Download - Social Learning and Consumer Demand
Social Learning and Consumer Demand
Markus Mobius (Harvard University and NBER)Paul Niehaus (Harvard University)Tanya Rosenblat (Wesleyan University and IAS)
CMPO, 2 June 2006
Motivation
We want to study social learning in the context of how consumer preferences form.
How strong are social learning effects absolutely and relatively compared to informative advertising?
How strong are social influence effects (on valuations) absolutely and relatively compared to persuasive advertising?
Which agents are influential?
Learning Persuasion
Strong Social Learning
Agents communicate directly about the product, sharing factual information:
“I didn’t buy it because it’s not Mac compatible”
“I’ve heard Sony makes the most reliable ones”
“They have a lot of vegetarian dishes on the menu”
Learning Persuasion
Strong Social Learning
Weak Social Learning
Agents observe their friends’ consumption decisions and enjoyment of products and make inferences about the products’ attributes.
“Greg got one for Christmas and I know he really liked it”
These inferences should be sharper when friends know their friend’s preferences well.
Learning Persuasion
Strong Social Learning
Weak Social Learning
Social Influence
Agents observe their friends’ consumption decisions and....
• Their private tastes are altered
• The status value of consuming the product is altered
Learning Persuasion
Strong Social Learning
Weak Social Learning
Social Influence
Persuasive Advertising
Informative Advertising
Agents observe advertising for the product. They may learn about objective features of the product or be persuaded to like it or be persuaded of its prestige value.
Methodology: basic paradigm
Stage 1: Measure the network (Harvard Undergraduates)
Stage 2: Distribute actual products and track social learning
Methodology
Measuring the Social Network
Measuring the Network
Rather than surveys, agents play in a trivia game
Leveraged popularity of www.thefacebook.comMembership rate at Harvard College over
90% *95% weekly return rate *
* Data provided by the founders of thefacebook.com
home search global social net invite faq logout
quick search go
sponsor
• offensive? tell us.• announcesomething
My Profile [ edit ]My FriendsMy GroupsMy PartiesMy MessagesMy AccountMy Privacy
Work from bed!
(Or desk, or kitchen)
Write short abstractsand earn royalties
www.shvoong.com
Paul Niehaus' Profile (This is you) Har
Picture [ edit ]
Visualize My Friends
Edit My Profile
My Account Prefer ences
My Privacy Preferences
Connection
This is you.
Access
Paul is currently logged in from a non-residential location.
Friends at Harvard [ edit ]
Paul has 80 friends at Harvard.[ see all of them ]
RohitChopra
Anna ByrneRussellAnello
ShannonChristmas
Zach LazarDaniel
Morales
Other Schools [ edit ]
Information [ ed
Account Info:
Name: Paul Niehaus
Member Since: May 18, 2004
Last Update: June 6, 2005
Basic Info: [ ed
School: Harvard '04
Geography: Boston, MA
Status: Grad Student
Sex: Male
Concentration: Economics
Birthday: 03/11/1982
High School: St. John's Prep '00
Contact Info: [ ed
Email:
Screenname: pfn007
Mobile: 508.335.5242
Website: http:/ /www.people.fas.harvard.edu/~nieha
Personal Info: [ ed
Relationship Status: In a Relationship with
Lauren Young (Berkeley)
Interests: visiting / talking to / daydreaming aboutLauren Young
Clubs and Jobs: Americans for Being Awesome
Favorite Music: donkey kong count ry II soundtrack
Favorite Books: The Bible, Development as Freedom, LOTRThe Screwtape Letters , Moneyball, MWG!
Favorite Movies: Kindergarten Cop , Office Space, Friday,Good Will Hunting, Pumping Iron 20thanniversary edition, pretty much any othermovies with Ahnold except Junior, Dr.Strangelove, Kujo's happy bi rthday movie
Favorite Quote: good advice I have received from friends:
"it'll be snowy and cold tomorrow, so kee pwarm and avoid slipperiness."- Yi Qian
"you should have proposed toa heterosexwoman."- Michael Baldwin
"go to grad school. I went, an d I loved it."- Elhanan Helpman
Summer Plans [ ed
Job/Activity: hanging out with Lauren
Location: Cambridge, MA, 02140
Additional Info:also catc
• Markus
• His Profile
• (Ad Space)
• His Friends
Trivia Game: Recruitment
1. On login, each Harvard undergraduate member of thefacebook.com saw an invitation to play in the trivia game.
2. Subjects agree to an informed consent form – now we can email them!
3. Subjects list 10 friends about whom they want to answer trivia questions.
4. This list of 10 people is what we’re interested in (not their performance in the trivia game)
Trivia Game: Trivia Questions
1. Subjects list 10 friends – this creates 10*N possible pairings.
2. Every night, new pairs are randomly selected by the computer
Example: Suppose Markus listed Tanya as one of his 10 friends, and that this pairing gets picked.
Trivia Game Example
a) Tanya (subject) gets an email asking her to log in and answer a question about herself
b) Tanya logs in and answers, “which of the following kinds of music do you prefer?”
