Tasks
  • Good Data Hard to Obtain
  • Analyze Data with Critical Eye

23
Additional Resources
  • http//en.wikipedia.org/wiki/Sampling_bias
  • http//en.wikipedia.org/wiki/United_States_preside
    ntial_election,_1936
  • How to Lie with Statistics Darrel Huff
  • Damned Lies and Statistics Joel Best
  • More Damned Lies and Statistics Joel Best
  • Predictably Irrational Daniel Arielly
  • Freakonomics Dubner Levitt
  • Books by Malcolm Gladwell
  • http//books.google.com

24
Classroom Exercises
  • Simulate an Open Ballot, Secret Ballot
  • Analyze a Product Review, Endorsement
  • Analyze Statistical Data for Lies
  • Examine Demographics

25
Tutorial Outline
  • Introduction
  • Module 1 What is Visual Analytics?
  • Module 2 Proper Visualization Construction and
    Use
  • Module 3 Misinformation and Lying with Graphics
  • Module 4 Department of Homeland Security
    (DHS)Why and How They Use Visualizations
  • Module 5 Understandable Applications
  • Module 6 Exercises and Resources for the
    Classroom
  • Conclusion
About PowerShow.com