McGraw-Hill/Irwin Copyright © 2014 by The McGraw-Hill Companies, Inc. All rights reserved. EXPLORING, DISPLAYING, AND EXAMINING DATA Chapter 16
Feb 24, 2016
McGraw-Hill/Irwin Copyright © 2014 by The McGraw-Hill Companies, Inc. All rights reserved.
EXPLORING, DISPLAYING, AND EXAMINING DATAChapter 16
16-2
Learning Objectives
Understand . . .That exploratory data analysis
techniques provide insights and data diagnostics by emphasizing visual representations of the data.
How cross-tabulation is used to examine relationships involving categorical variables, serves as a framework for later statistical testing, and makes an efficient tool for data visualization and later decision-making.
16-3
Pull Quote
“On a day-to-day basis, look for inspiration and ideas outside the research industry to influence your thinking. For example, data visualization could be inspired by an infographic you see in a favorite magazine, or even a piece of art you see in a museum.”
Amanda Durkee, partnerZanthus
16-4
Researcher Skill Improves Data Discovery
DDW is a global player in research services. As this ad proclaims, you can “push data into a template and get the job done,” but you are unlikely to make discoveries using a template process.
16-5
Exploratory Data Analysis
ConfirmatoryExploratory
16-6
Data Exploration, Examination, and Analysis in the Research Process
16-7
Research Values the Unexpected
“It is precisely because the unexpected jolts us out of our preconceived notions, our assumptions, our certainties, that it is such a fertile source of innovation.”
Peter Drucker, authorInnovation and Entrepreneurship
16-8
Frequency: Appropriate Social Networking Age
16-9
Bar Chart
16-10
Pie Chart
16-11
Frequency Table
16-12
Histogram
16-13
Stem-and-Leaf Display455666788889124667990223567802268
240183106336
3
68
56789
101112131415161718192021
16-14
Pareto Diagram
16-15
Boxplot Components
16-16
Diagnostics with Boxplots
16-17
Boxplot Comparison
16-18
Mapping
16-19
SPSS Cross-Tabulation
16-20
Percentages in Cross-Tabulation
16-21
Guidelines for Using Percentages
Don’t average percentages
Don’t use too large a percentage
Don’t use too small a base
Changes should never exceed 100%
Higher number is the denominator
16-22
Cross-Tabulation with Control and Nested Variables
16-23
Automatic Interaction Detection (AID)
16-24
Exploratory Data Analysis
This Booth Research Services ad suggests that the researcher’s role is to make sense of data displays.
Great data exploration and analysis delivers insight from data.
16-25
Key Terms
Automatic interaction detection (AID)
BoxplotCellConfirmatory data
analysisContingency tableControl variableCross-tabulationExploratory data
analysis (EDA)
Five-number summary
Frequency tableHistogramInterquartile range
(IQR)MarginalsNonresistant
statisticsOutliersPareto diagramResistant statisticsStem-and-leaf
display
McGraw-Hill/Irwin Copyright © 2014 by The McGraw-Hill Companies, Inc. All rights reserved.
ADDITIONAL DISCUSSION OPPORTUNITIESChapter 16
16-27
Snapshot: Novation
No standarded vocabulary across companies
Serve variety of users
Ad hoc analysis with sophisticated visualizations
Big data with sophisticated analytical tool.
16-28
Snapshot: Digital Natives vs. Digital Immigrants
30 subjects = 15 natives, 15 immigrants
Monitored media behaviors
300 hours of real-time data
Biometric Monitoring: emotional engagement
16-30
Snapshot: Internet-age Researchers
“The ability to take data—to be able to understand it, to process it, to extract value from it, to visualize it, to communicate it—that’s going to be a hugely important skill in the nextdecades . . . .”
16-31
Research Thought Leader
“As data availability continues to increase, theimportance of identifying/filtering and analyzingrelevant data can be a powerful way to gain aninformation advantage over our competition.”
Tom H.C. Anderson founder & managing partner
Anderson Analytics, LLC
16-32
PulsePoint: Research Revelation
65 The percent boost in company revenue created by best practices in data quality.
16-33
Geograph: Digital Camera Ownership
16-34
CloseUp: Working with Data Tables
16-35
CloseUp: Original Data Table
16-36
CloseUp: Arranged by SpendingMost to Least
16-37
CloseUp: Arranged by Average Annual Purchases, Most to Least
16-38
CloseUp: Arranged by Average Transaction, Most to Least
16-39
CloseUp: Arranged by Estimated Average Transaction, Least to Most
McGraw-Hill/Irwin Copyright © 2014 by The McGraw-Hill Companies, Inc. All rights reserved.
EXPLORING, DISPLAYING, AND EXAMINING DATAChapter 16
16-41
Photo AttributionsSlide Source
4 Courtesy of Radius Global Market Research18 Courtesy of RealtyTrac21 Vstock/Alamy24 Courtesy of Booth Research Services27 Courtesy of Novation
28 Realistic Reflections29 Courtesy of DecisionPro; Digital Vision/Getty
Images30 Vstock LLC/Getty Images