Trivia Game Example (cont.)
c) Once Tanya has answered, Markus gets an email inviting him to log in and answer a question about one of his friends.
d) After logging in, Markus has 20 seconds to answer “which of the following kinds of music does Tanya prefer?”
Trivia Game Example (cont.)
e) If Markus’ answer is correct, he and Tanya are entered together into a nightly drawing to win a prize.
Trivia Game: Summary
Subjects have incentives to list the 10 people they are most likely to be able to answer trivia questions about
This is our (implicit) definition of a “friend” This definition is suited for measuring social learning
about products. Answers to trivia questions are unimportant
ok if people game the answers as long as the people it’s easiest to game with are the same as those they know best.
Roommates were disallowed 20 second time limit to answer On average subjects got 50% of 4/5 answer multiple choice
questions right – and many were easy
Recruitment
In addition to invitations on login, Posters in all hallways Workers in dining halls with laptops to step through
signup Personalized snail mail to all upper-class students Article in The Crimson on first grand prize winner
Average acquisition cost per subject ~= $2.50
Network Data
23,600 links from participants 12,782 links between participants 6,880 of these symmetric (3,440 coordinated friendships)
Similar to 2003 results Construct the network using “or” link definition
5576 out of 6389 undergraduates (87%) participated or were named
One giant cluster Average path length between participants = 4.2 Cluster coefficient for participants = 17%
Lower than 2003 results – because many named friends are in different houses
Number of Roommate links, friend (N1), indirect friend (N2), and friends of distance 3 (N3) for an average subject (OR network on all participants of trivia game)Type of Link Number of
LinksRatio
Roommate .96 1
N1 7.68 8
N2 57.91 60.32
N3 347.14 361.6
Methods in Comparison
2003 House Experiment in 2 undergraduate houses
Email-data: Sacerdote and Marmaris (QJE 2006)
Mutual-friend methods with facebook data? (Glaeser et al, QJE 2000)
Methodology
Seeding Information
Seeding Information
1. Elicit subjects’ initial valuations Center empirical estimates Decompose valuations (hedonics)
2. Randomized treatments Distribute product samples Information / instructions
3. Randomized advertising Print (Crimson) and online
(thefacebook.com) Informative and persuasive
4. Elicit subjects’ final valuations
Example
A hypothetical subject “Paul” might be exposed to the following treatments: A friend of Paul’s of social distance 2 used a PDA The friend was told about the PDA’s instant
messenger capabilities Paul saw an advertisement for the PDA in the
newspaper that emphasized it’s hip-ness Paul did not see online advertising for the PDA
Product Samples
We want new products to maximize the potential for social learning.
Want to vary products byLikely demographic appealPotential for strong learning (need a manual?)Potential for weak learning and social
influence – the “buzz factor”
DurablesT-Mobile Sidekick II
Philips Key019 Digital Camcorder
Philips ShoqBox
Perishables Student Advantage Discount Card
Qdoba Meal Vouchers
Baptiste Studios Yoga Vouchers
Step I: Elicit Valuations
We want to elicit valuations for a product without telling subjects what the product is.
Our solution: We treat a product as a vector of attributes which span a space containing the specific product.
We can elicit valuations for each attribute without revealing product.
Step I: Configurators
Familiar examples with posted menus of pricesmany computer manufacturers (e.g. Dell)some car manufacturers
Here, subjects bid for featuresBaseline bid for “featureless” product Incremental bids for distinct features
Constructed Bids Subjects told that either this bid or their bid in the
followup will be entered into a uniform-price auction with equal probability
Construction:
Incentives: bid as accurately as possible Extension: interactions between features
Potential additional featuresfor this product include:
Amtrak discounts: studentdiscounts on Amtrak trains.
Textbook discounts: ontextbook purchases atBarnes&Noble.com
Greyhound discounts:student discounts onGreyhound trains.
Online guides: websiteprovides a guide todiscounts by product typeand by city.
Clothing discounts: studentdiscounts at UrbanOutfitters.
14
Baseline bid for StudentDiscount Card
Textbook discounts
6
Clothing discounts
12
Greyhound discounts
0
Amtrak discounts
0
Online guides
0
Feature descriptions
Baseline bid
Feature bids
Card Yoga Food Camcorder ShoqBox Sidekick
-50
05
01
00
15
02
00
25
03
00
Distributions of Imputed Bids
$
($20) ($50) ($35) ($150) ($150) ($250)(Price)
Distributions of Imputed Bids
Results from configurators look sensible In each case, market prices lie between
median bid and upper tailT-Mobile and Philips confirmed that demand
curves for their products are similar to results from more traditional analysis
Step 2: Randomized Product Trials
Perishables½ year Student Advantage cards5 yoga vouchers5 meal vouchers
DurablesTry out for approximately 4 weeks during end
of term
Randomization
Blocked by year of graduation, gender, and residential house
Email invitations to come pick up samples
Invitation times varied to vary strength of exposure (April 26th – May 3rd)
Info Treatments
Varied information communicated verbally by workers doing distribution
Information treatments correspond to product features in our configurators (5 or 6 features for each product).
