Behavioral Science
Prof. Dr. Dan Schley
1
A BIT ABOUT ME…
– CREDENTIALS:
• EDUCATION:
• PhD and MA (Quantitative Psychology)
• BA (Economics)
• EXPERIENCE:
• Sales
• Consult for companies and the US government
2
A BIT ABOUT ME…
3
A BIT ABOUT ME…
– CREDENTIALS:
• EDUCATION:
• PhD and MA (Quantitative Psychology)
• BA (Economics)
• EXPERIENCE:
• Sales
• Consult for companies and the US government
4
A BIT ABOUT ME…
– CREDENTIALS:
• EDUCATION:
• PhD and MA (Quantitative Psychology)
• BA (Economics)
• EXPERIENCE:
• Sales
• Consult for companies and the US government
5
“… wrote a series of breathtakingly original
studies undoing our assumptions about the
[human] decision-making process...
Their work created the field of behavioral
economics, revolutionized Big Data studies,
advanced evidence-based medicine, led to a
new approach to government regulation,...
They became one of the greatest partnerships
in the history of science, working together so
closely that they couldn’t remember whose
brain originated which ideas, or who should
claim credit.”
Daniel Kahneman Amos Tversky
Nobel Prize (2002) 1937 - 1996
Why are we here?
• Better understand your own judgments and
decisions.
• Better predict and influence behavior.
• Improve your success as decision maker (manager,
entrepreneur, technocrat, leader, consumer, private
individual).
• Learn a lot of interesting things about people.
10
CONSUMER PUZZLES
11
CONSUMER PUZZLES
Apple
Quantity Sold
Product Price
12
How will we achieve these objectives?
• Discover some systematic biases when judging and
choosing (i.e., the phenomena)
• Understand fundamental psychological principles
behind these (i.e., the explanation).
• Learn about behavioral experiments (i.e., the
method)
13
How does behavioral science relate
economics?
• Theoretical assumptions:
– normative: optimize (rational)
– descriptive: describe and backwards engineer
• Different data:
real-world correlational vs. laboratory experimental
• Complimentary strengths and weaknesses:
– Unifying theory, but inferences to fit data to theory
– Many micro-theories, yet evidence to refute theory
• From antipathy to synergy:
behavioral economics, behavioral… 14
Nobel Prizes in economics for behavioral
science
• 1978 Herbert Simon "for his pioneering
research into the decision-making
process within economic organizations.”
• 2002 Daniel Kahneman “for having
integrated insights from psychological
research into economic science”
• 2017 Richard Thaler “for his contributions
to behavioural economics.”
15
If you are interested:
If you are interested:
Course logistics (1)
• Readings
– Book & readings on course webpage
– First lecture, then readings
• Slides
– outline, not a textbook
– take notes, but not verbatim
18
Course logistics (2)
• Classroom etiquette - respect your peers:
– Do not record in class
– If content on your monitor distracts peers, sit in the back
– If you arrive late, sneak into a seat on the side
• Contact (see course outline)
– email teaching assistant
– book my office hour directly via email
• Grade: Exam based on readings and lecture
19
Do not think of the slides as
lecture notes!
20
SOME REAL TALK…
COURSE EXPECTATIONS
21
If you are interested:
Behavioral Marketing Curriculum
([Link]/de/marketing/courses-masters-thesis/)
Mandatory for Thesis in
Optional Courses
Behavioral Marketing
1 Bachelor Course:
Behavioral Science (Schley) Applied Statistics for the Behavioral Sciences
(hands on analyses in R)
HS 1 2
Consumer Behavior: Consumer Behavior:
Theoretical Foundations (Schley) Marketplace Applications (Laran)
3
Psychological Theory in Consumer
Behavior (Brendl)
FS 1 4
Master Seminar Behavioral
Marketing (Brendl)
summer/ HS 2 Thesis or during HS 2
22
Course Overview
I. Introduction
II. Judgment by Heuristics
III. Psychological Utility
IV. Making Decisions
23
Today’s Overview
• Opening example to motivate the topic
• Introduction to experimental design
24
Your device showed you these instructions:
Imagine that you are looking for a new opening-
screen image for your computer. Please look at each
of the following six pictures. We will present them for
about 15 seconds each. For each picture please…
Your device then continued with either…
A: … or B: …
I will now show you the images.
