A Presenta*on from The NewMR Behaviour Economics Event 19 April 2012 Be#er Surveys through Behavioral Economics Jeffrey Henning, Affinnova Event sponsored by Greenbook All copyright owned by The Future Place and the presenters of the material For more informa>on about Greenbook visit www.greenbookblog.org For more informa>on about NewMR events visit newmr.org
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
A Presenta*on from The NewMR Behaviour Economics Event
19 April 2012
Be#er Surveys through Behavioral Economics Jeffrey Henning, Affinnova
Event sponsored by Greenbook All copyright owned by The Future Place and the presenters of the material
For more informa>on about Greenbook visit www.greenbookblog.org For more informa>on about NewMR events visit newmr.org
Speaker: Jeffrey Henning, Affinnova, US NewMR Behavioural Economics Event, 19 April 2012, Session 3
Better Surveys through Behavioral Economics
Jeffrey Henning Affinnova
Speaker: Jeffrey Henning, Affinnova, US NewMR Behavioural Economics Event, 19 April 2012, Session 3
Agenda
Surveys & BE
Framing Issues
Research on Research
Some Sugges>ons
Speaker: Jeffrey Henning, Affinnova, US NewMR Behavioural Economics Event, 19 April 2012, Session 3
Classical Surveys
Speaker: Jeffrey Henning, Affinnova, US NewMR Behavioural Economics Event, 19 April 2012, Session 3
Significant Overlap
Good Survey
Methodology
Good Behavioral Economics
Speaker: Jeffrey Henning, Affinnova, US NewMR Behavioural Economics Event, 19 April 2012, Session 3
Surveys & BE are Hard
• Even Dan Ariely gets it wrong
• Dan Ariely’s research uses a 101-point scale
• For unipolar scales, 5-point, fully labeled scales are the most reliable (Krosnick & Tahk)
Speaker: Jeffrey Henning, Affinnova, US NewMR Behavioural Economics Event, 19 April 2012, Session 3
3 Key BE Themes
Framing Heuristics Market Inefficiencies
Speaker: Jeffrey Henning, Affinnova, US NewMR Behavioural Economics Event, 19 April 2012, Session 3
Framing
Q: Name all of your favorite foods.
Q: You have an opportunity to go to the supermarket with an unlimited budget and buy all of your favorite foods. The catch is, you only have two minutes.
Source: Jon Puleston, GMI
Speaker: Jeffrey Henning, Affinnova, US NewMR Behavioural Economics Event, 19 April 2012, Session 3
They’ll Answer Anything
Speaker: Jeffrey Henning, Affinnova, US NewMR Behavioural Economics Event, 19 April 2012, Session 3
Why That Wine?
Speaker: Jeffrey Henning, Affinnova, US NewMR Behavioural Economics Event, 19 April 2012, Session 3
Trust Respondents’ Choices, Not Reasons
Speaker: Jeffrey Henning, Affinnova, US NewMR Behavioural Economics Event, 19 April 2012, Session 3
Predicting Share for Frozen Dinners
Top 4 selling SKUs
Top selling varieties (based on distribution
and facings)
Price (EDP, promotional price and activity)
Logo
Speaker: Jeffrey Henning, Affinnova, US NewMR Behavioural Economics Event, 19 April 2012, Session 3
Monadic vs. Discrete Choice
Predicted Market Share
Likert Scale Scores Discrete Choice Data
Speaker: Jeffrey Henning, Affinnova, US NewMR Behavioural Economics Event, 19 April 2012, Session 3
Frozen Dinners (73% category coverage)
Speaker: Jeffrey Henning, Affinnova, US NewMR Behavioural Economics Event, 19 April 2012, Session 3
Correlation Analysis
Monadic Correlation .76 DC Correlation .94
Boston Market
17%; 8.08
19%; 29.71
43%; 113.96
Amy's
HC Café Steamers HC Complete Selections
Hungry Man
Marie Callender's
Stouffers
LC Café Cuisine
LC Comfort Cuisine
LC Simple Favorites
LC Spa Cuisine
Smart Ones
Smart Ones Bistro
29%; 68.06
33%; 42.1134%; 47.32
31%; 31.6128%; 30.53
30%; 23.7133%; 22.99
26%; 16.09
36%; 65.46
33%; 54.46
0
20
40
60
80
100
120
10% 15% 20% 25% 30% 35% 40% 45% 50%
46; 8.08
203; 113.96
Amy's
Boston Market
HC Café Steamers
HC Complete Selections
Hungry Man
Marie Callender's
Stouffers
LC Café Cuisine
LC Comfort Cuisine
LC Simple Favorites
LC Spa Cuisine
Smart Ones
Smart Ones Bistro93; 16.09
121; 47.32
147; 54.46
165; 65.46135; 68.06
108; 22.9988; 23.71
92; 29.71 109; 31.61
111; 42.11
107; 30.53
0
20
40
60
80
100
120
40 60 80 100 120 140 160 180 200 220
Discrete Choice Index of Linear Utilities
Volum
e Sale
s (in
millio
ns)
Monadic Weighted T2Box Purchase Intent
Volum
e Sale
s (in
millio
ns)
Speaker: Jeffrey Henning, Affinnova, US NewMR Behavioural Economics Event, 19 April 2012, Session 3
Marie Callender's
46; 17%
165; 36%
203; 43%
Amy's
Boston Market
HC Café Steamers
HC Complete
Selections
Hungry Man
Stouffers
LC Café CuisineLC Comfort Cuisine
LC Simple FavoritesLC Spa Cuisine
Smart OnesSmart Ones Bistro
147; 33%
111; 33%108; 33%
109; 31%
121; 34%
107; 28% 135; 29%
88; 30%
93; 26%
92; 19%
10%
15%
20%
25%
30%
35%
40%
45%
50%
40 60 80 100 120 140 160 180 200 220
Pairwise Prediction
Correct Predictions
Pairings Monadic DC
Methods Agree 67 61 61 86% 91% 91%
Methods Disagree 11 1 10 14% 9% 91%
Total 78 62 71
79% 91%
Discrete Choice Index of Linear Utilities
Mona
dic W
eighte
d T2B
ox P
urch
ase I
ntent
Speaker: Jeffrey Henning, Affinnova, US NewMR Behavioural Economics Event, 19 April 2012, Session 3
Signal-to-Noise Analysis
Monadic DC WPI Raw Utility
Signal* StdDev across Pts. 5.6 46.3 Highest point 41 63
Lowest point 21 -59
Noise Standard Error 2.4 4.4 as % of signal range 12% 4%