Reinforced this information treatment with reminder emails
Each treatment given with 50% probability to each subject
“Buzz” Treatments
Product-specific treatments without information content
Intended to increase subject’s enjoyment of the product
Examples Subway tokens for yoga, Qdoba 5 free MP3s on ShoqBox Extra pre-paid balance on Sidekicks Special one-store subsidy on Student Advantage
cards Given with 50% probability to each subject
home search global social net invite faq logout
quick search go
sponsor
• offensive? tell us.• announcesomething
My Profile [ edit ]My FriendsMy GroupsMy PartiesMy MessagesMy AccountMy Privacy
Work from bed!
(Or desk, or kitchen)
Write short abstractsand earn royalties
www.shvoong.com
Paul Niehaus' Profile (This is you) Har
Picture [ edit ]
Visualize My Friends
Edit My Profile
My Account Prefer ences
My Privacy Preferences
Connection
This is you.
Access
Paul is currently logged in from a non-residential location.
Friends at Harvard [ edit ]
Paul has 80 friends at Harvard.[ see all of them ]
RohitChopra
Anna ByrneRussellAnello
ShannonChristmas
Zach LazarDaniel
Morales
Other Schools [ edit ]
Information [ ed
Account Info:
Name: Paul Niehaus
Member Since: May 18, 2004
Last Update: June 6, 2005
Basic Info: [ ed
School: Harvard '04
Geography: Boston, MA
Status: Grad Student
Sex: Male
Concentration: Economics
Birthday: 03/11/1982
High School: St. John's Prep '00
Contact Info: [ ed
Email:
Screenname: pfn007
Mobile: 508.335.5242
Website: http:/ /www.people.fas.harvard.edu/~nieha
Personal Info: [ ed
Relationship Status: In a Relationship with
Lauren Young (Berkeley)
Interests: visiting / talking to / daydreaming aboutLauren Young
Clubs and Jobs: Americans for Being Awesome
Favorite Music: donkey kong count ry II soundtrack
Favorite Books: The Bible, Development as Freedom, LOTRThe Screwtape Letters, Moneyball, MWG!
Favorite Movies: Kindergarten Cop, Office Space, Friday,Good Will Hunting, Pumping Iron 20thanniversary edition, pretty much any othermovies with Ahnold except Junior, Dr.Strangelove, Kujo's happy bi rthday movie
Favorite Quote: good advice I have received from friends:
"it'll be snowy and cold tomorrow, so kee pwarm and avoid slipperiness."- Yi Qian
"you should have proposed toa heterosexwoman."- Michael Baldwin
"go to grad school. I went, an d I loved it."- Elhanan Helpman
Summer Plans [ ed
Job/Activity: hanging out with Lauren
Location: Cambridge, MA, 02140
Additional Info:also catc
Step 2: Advertising
Delivered via thefacebook.com
Mixed in with normal paid advertising
65% of subjects saw ads 232,736 impressions
(approx. 300 per treated subject)
136 clicks (in line with averages)
Online Advertising
Advertising Content Content from sponsor
companies Tweaked to vary
informational content in line with product features
Also non-informative versions
Step 2: Advertising
Inlets in The Crimson, Harvard’s student newspaper
One of nation’s largest student papers, daily readership approx. 14,000
Delivered to undergrad students’ rooms Inlets allow randomization across
residential houses
Print Advertising
All ads for a product has the same styleand differed only in the informational content.
Print advertising
4 inlets with two ads each.
3 ads emphasizing a single feature of a product.
Residents in a house were exposed to either 2 or 3 impressions of the same print ad.
Step 4: Final Valuations
Subjects receive full product descriptions and submit a second round of bids, which go into the auctions with 50% probability
Subjects also… Predict what the average bid will be Predict what a sample of their friends will bid in the
auction Answer factual questions about each product Indicate their confidence in these answers
Facebook Experiment
First Product
Personal Sound Systemwith MP3 players
Time left: 46
This product is a Personal Sound System,an MP3 player with two inbuilt speakers loudenough to fill a room. It is small enough to fitin your pocket and you can upload songsdirectly from your computer.
Please submit your bid for this product:______ Dollars
You can increase your earnings by 50 cents if youranswer to the following question is not more than10 percent off.
What is your best guess for the averagebid of all other participants?: ______ Dollars
Facebook Experiment
First Product
Personal Sound System with MP3 playersFor each of the following students please predict how they will bid in the auction. For each student if your answer is within10 percent of their true bid we will add10 cents to your earnings.
