Chapter 18

MEASURES OF ASSOCIATION

McGraw-Hill/Irwin Copyright 2014 by The McGraw-Hill Companies, Inc. All rights reserved.
Pearsons 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?

18-2
Measures of Association:
Interval/Ratio Data
Pearson correlation
coefficient
Correlation ratio (eta)
Biserial
Partial correlation
Multiple correlation
Bivariate linear regression
For continuous linearly related
variables
For nonlinear data or relating a
main effect to a continuous
dependent variable
One continuous and one
dichotomous variable with an
underlying normal distribution
Three variables; relating two with
the thirds effect taken out
Three variables; relating one
variable with two others
Predicting one variable from
anothers scores
18-3
Measures of Association:
Ordinal Data

Based on concordant-discordant
Gamma pairs; proportional reduction in
error (PRE) interpretation
P-Q based; adjustment for tied
Kendalls tau b
ranks

P-Q based; adjustment for table

Kendalls tau c
dimensions

P-Q based; asymmetrical

Somerss d
extension of gamma
Product moment correlation for
Spearmans rho
ranked data
18-4
Measures of Association:
Nominal Data

Phi Chi-square based for 2*2 tables

CS based; adjustment when one table
Cramers V
dimension >2
CS based; flexible data and distribution
Contingency coefficient C
assumptions

Lambda PRE based interpretation

PRE based with table marginals

Goodman & Kruskals tau
emphasis
Uncertainty coefficient Useful for multidimensional tables

Kappa Agreement measure

18-5
Scatterplots
of
Relationships

18-6
Scatterplots

18-7
Plot of
Forbes 500
Net Profits
with Cash
Flow

18-8
Diagram of Common Variance

18-9
Interpretation of Correlations

X causes Y

Y causes X

X and Y are activated by

one or more other variables

X and Y influence each

other reciprocally
18-10
Artifact
Correlations

18-11
Interpretation of Coefficients

A coefficient is not remarkable

simply because it is statistically
significant!
It must be practically meaningful.

18-12
Comparison
of Bivariate
Linear
Correlation
and
Regression

18-13
Examples of Different Slopes

18-14
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

18-15
Plot of Wine Price by Average
Temperature

18-16
Distribution
of Y for
Observation
of X

18-17
Wine
Price
Study
Example

18-18
Least
Squares Line:
Wine Price
Study

18-19
Plot of Standardized Residuals

18-20
Prediction and Confidence Bands

18-21
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

18-22
Components of Variation

18-23
F Ratio in Regression

18-24
F Ratio in Regression

18-25
Coefficient of Determination: r2

Total proportion of variance in Y

explained by X
Desired r2: 80% or more

18-26
Chi-
Square
Based
Measures

18-27
Proportional
Reduction of
Error
Measures

18-28
Statistical Alternatives for Ordinal
Measures

18-29
Calculation of
Concordant (P),
Discordant (Q),
Tied (Tx,Ty),
and Total Paired
Observations:
KeyDesign
Example

18-30
KDL Data for Spearmans 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 .

18-31