Aiding Consumer Decisions

1
Aiding Consumer Decisions on the Web Gary
McClelland University of Colorado _at_
Boulder with assistance from Barbara Fasolo
Katharine Lange Presented at The Wharton
School University of Pennsylvania 26 February 2001
2
Todays Tour

• Consumer-Aiding Websites
• Winnowing via EBA, LEX, MCD, WADD, MAUT, etc.
• Why Attribute Correlations Matter
• Typical Attribute Correlations
• Effects of Attribute Correlations

3
Decision Guides
4
Decision Guides
5
MouseLab IDB (ca. 1990)
6
Consumer Reports
7
Consumer Reports Online

• Paid Subscriptions
• 532,000 in Feb 2001

8
IDBOptions x Attributes
www.decide.com
9
IDBAttributes x Options
www.point.com
10
IDB46 x 5
www.activebuyersguide.com
11
IDBContinued
www.activebuyersguide.com
12
IDBStill More
www.activebuyersguide.com
13
IDBMore Yet!
www.activebuyersguide.com
14
IDBThe End
www.activebuyersguide.com
15
Many, Many Options
www.personalogic.com
www.point.com
16
Aiding the Consumer

• Winnowing
• Comparing
• Evaluating
• Recommending
• Choosing

17
Company SitesNot Much Help
www.panasonic.com
18
Compared to What?
www.panasonic.com
19
Additional Info
www.panasonic.com
20
Compared to Other Phone?
www.panasonic.com
21
Automobile SitesThis is Decision Help?
22
IDBs Designed for Decision Making
Sort by Attribute, Eliminate Options, Choose
www.cdw.com
23
Decision Sites in Transition
24
Retail Sales 20004Q

• Online Up 36 to 8.7 billion
• Online 1.1 of total retail
• Total Up 5.4

Source Reuters, 16 Feb 2001
25
Todays Tour

• Consumer-Aiding Websites
• Winnowing via EBA, LEX, MCD, WADD, MAUT, etc.
• Why Attribute Correlations Matter
• Typical Attribute Correlations
• Effects of Attribute Correlations

26
Winnowing Options

• Setting Attribute Cutoffs (EBA)
• Sorting along Attributes (LEX,TB)
• Weighting Attributes (WADD)
• Measuring Tradeoffs (MAUT)

27
WinnowingLexicographic
28
WinnowingElimination by Aspects
29
WinnowingSatisficing
30
Winnowing Most Confirming Dimensions
31
WinnowingAdding (Equal Wts)
32
WinnowingImplications
33
Attribute Processing
EBA or LEX or TakeBest
34
Elimination-by-Aspects
www.point.com
35
Elimination-by-Aspects
www.point.com
36
EBAAre You Sure?
www.activebuyersguide.com
37
EBA LEX
www.decide.com
38
Option Processing
WADD or MAUT Score
39
WADDGetting the Wts
www.personalogic.com
40
WADDWeights Values
mro.frictionless.com
41
Weight Profiles
mro.frictionless.com
42
WADDOption Score
mro.frictionless.com
43
MAUTTradeoffs
www.activebuyersguide.com
44
MAUTGlobal
www.activebuyersguide.com
45
Collaborative Filtering

• www.amazon.com
• movielens.umn.edu
• www.imdb.com

46
Todays Tour

• Consumer-Aiding Websites
• Winnowing via EBA, LEX, MCD, WADD, MAUT, etc.
• Why Attribute Correlations Matter
• Typical Attribute Correlations
• Effects of Attribute Correlations

47
Positive Correlation Friendly Decision
r .21
48
Weight Insensitivity
3 X1 X2
X1 3 X2
r .21
49
Attribute Agreement
r .21 ? .24
50
Equal Weights History

• Wilks (1938)
• Gulliksen (1950)
• Dawes Corrigan (1974)
• Einhorn Hogarth (1975)
• Wainer (1976)
• Meehl (1999)

