CH15 Measures of Association

Chapter 15
STAGE 4: MEASURES OF ASSOCIATION
Copyright © 2019 by The McGraw-Hill Companies, Inc. All rights reserved. 15-1
Learning Objectives
Understand . . .
 How correlation analysis may be applied to study relationships
between two or more variables.
 The uses, requirements, and interpretation of the product
moment correlation coefficient.
 How predictions are made with regression analysis using the
method of least squares to minimize errors in drawing a line of
best fit.
 How to test regression models for linearity and whether the
equation is effective in fitting the data.
 The nonparametric measures of association and the
alternatives they offer when key assumptions and
requirements for parametric techniques cannot be met.
Research Thought Leader
“The human brain is an incredible

pattern-matching machine.”
Jeff Bezos, CEO,

Amazon
Measures of Association:
Interval/Ratio Data
Pearson correlation For continuous linearly related
coefficient variables
For nonlinear data or relating a
Correlation ratio (eta) main effect to a continuous
dependent variable
One continuous and one
Biserial dichotomous variable with an
underlying normal distribution
Three variables; relating two with
Partial correlation
the third’s effect taken out
Three variables; relating one
Multiple correlation
variable with two others
Predicting one variable from
Bivariate linear regression
another’s scores
Ordinal Data
Based on concordant-discordant
Gamma pairs; proportional reduction in
error (PRE) interpretation
P-Q based; adjustment for tied
Kendall’s tau b
ranks
P-Q based; adjustment for table

Kendall’s tau c
dimensions
P-Q based; asymmetrical

Somers’s d
extension of gamma
Product moment correlation for
Spearman’s rho
ranked data
Nominal Data
Phi Chi-square based for 2*2 tables
CS based; adjustment when one table

Cramer’s V
dimension >2
CS based; flexible data and distribution
Contingency coefficient C
assumptions
Lambda PRE based interpretation
PRE based with table marginals

Goodman & Kruskal’s tau
emphasis
Uncertainty coefficient Useful for multidimensional tables
Kappa Agreement measure
Researchers Search for Insights
Burke, one of the world’s

leading research companies,
claims researchers add the
most value to a project when
they look beyond the raw
numbers to the shades of
gray…discovering what the
data really mean.
Pearson’s Product Moment Correlation r
Is there a relationship between X and Y?
What is the magnitude of the relationship?
What is the direction of the relationship?
Connections and Disconnections
“To truly understand consumers’ motives

and actions, you must determine
relationships between what they think
and feel and what they actually do.”
David Singleton, vp of insights

Zyman Marketing Group
Scatterplots of
Relationships
Scatterplots
Plot of Forbes 500 Net Profits with Cash Flow
Diagram
of
Common
Variance
Interpretation of Correlations
X
X causes
causes Y
Y
Y
Y causes
causes X
X
X
X and
and YY are
are activated
activated by
by
one
one or
or more
more other
other variables
variables
X
X and
and YY influence
influence each
each
other
other reciprocally
reciprocally
Artifact
Correlations
Interpretation of Coefficients
A coefficient is not remarkable simply

because it is statistically significant!
It must be practically meaningful.
Comparison of
Bivariate
Linear
Correlation
and
Regression
Examples of Different Slopes
Concept Application
Y
X
Price per Case
Average Temperature (Celsius)
(FF)
12 2,000
16 3,000
20 4,000
24 5,000
Mean =18 Mean = 3,500
Plot of Wine Price by Average Temperature
Distribution of
Y for
Observation of
X
Wine Price
Study
Example
Least Squares
Line:
Wine Price
Study
Plot of Standardized Residuals
Prediction and Confidence Bands
Testing Goodness of Fit
Y is completely unrelated to X
and no systematic pattern is evident
There are constant values of

Y for every value of X
The data are related but

represented by a nonlinear function
Components of Variation
F Ratio in Regression
F Ratio in Regression
Coefficient of Determination: r2
Total proportion of
variance in Y explained by X
Desired r2: 80% or more
Chi-Square
Based
Measures
Proportional
Reduction of
Error
Measures
Statistical Alternatives for Ordinal Measures
Calculation of
Concordant (P),
Discordant (Q), Tied
(Tx,Ty), and Total
Paired Observations:
KeyDesign Example
Calculation of
Concordant (P),
Discordant (Q), Tied
(Tx,Ty), and Total
Paired Observations:
KeyDesign Example
Commonly Used
Measures of
Association
KDL Data for Spearman’s Rho
_______ _____ Rank By_____ _____

_____
Applicant Panel x Psychologist y d d2
1 3.5 6.0 -2.5 6.25
2 10.0 5.0 5.0 25.00
3 6.5 8.0 -1.5 2.52
4 2.0 1.5 .05 0.25
5 1.0 3.0 -2 4.00
6 9.0 7.0 2.0 4.00
7 3.5 1.5 2.0 4.00
8 6.5 9.0 -2.5 6.25
9 8.0 10.0 -2 4.00
10 5.0 4.0 1.0 _1.00_
57.00 .
Key Terms
18-38
 Artifact correlations  Concordant