Danielle Sassoon (FR, Canaday) ______ Dollars Skyler Johnson (FR, Canaday) ______ Dollars
Rachel Thornton (FR, Canaday) ______ Dollars Danny Mou (FR, Canaday) ______ Dollars
Eliciting Confidence Levels
Meet “Bob the Robot” and his clones Bob 1 – Bob 100
Subjects are randomly paired with an (unknown) Bob
Subjects indicated a “cutoff Bob” at which they are indifferent about who should answer the question
If assigned Bob is better than the cutoff, Bob answers the question; otherwise we use subject’s answer
Incentive-compatible mechanism to elicit subject’s belief that he/she will get the question right
Facebook Experiment
Second Product
T-Mobile Sidekick IITime left: 36
How confident are you that you can answer some YES/NOquestions about this product correctly?Your confidence: ______ percent
You can increase your earnings by 50 cents if your answer to the followingquestion is not more than 10 percent off.
Please estimate the average confidence of other participants in thisstudy to answer some YES/NO question about this product correctly?______ percent
Next Page >>
Facebook Experiment
Second Product
T-Mobile Sidekick IITime left: 84
Question 1Does the Sidekick include AOL messenger?
YES NO
Your confidence:______ percent
Question 2Does the Sidekick have a color screen?
YES NO
Your confidence:______ percent
Question 3Does the Sidekick have 10 or more hours of batterylife?
YES NO
Your confidence:______ percent
Question 4Does the Sidekick have a QWERTY keyboard?
YES NO
Your confidence:______ percent
Question 5Does the Sidekick include a camera?
YES NO
Your confidence:______ percent
Question 6Does the Sidekick use the Pocket PC OS?
YES NO
Your confidence:______ percent
Analysis
Measuring Learning
Analysis
Stage I: Check whether info and ad treatments affected a subject’s knowledge.
Stage II: Use info treatments as instruments to measure social learning.
Analysis
Stage I: Check whether info and ad treatments affected a subject’s knowledge.
Product Group (PG) – Likelihood of answering a question about a feature correctly if primed about that feature at distribution
Non-Product Group (NPG) – Likelihood of answering a question about a feature correctly if exposed to informative advertising about that feature
Stage I: Effect of Info Treatments on Knowledge (PG)
Stage I: Effect of Info Treatments on Knowledge (PG)
85.294.2
Stage I: Effect of Info Treatments on Knowledge (PG)
85.294.2
Subjects who received a product and were primed on a Feature are about 9% more likely to answer the question about the feature correctly.
Stage I: Info-TreatmentsFCONFIDENCE FCORRECTANSWER
(1) (2) (3) (4) (5) (6)
NUMTREATED .748*
(.373)
.766
(.505)
.007
(.007)
.007
(.007)
FTREATED 7.057*
(.825)
7.087*
(.825)
7.080*
(.825)
.082**
(.015)
.083**
(.014)
.085**
(.014)
Intercept 85.468*
(1.065)
85.361*
(1.065)
85.645*
(1.065)
.838**
(.019)
.837**
(.021)
.856**
(.010)
Fixed effects None RE FE None RE FE
N 1927 1927 1927 1930 1930 1930
R2 .054 .056 .058 .022 .023 .022
Significance Levels: *: 5% **: 1%
Stage I: Info-TreatmentsFCONFIDENCE FCORRECTANSWER
(1) (2) (3) (4) (5) (6)
NUMTREATED .748*
(.373)
.766
(.505)
.007
(.007)
.007
(.007)
FTREATED 7.057*
(.825)
7.087*
(.825)
7.080*
(.825)
.082**
(.015)
.083**
(.014)
.085**
(.014)
Intercept 85.468*
(1.065)
85.361*
(1.065)
87.645*
(1.065)
.838**
(.019)
.837**
(.021)
.856**
(.010)
Fixed effects None RE FE None RE FE
N 1927 1927 1927 1930 1930 1930
R2 .054 .056 .058 .022 .023 .022
Significance Levels: *: 5% **: 1%
Both confidence and knowledge increases with info treatments.
Stage I: Effect of Online Ad on Knowledge (NPG)
Effect of online ads on subjects who did not receive products or print ads.
Stage I: Effect of Online Ad on Knowledge (NPG)
Effect of online ads on subjects who did not receive products or print ads.
64.7 %
73.5 % 71.0 %
Stage I: Effect of Online Ad on Knowledge (NPG)
Effect of online ads on subjects who did not receive products or print ads.
64.7 %
73.5 % 71.0 %
Subjects who received online ads are about 5-8% more likely to answer the question about the feature correctly.
Stage I: Effect of Print Ad on Knowledge (NPG)
Effect of print ads on subjects who did not receive products or online ads.
Stage I: Effect of Print Ad on Knowledge (NPG)
64.8%71.3%
79.8%
Effect of print ads on subjects who did not receive products or online ads.
Stage I: Effect of Print Ad on Knowledge (NPG)
64.8%71.3%
79.8%
Effect of print ads on subjects who did not receive products or online ads.
Subjects who received print ads are about 8-15% more likely to answer the question about the feature correctly.The effect is increasing in intensity of exposure.