Follow the instructions you have just read.
Source: cf. Wilson et al. (1993), PSPB
Now, imagine you would actually
choose one image.
32
Imagine that you are looking for a new opening-
screen image for your computer. Please look at each
of the following six pictures. We will present them for
about 30 seconds each. For each picture please…
A: Overall Impression B: Reasons
…form a general …describe the reasons
impression of how much why you like or dislike
you would like each one each picture for your
for your computer computer screen. Write
screen. down these reasons.
Source: cf. Wilson et al. (1993), PSPB
Procedure
Random Assignment to One of Two Between-Subjects
Experimental Conditions
Overall Impression Reasons
… Imag Imag Image
Image I …
e VI eI VI
Choose one of six images
34
Procedure
Random Assignment to One of Two Between-Subjects
Experimental Conditions
Overall Impression Reasons
Imag … Imag Imag Imag
…
eI e VI eI e VI
Choose one of six images
Wait two weeks
Rating: How satisfied are you with the image you chose?
very dissatisfied (-3) (-2) (-1) ( 0) ( 1) ( 2) ( 3) very satisfied
35
Summary and conclusions
• In a similar experiment by Wilson et al. (1993), which
group was more satisfied?
Overall Impressions Reasons
• …
36
Summary and conclusions
• In a similar experiment by Wilson et al. (1993), which
group was more satisfied?
Overall Impressions Reasons
• Two kinds of reasoning: intuitive versus analytical
– which “Thinking Fast and Slow” alludes to
• Why this result?
• Is either kind of reasoning better?
• How do these kinds of reasoning differ?
37
Today’s Overview
• Opening example to motivate the topic
• Introduction to experimental design
38
Why do experiments? (1)
39
Why do experiments? (2)
Correlation (A, B) Causations
A B
A B
C A
B
40
Does a paid search online-advertisement
increase sales?
Paid Links/
Search Ads
Organic Links/
Search Results
42
Does a paid search online advertisement
increase sales?
times ad is shown per person
43
Correlational Approach
like books don’t like books invisible
visible
many ad
few ad servings
servings
buy don’t buy
44
CORRELATION AND CAUSATION
• BLAH, BLAH, CORRELATION ≠ CAUSATION, BLAH
• REICHENBACH COMMON CAUSE PRINCIPLE (1956)
Mediator
Female Lead Male Lead
Sick Sick
45
CORRELATION AND CAUSATION
• BLAH, BLAH, CORRELATION ≠ CAUSATION, BLAH
• REICHENBACH COMMON CAUSE PRINCIPLE (1956)
Common Cause
Female Lead Male Lead
Sick Sick
46
Experiment Approach
like books don’t like books like books don’t like books
show ad do not show ad
Buy don’t buy
47
Why do experiments? (3)
Correlation (A, B) Causations
A B
A B
C A
B
Warning:
Constellation
C1, C2,…Cn A 48
Why should we care about causation?
Intervention
Prediction
Explanation
49
THE “EXPERIENTIAL ADVANTAGE”
50
EXAMPLE: EXPERIENCES MAKE US HAPPY?
51
THE MODEL
Experiences/ Happiness
Material
52
DIRECTED ACYCLIC GRAPH OF THE MODEL
Treatment Outcome
𝑈𝐴 (Y) 𝑈𝑦
(A)
𝑋෨ 𝑌෨
𝑐𝑜𝑣 𝑈 = 𝐼2
53
WHAT’S “IN” AN EXPERIMENT?
Remember
material/
experiential Rated
purchases Happiness
𝑈𝐴 𝑈𝑦
𝑋෨ 𝑌෨
𝑐𝑜𝑣 𝑈 = 𝐼2
54
WHAT’S “IN” AN EXPERIMENT?
Present
“material” or
“experiential” Rated
products Happiness
𝑈𝐴 𝑈𝑦
𝑋෨ 𝑌෨
𝑐𝑜𝑣 𝑈 = 𝐼2
55
WHAT’S “IN” AN EXPERIMENT?