 Bivariate correlation  Correlation matrix
analysis  Discordant
 Bivariate normal  Error term
distribution  Goodness of fit
 Chi-square-based measures  Lambda
 Contingency coefficient C  Linearity
 Cramer’s V  Method of least squares
 Phi
 Coefficient of determination
(r2)
Key Terms
18-39
 Ordinal measures  Regression coefficients

 Gamma  Intercept
 Somers’s d  Slope
 Spearman’s rho  Residual
 Pearson correlation  Scatterplot
coefficient  Simple prediction
 Prediction and  Tau
confidence bands  Tau b
 Proportional reduction in  Tau c
error (PRE)
 Regression analysis
Chapter 15
DISCUSSION OPPORTUNITIES
Snapshot:
Discovering Best Practices
Best distribution practices

for Latin America.
Slower growth
Rising Inflation.
Devaluation of currency
Small stores = 40% CPG
Snapshot:
35 CPGs doing business in

Latin America
Customer & Channel
Management Survey
Who was performing well
What practices used by

top performing companies
Snapshot:
Best Practice Companies

More evaluative criteria
More forward-looking criteria
Snapshot:
Best Practices
Power partnerships
Multifunctional service teams
Focus on sales generation
Pay staff more = lower turnover
Centrally manage price and promotion
Track competitor prices
Vary trade pricing
Responsive to local conditions
Snapshot:
Is Oscar Showing Some Tarnish
Does the Oscar have any measurable

effect on movie viewership?
Brief online survey via OmniPulse.
Event hype only a small influence.
Do women respond differently than men?
Snapshot:
Is Oscar Showing Some Tarnish
PicProfile: Constellation Wines
Recession affected
wine behavior.
Positioning research
using focus groups.
Word sorts to reveal how
Blackstone compared to other brands.
‘Masculine’ wasn’t threatening

but a strength.
Three ads by Amazon Advertising were

tested; “Count on it” strongest.
Snapshot:
Advanced Statistics Increase Satisfaction
& Release More Funds through ATMs
“A risk score model was embedded

in the daily transactions processing
system to automatically determine
how much cash each member can
withdraw from an ATM or receive
when making deposits.”
Snapshot: Envirosell
Does how you drive affect how you shop?
Envirosell tracked shoppers’ behaviors.
Observation studies.
Brits and Aussies shop as they drive.
Photo Attributions
Slide Source
7 ©McGraw-Hill Education
8 ©Thinkstock/Getty Images
14 © Andy Dean Photography / Alamy

41 ©Blend Images/Dave and Les Jacobs/Getty Images
45 ©Shutterstock / CREATISTA
47 ©Ariel Skelley/Getty Images
48 ©Onoky/SuperStock
49 ©Shutterstock / melis
Chapter 15

CH15 Measures of Association

Uploaded by

Copyright:

Available Formats

CH15 Measures of Association

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

CH15 Measures of Association

Uploaded by

Copyright:

Available Formats

Chapter 15

STAGE 4: MEASURES OF ASSOCIATION

“The human brain is an incredible

Jeff Bezos, CEO,

P-Q based; adjustment for table

P-Q based; asymmetrical

Phi Chi-square based for 2*2 tables

CS based; adjustment when one table

Lambda PRE based interpretation

PRE based with table marginals

Uncertainty coefficient Useful for multidimensional tables

Kappa Agreement measure

Burke, one of the world’s

Is there a relationship between X and Y?

What is the magnitude of the relationship?

What is the direction of the relationship?

“To truly understand consumers’ motives

David Singleton, vp of insights

A coefficient is not remarkable simply

Mean =18 Mean = 3,500

There are constant values of

The data are related but

_______ _____ Rank By_____ _____

 Artifact correlations  Concordant

 Ordinal measures  Regression coefficients

Best distribution practices

Small stores = 40% CPG

35 CPGs doing business in

Who was performing well

What practices used by

Best Practice Companies

Does the Oscar have any measurable

Brief online survey via OmniPulse.

Event hype only a small influence.

Do women respond differently than men?

‘Masculine’ wasn’t threatening

Three ads by Amazon Advertising were

“A risk score model was embedded

Does how you drive affect how you shop?

Envirosell tracked shoppers’ behaviors.

Brits and Aussies shop as they drive.

14 © Andy Dean Photography / Alamy

42 ©Blend Images/Dave and Les Jacobs/Getty Images

43 ©Blend Images/Dave and Les Jacobs/Getty Images

44 ©Blend Images/Dave and Les Jacobs/Getty Images

47 ©Ariel Skelley/Getty Images

You might also like

_____ _ Rank By_ ___