Stage I: Ad-Treatments FCONFIDENCE FCORRECTANSWER
(1) (2) (3) (4) (5) (6)
PIMPRESSIONS 1.108
(.698)
1.142
(1.133)
-.022#
(.012)
-.022
(.014)
FIMPRESSIONS 2.278
(1.525)
2.198*
(1.075)
2.182*
(1.075)
.121**
(.026)
.121**
(.026)
.120**
(.025)
PCRIMSONNUMADS -.520**
(.146)
-.496*
(.243)
- .008**
(.003)
- .008**
(.003)
FCRIMSONNUMADS 1.883**
(.264)
1.659**
(.187)
1.614**
(.187)
.052**
(.005)
.051**
(.004)
.048**
(.004)
Intercept 63.496**
(0.249)
63.509**
(0.439)
63.144**
(0.138)
.650**
(.004)
.650**
(.005)
.640**
(.003)
Fixed effects None RE FE None RE FE
N 22,959 22,959 22,959 22,995 22,995 22,995
R2 .003 .003 .004 .006 .007 .008
Significance Levels: #:10% *: 5% **: 1%
Stage I: Ad-Treatments FCONFIDENCE FCORRECTANSWER
(1) (2) (3) (4) (5) (6)
PIMPRESSIONS 1.108
(.698)
1.142
(1.133)
-.022#
(.012)
-.022
(.014)
FIMPRESSIONS 2.278
(1.525)
2.198*
(1.075)
2.182*
(1.075)
.121**
(.026)
.121**
(.026)
.120**
(.025)
PCRIMSONNUMADS -.520**
(.146)
-.496*
(.243)
- .008**
(.003)
- .008**
(.003)
FCRIMSONNUMADS 1.883**
(.264)
1.659**
(.187)
1.614**
(.187)
.052**
(.005)
.051**
(.004)
.048**
(.004)
Intercept 63.496**
(0.249)
63.509**
(0.439)
63.144**
(0.138)
.650**
(.004)
.650**
(.005)
.640**
(.003)
Fixed effects None RE FE None RE FE
N 22,959 22,959 22,959 22,995 22,995 22,995
R2 .003 .003 .004 .006 .007 .008
Significance Levels: #:10% *: 5% **: 1%
Both confidence and knowledge increases with ad treatments.
Stage I: Buzz-TreatmentsBID
All Products
Services Gadgets
BUZZ 8.504*
(4.206)
1.516
(1.561)
23.706*
(9.176)
NUMTREATED 3.780*
(1.886)
.822
(.669)
5.837*
(4.526)
N 373 227 146
R2 .019 .01 .048
Significance Levels: *: 5% **: 1%
Stage I: Buzz-TreatmentsBID
All Products
Services Gadgets
BUZZ 8.504*
(4.206)
1.516
(1.561)
23.706*
(9.176)
NUMTREATED 3.780*
(1.886)
.822
(.669)
5.837*
(4.526)
N 373 227 146
R2 .019 .01 .048
Significance Levels: *: 5% **: 1%
Buzz treatments raise valuations for gadgets.
Analysis: stage II
Use successful first stage as instruments for measuring the effects of social learning.
Regress confidence or correct answers of every NPG member on sum friends’ knowledge (PG) at various social distance using sum of info treatments as instruments.
Confidence FCONFIDENCE
(1) (2)
PGFCONFIDENCE_R .064*
(.029)
.057#
(.031)
PGFCONFIDENCE_NW1 .040**
(.013)
.034*
(.014)
PGFCONFIDENCE_NW2 .005
(.005)
.008#
(.005)
PGFCONFIDENCE_NW3 .003**
(.001)
.009**
(.001)
Control for # of Eligible NO YES
Intercept 59.628**
(.826)
67.870**
(1.197)
N 8,982 8,982
R2 0.018 0.045
Significance Levels: #:10% *: 5% **: 1%
FCONFIDENCE FCONFIDENCE
(1) (2)
PGFCONFIDENCE_R .064*
(.029)
.057#
(.031)
PGFCONFIDENCE_NW1 .040**
(.013)
.034*
(.014)
PGFCONFIDENCE_NW2 .005
(.005)
.008#
(.005)
PGFCONFIDENCE_NW3 .003**
(.001)
.009**
(.001)
Control for # of Eligible NO YES
Intercept 59.628**
(.826)
67.870**
(1.197)
N 8,982 8,982
R2 0.018 0.045
Significance Levels: #:10% *: 5% **: 1%
FCONFIDENCE FCONFIDENCE
(1) (2)
PGFCONFIDENCE_R .064*
(.029)
.057#
(.031)
PGFCONFIDENCE_NW1 .040**
(.013)
.034*
(.014)
PGFCONFIDENCE_NW2 .005
(.005)
.008#
(.005)
PGFCONFIDENCE_NW3 .003**
(.001)
.009**
(.001)
Control for # of Eligible NO YES
Intercept 59.628**
(.826)
67.870**
(1.197)
N 8,982 8,982
R2 0.018 0.045
Significance Levels: #:10% *: 5% **: 1%
Control for # of subjects who were eligible to receive products at distance R, NW1, NW2 and NW3.