51
Equal Wts Correlations
52
Markets - Nondominated Options
r -.87
53
Weight Sensitivity!
3 X1 X2
X1 3 X2
r -.87
54
Nondominated Shapes
55
Equal Wts Value Loss
56
Attribute Disagreement
X1 X2 A E B
D C C D
B E A
r -.87 ? -1.0
57
Todays Tour

• Consumer-Aiding Websites
• Winnowing via EBA, LEX, MCD, WADD, MAUT, etc.
• Why Attribute Correlations Matter
• Typical Attribute Correlations
• Effects of Attribute Correlations

58
Real Choice Sets

• What are their attribute correlations?
• Are they approximately nondominated sets?
• Consumer Reports

59
Consumer Reports
60
Example CR Choice SetMtn Bikes
r -.82
61
-P/Q Correlations from CR
-.17 Air Conditioners -.27 Bike Helmets .16
Dishwashers -.82 Mtn Bikes -.74 Printers
-.49 Pro Ranges -.28 Fridge -.59 27 TVs -.16
Vacuums -.56 Wall Oven
62
-P/Q Correlations
63
Attribute Correlations from CR
.35 Air Conditioners .07 Bike Helmets -.03
Dishwashers .37 Mtn Bikes .18 Printers
.47 Pro Ranges -.06 Fridge .05 27 TVs .12
Vacuums .04 Wall Oven
64
Average r .05
65
Todays Tour

• Consumer-Aiding Websites
• Winnowing via EBA, LEX, MCD, WADD, MAUT, etc.
• Why Attribute Correlations Matter
• Typical Attribute Correlations
• Effects of Attribute Correlations

66
WebIDB
67
Sample Data Streams

• Att Opt Time
• 0 1 276
• 0 0 1595
• 1 0 851
• 0 0 859
• 0 1 836
• 0 0 535
• 0 1 975
• 0 2 652
• 0 3 652
• 0 4 543

• Att Opt Time
• 1 4 557
• 1 3 166
• 0 2 234
• 1 2 472
• 1 1 765
• 2 1 2111
• 2 2 2519
• 2 3 1031
• 2 4 448

Attribute Focus
68
WebIDB cf. MouseLab

• WebIDB replicated MouseLab results
• Attribute Focus is the default strategy
• Increasing Attributes - Attribute Focus
• Increasing Options - Less info, more var.
• Different result
• Somewhat more information viewed in WebIDB
• Somewhat greater attribute focus in WebIDB

69
Prior Research on Correlation Effects

• Johnson, Meyer Ghose (1989)
• Theory Negative Attribute-based
• Results Null

• Bettman, Johnson, Luce Payne (1993)
• Theory Negative Option-based
• Results Negative Option-based

70
Experiments

• Study 1
• 8 att x 5 opts
• Attribute Correlation Pos (.5) vs. Neg (-.14)
• 8 Matrices
• Between Subjects

• Study 2
• Within Subjects Switch after 4 Matrices

71
Proportion of Cells Visited
72
Attribute Visit S.D.
Selective Attention
Equal Attention
73
Payne Index (Opt-Att)/(OptAtt)
Option Focus
Attribute Focus
74
Self-Ratings
75
Switch Attribute Correlation
76
Results Summary

• Default Strategy is Attribute
  Processing
• Negative Correlation Option
  Processing
• Immediate Sensitivity to Correlation
• Quickly Switch to Option Processing
• Amount of Information Constant

77
Research Questions

• What winnowing strategies do consumers use?
• Attribute-based unless forced towards
  Option-based by negative attribute correlations

78
Research Questions

• What winnowing strategies might consumers be
  willing to use if aided?
• And how do attribute correlations affect the use
  of such aids?

79
Correlations and EBA
80
Correlations and WADD
81
End of Tour

• Consumer-Aiding Websites
• Winnowing via EBA, LEX, MCD, WADD, MAUT, etc.
• Why Attribute Correlations Matter
• Typical Attribute Correlations
• Effects of Attribute Correlations