Outcome
𝑈𝐴 Treatment 𝑈𝑦
(Y)
(A)
𝑋෨ 𝑌෨
56
WHAT’S “IN” AN EXPERIMENT?
Outcome
(Y) 𝑈𝑦
𝐴
𝐵𝑖𝑜𝑡𝑖𝑛 𝑌෨
𝐵6 𝐵12
𝐶 𝐷3
57
WHAT’S “IN” AN EXPERIMENT?
Experiences Material Outcome 𝑈𝑦
(Y)
𝑀𝑒𝑚𝑜𝑟𝑖𝑒𝑠
𝐹𝑢𝑛𝑔𝑙𝑖𝑏𝑙𝑒 𝑌෨
𝑆ℎ𝑎𝑟𝑒𝑑 𝑂𝑝𝑝𝑜𝑟𝑡𝑢𝑛𝑖𝑡𝑦
𝐶𝑜𝑠𝑡𝑠
𝑆𝑜𝑐𝑖𝑎𝑙 𝐹𝑢𝑛𝑐𝑡𝑖𝑜𝑛𝑎𝑙
58
Some experimental design terminology
59
Types of Reasons Experiment Procedure
Random Assignment to One of Two Between-Subjects
Experimental Conditions
Overall Impression Reasons
Image … Image Image Image
…
I VI I VI
Choose one of six images
Wait two weeks
Rate satisfaction with choice.
60
Procedure
Independent Variable #1 (manipulated)
Random Assignment to One of Two Between-Subjects
Experimental Conditions
Overall Impression Reasons
Image IndependentImage
VariableImage
#2 (manipulated) Image
… …
I VI I VI
Dependent Variable #1 Choose one of six images
Wait two weeks
Dependent Variable #2
Rate satisfaction with choice.
61
Experimental Variables
Independent Variables (IVs)
between- Impression: overall impression vs. reasons
subjects
within- six different images
subjects
Dependent Variables (DVs)
Choice of one image
Satisfaction Ratings
62
Experimental Design
Overall Impression Reasons
12 experimental conditions
2x6 mixed orthogonal design of Type of Impression (overall impression vs.
reasons) x Image (six images), manipulated between-subjects and within-
subjects, respectively; and with two dependent variables (DV’s): choice of one
image and satisfaction ratings (7-point scale from -3 to +3).
• “Mixed” refers to mixing between- and within-
subjects independent variables (IV’s).
63
Which conclusions can we draw from the
results?
• Did the reasoning manipulation have a causal
influence on ratings of satisfaction?
• Did the image that participants chose have a causal
influence on ratings of satisfaction? (Why should we
care about this question?)
• Does the experiment tell us why respondents in the
two reasoning conditions made different ratings of
satisfaction?
64
Causal description and causal
explanation
65
Source: Shadish, Cook, & Campbell (2002)
Causal description and causal
explanation
Causal explanation: closes electric circuit
➞ electricity flows through bulb
Causal description: flick switch ➞ lights bulb
66
Causal description and causal
explanation
Causal explanation: mechanism,
hypothesis, theory,
not observable,
renders effect unsurprising
Causal description: observed cause-effect in
experiment
67
Hypothesis and prediction
Causal explanation entails hypotheses.
-“universal” (Popper, 1959)
/ abstract
Causal description is a prediction,
if you have not yet conducted
the experiment.
-specific / observable
68
For details, see Calder, Brendl, Tybout, & Sternthal (2021)
Hypothesis, explanation, and prediction
Hypothesis
explained by predicts
past observations future observations
69
For details see Calder, Brendl, Tybout, & Sternthal (2021)
Session Summary
• Experiment:
– Why do it?
– What is it?
– Types of experiment variables, random assignment
• Theory of science: Hypothesis, prediction, explanation
70
Homework
• Read:
• Snoddgrass et al. (1985): pp. 18-30
• Shadish, Cook, & Campbell (2002): pp: 3-12
• Optional reading:
– Introduction of Thinking Fast and Slow
71
Course Overview
I. Introduction
1. Introduction and Experimental Design
2. Fundamentals of Cognition
3. System 1 versus System 2
4. System 1 Preference Construction
II. Judgment by Heuristics
III. Psychological Utility
IV. Making Decisions 72