FCORRECTANSWER FCORRECTANSWER
(1) (2)
PGFCORRECTANSWER_R .108**
(.026)
.070*
(.030)
PGFCORRECTANSWER_NW1 .041**
(.013)
.018
(.014)
PGFCORRECTANSWER_NW2 .019**
(.005)
.020**
(.005)
PGFCORRECTANSWER_NW3 .007**
(.001)
.018**
(.002)
Control for # of Eligible NO YES
Intercept .567**
(.010)
0.696**
(0.014)
N 9,006 9,006
R2 0.033 0.064
Significance Levels: #:10% *: 5% **: 1%
FCORRECTANSWER FCORRECTANSWER
(1) (2)
PGFCORRECTANSWER_R .108**
(.026)
.070*
(.030)
PGFCORRECTANSWER_NW1 .041**
(.013)
.018
(.014)
PGFCORRECTANSWER_NW2 .019**
(.005)
.020**
(.005)
PGFCORRECTANSWER_NW3 .007**
(.001)
.018**
(.002)
Control for # of Eligible NO YES
Intercept .567**
(.010)
0.696**
(0.014)
N 9,006 9,006
R2 0.033 0.064
Significance Levels: #:10% *: 5% **: 1%
FCORRECTANSWER FCORRECTANSWER
(1) (2)
PGFCORRECTANSWER_R .108**
(.026)
.070*
(.030)
PGFCORRECTANSWER_NW1 .041**
(.013)
.018
(.014)
PGFCORRECTANSWER_NW2 .019**
(.005)
.020**
(.005)
PGFCORRECTANSWER_NW3 .007**
(.001)
.018**
(.002)
Control for # of Eligible NO YES
Intercept .567**
(.010)
0.696**
(0.014)
N 9,006 9,006
R2 0.033 0.064
Significance Levels: #:10% *: 5% **: 1%
One standard deviation increase in each friend’s knowledge (about 30%)raises my knowledge by 1% to 2%. The total effect is about 9% because subjects are influenced by severaltreated subjects on average.
Alternative approach: Regressing knowledge on friends’ knowledge only measures average
amount of social learning.
We can instead measure social learning conditional on two subjects having reported to have talked to each other (collected during follow-up – 350 NPG subjects listed specific PG subjects whom they had talked to).
We exploit the fact that we both randomly distributed products and randomized information for each subject who received a product.
We assume that a NPG-subject’s pre-information is uncorrelated with the info treatment received by the PG-subject whom he or she talks to about the product.
This excludes the following situation: If I know that a Sidekick has AOL messenger I will specifically seek out subjects who received a product and whom we told about the AOL messenger capability of the Sidekick.
Effect of Info-Treated Friends on Knowledge (NPG)
Effect of PG-subject’s info-treatment on NPG-subject’s knowledge (only for subjects whoReported to have talked to specific PG subject)
Effect of Info-Treated Friends on Knowledge (NPG)
Effect of PG-subject’s info-treatment on NPG-subject’s knowledge (only for subjects whoReported to have talked to specific PG subject and seen PG subject with product)
68.474.3
Effect of Info-Treated Friends on Knowledge (NPG)
Effect of PG-subject’s info-treatment on NPG-subject’s knowledge (only for subjects whoReported to have talked to specific PG subject and seen PG subject with product)
68.474.3
Subjects who reported to have talked to a friend who had the productand whom they have seen use the product are 6% more likely to correctly answer a question about the feature if their friend had received an info treatment.
IV-Regression – confidence in answer FCONFIDENCE
Talked about OR
seen
(all)
Talked about OR seen
(services)
Talked about OR seen
(gadget)
Talked about AND
seen
FR_FCONFIDENCE .142**
(.054)
.124
(.100)
.151*
(.064)
.184*
(.074)
Intercept 61.617**
(5.626)
67.697**
(10.124)
59.495**
(6.795)
57.503**
(7.790)
N 1,912 400 1,511 1,207
Significance Levels: #:10% *: 5% **: 1%
IV-Regression – confidence in answer FCONFIDENCE
Talked about OR
seen
(all)
Talked about OR seen
(services)
Talked about OR seen
(gadget)
Talked about AND
seen
FR_FCONFIDENCE .142**
(.054)
.124
(.100)
.151*
(.064)
.184*
(.074)
Intercept 61.617**
(5.626)
67.697**
(10.124)
59.495**
(6.795)
57.503**
(7.790)
N 1,912 400 1,511 1,207
Significance Levels: #:10% *: 5% **: 1%
IV-Regression - knowledge FCORRECTANSWER
Talked about OR
seen
(all)
Talked about OR seen
(services)
Talked about OR seen
(gadget)
Talked about AND
seen
FR_FCORRECTANSWER .180**
(.067)
.011
(.106)
.246**
(.077)
.325**
(.112)
Intercept .567**
(.068)
.890**
(.107)
.461**
(.077)
.400**
(.109)
N 1,919 400 1,519 1,209
Significance Levels: #:10% *: 5% **: 1%
IV-Regression - knowledge FCORRECTANSWER
Talked about OR
seen
(all)
Talked about OR seen
(services)
Talked about OR seen
(gadget)
Talked about AND
seen
FR_FCORRECTANSWER .180**
(.067)
.011
(.106)
.246**
(.077)
.325**
(.112)
Intercept .567**
(.068)
.890**
(.107)
.461**
(.077)
.400**
(.109)
N 1,919 400 1,519 1,209
Significance Levels: #:10% *: 5% **: 1%
Info-treatment of friend is used as instrument. Estimated social-learning effects are about 3-15 times greater than the average effects estimated across all subjects.
Observations
Conditional on having communicated about the product social learning seems strongest for gadgets rather than services.
This might indicate that visual observation is important for social learning.
It is also possible that our feature set for gadgets provides a more natural decomposition of real-world communication than our feature set for services.
Analysis
Alternative Model
Model
An untreated (uninformed) subject has a probability p of interacting with some treated (informed) subject.
The interaction probability p depends on the social distance between uninformed and informed subject.
We distinguish three types of social distances: room mates (M), direct friends (NW1) and indirect friends (NW2).
Model
We define knowledge as the subjective or objective probability of answering a question about the product correctly.
If an informed and uninformed subject interact the knowledge of the informed subject is transferred to the uninformed subject (informed = treated with a product).
Model
We define knowledge as the subjective or objective probability of answering a question about the product correctly.
If an informed and uninformed subject interact the knowledge of the informed subject is transferred to the uninformed subject (informed = treated with a product).
After interacting the uninformed subject has the same probability of answering a question correctly as the informed subject.
Model Assume that the knowledge of an informed subject is and the
knowledge of an uninformed subject is .
Assume that the uninformed’s probability of interacting with some informed subject is X. Then we can express the final expected knowledge of the uninformed agent as:
UniformedInformedFinal FXFXF )1(
InfFUnifF
What is X?Assume that the uninformed agent has room mates who were
offered a product, direct friends and indirect friends. Then we can express X as:
21 )1()1()1(1 21NWNWR n
NWn
NWn
R pppX
Rn1NWn 2NWn
What is X?Assume that the uninformed agent has room mates who were
offered a product, direct friends and indirect friends. Then we can express X as:
21 )1()1()1(1 21NWNWR n
NWn
NWn
R pppX
Rn1NWn 2NWn
The probability of interacting with some informed subject is 1 minus theprobability of interacting with none of them.
Model We obtain:
21 )1()1()1)(( 21NWNWR n
NWn
NWn
RUniformedInformedFinalInformed pppFFFF
• We observe and in the followup survey.InfF FinalF
Model We obtain:
21 )1()1()1)(( 21NWNWR n
NWn
NWn
RUniformedInformedFinalInformed pppFFFF
• We observe and in the followup survey.
• We do not observe because we cannot do a baseline quiz without revealing the product.
InfF FinalF
UniformedF
Model We obtain expression (*):
21 )1()1()1)(( 21NWNWR n
NWn
NWn
RUniformedInformedFinalInformed pppFFFF
• We observe and in the followup survey.
• However, we do not observe because we cannot do a baseline quiz without revealing the product.
• Moreover, we expect the information of uninformed agents to vary with the number of eligible neighbors (and hence the number of neighbors who were offered a treatment) due to selection.
InfF FinalF
UniformedF
We instead compare agents in similar “cells”:
NW2 friends: Eligible, Treated
NW1 friends: Eligible, Treated
Roommate (M) friends: Eligible , Treated
Subject without product
We instead compare untreated agents in similar “cells”:
NW2 friends: Eligible, Treated
NW1 friends: Eligible, Treated
Roommate (M) friends: Eligible , Treated
Subject without product
We say the green subject lives in a (1,4+,4) cell to indicate that she has onetreated room-mate, and four treated NW1 and NW2 friends AND she has at least one more eligible (but non-treated) NW1 friend (indicated by plus sign).
For example, compare a (1,4+,4) cell with a (1,5,4) cell:
NW2 friends: Eligible, Treated
NW1 friends: Eligible, Treated
Roommate (M) friends: Eligible , Treated
Subject without product
NW2 friends: Eligible, Treated
NW1 friends: Eligible, Treated
Roommate (M) friends: Eligible , Treated
Subject without product
For example, compare a (1,4+,4) cell with a (1,5,4) cell:
NW2 friends: Eligible, Treated
NW1 friends: Eligible, Treated
Roommate (M) friends: Eligible , Treated
Subject without product
NW2 friends: Eligible, Treated
NW1 friends: Eligible, Treated
Roommate (M) friends: Eligible , Treated
Subject without product
The green agent on the right faces the same neighborhood as the agent on the leftbut the randomization turned one eligible, untreated agent into a treated agent.
Model
By dividing expression (*) for all agents in cell (1,5,4) by expression (*) for all agents in cell (1,4+,4) we obtain the marginal impact of treating one more NW1 neighbor:
1)4,4,1()4,4,1(
)4,5,1()4,5,1(
1 NWFinalInformed
FinalInformed pFF
FF
Model
By dividing expression (*) for all agents in cell (1,5,4) by expression (*) for all agents in cell (1,4+,4) we obtain the marginal impact of treating one more NW1 neighbor:
1)4,4,1()4,4,1(
)4,5,1()4,5,1(
1 NWFinalInformed
FinalInformed pFF
FF
Since we only have finitely many observations per cell we get an estimate forp. For each marginal comparison between two neighboring cells we get a newestimate. From this we can construct an estimate for p and a confidence interval.
Model
By dividing expression (*) for all agents in cell (1,5,4) by expression (*) for all agents in cell (1,4+,4) we obtain the marginal impact of treating one more NW1 neighbor:
1)4,4,1()4,4,1(
)4,5,1()4,5,1(
1 NWFinalInformed
FinalInformed pFF
FF
By comparing neighboring cells we are essentially differing out the unobserved knowledge of the uninformed agent.
Analysis
Results
Results
We are estimating the interaction probabilities separately for each product.
We use both subjective knowledge (“What is the probability that you can answer a Yes/No question correctly?”) and objective knowledge (“Actual share of correctly answered questions in the quiz”).
Results - Card
0.45
0.51
0.09
0.14 0.13
0.01
0.1
.2.3
.4.5
Inte
raction P
robability
M NW1 NW2
card
Objective Knowledge Subjective Knowledge
Results - Card
0.45
0.51
0.09
0.14 0.13
0.01
0.1
.2.3
.4.5
Inte
raction P
robability
M NW1 NW2
card
Objective Knowledge Subjective Knowledge
SE (0.16)* (0.21)* (0.02)* (0.04)* (0.09) (0.03)
Results - Yoga
0.49
0.61
0.11
0.20
0.01
0.12
0.2
.4.6
Inte
raction P
robabili
ty
M NW1 NW2
yoga
Objective Knowledge Subjective Knowledge
SE (0.19)* (0.23)* (0.04)* (0.03)* (0.03) (0.05)*
Results – Restaurant
0.30
0.24
0.120.10
0.01 0.01
0.1
.2.3
Inte
raction P
robability
M NW1 NW2
food
Objective Knowledge Subjective Knowledge
SE (0.03)* (0.08)* (0.03)* (0.04)* (0.02) (0.01)
Results – Camcorder
0.62
0.67
0.12 0.13
0.04 0.05
0.2
.4.6
.8In
tera
ction P
robability
M NW1 NW2
camcorder
Objective Knowledge Subjective Knowledge
SE (0.02)* (0.02)* (0.02)* (0.03)* (0.02)* (0.02)*
Results – MP3
0.58
0.52
0.120.08
0.04 0.04
0.2
.4.6
Inte
raction P
robability
M NW1 NW2
sound
Objective Knowledge Subjective Knowledge
SE (0.06)* (0.07)* (0.03)* (0.04)* (0.02)* (0.01)*
Results – PDA
0.36
0.45
0.120.16
0.06 0.05
0.1
.2.3
.4.5
Inte
raction P
robability
M NW1 NW2
pda
Objective Knowledge Subjective Knowledge
SE (0.04)* (0.07)* (0.03)* (0.04)* (0.02) (0.02)
Results
For “private products” the interaction probability for NW2 neighbors is usually insignificant.
For “public products” the NW2 effect is small but significant.
NW2 neighborhoods are also 7-times as large as NW1 neighborhoods! Therefore, the expected number of influenced NW2 agents can be large.
Who is influenced the most by social learning (close or distant neighbors)?(expected number of interactions taking Nhood size into account; subjective knowledge and significant probabilities only)
M NW1 NW2 TOTAL
CARD 0.50 1.12 1.62
YOGA 0.60 1.60 1.20
FOOD 0.24 0.80 1.04
CAM. 0.65 1.12 2.85 4.62
SOUND 0.50 0.64 2.28 3.42
PDA 0.45 1.44 2.85 4.74
Who is influenced the most by social learning (close or distant neighbors)?(expected number of interactions taking Nhood size into account; subjective knowledge and significant probabilities only)
M NW1 NW2 TOTAL
CARD 0.50 1.12 1.62
YOGA 0.60 1.60 1.20
FOOD 0.24 0.80 1.04
CAM. 0.65 1.12 2.85 4.62
SOUND 0.50 0.64 2.28 3.42
PDA 0.45 1.44 2.85 4.74
Although there is a greater probability to interact with close agents the expected number of interactions increases with distance.
Summary Three methodological contributions
Application–specific measure of social connectedness Hedonic analysis using configurators Measure of confidence using the Bobs
Advertising increases information. Social learning is as important as effects of
advertising. Future work:
Disentangle weak and strong social learning channels Measure social influence.