61
Research Using Behavioral Big Data A Tour and Why Mechanical Engineers Should Care Galit Shmueli

Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care

Embed Size (px)

Citation preview

Page 1: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care

Research Using Behavioral Big Data A Tour and Why Mechanical Engineers Should Care

Galit Shmueli

Page 2: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care

I’m not a mechanical engineerPhD in Statistics, Technion IE&MCMU Statistics DeptU of Maryland Business SchoolIndian School of BusinessNational Tsing (“Ching”) Hua U, Inst. Service Science

פרופ' מנחם שמואלי )ז"ל(

1935-1980הנדסת מכונות, טכניון

Page 3: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care

Research in Data Analytics‘Entrepreneurial’ statistical & data mining modeling (for today’s problems)

Interdisciplinary Research

Statistical StrategyTo Explain or To Predict?Information QualityData Mining and Causality

Page 4: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care

What is Behavioral Big Data (BBD)

Special type of Big DataBehavioral: people’s actions, interactions, self-reported opinions, thoughts, feelings

Human and social aspects: Intentions, deception, emotion, reciprocation, herding,…

When aware of data collection -> modify behavior (legal risks, embarrassment, unwanted solicitation)

Page 5: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care

BBD vs. Medical Big Data

• Physical measurements

• Data collection timing often set by medical system

• Clinical trials: awareness & vested interest

• People’s daily actions, interactions, self-reported feelings, opinions, thoughts

• Data generation timing often chosen by user

• Experiments: users often unaware; goal not always in user’s interest

Page 6: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care

BBD on Citizens and Customers – old storyGovernments law enforcement, security, traffic (cameras, sensors)

Financial Institutionsfraud, loans (IT systems, cameras)

Telecoms fraud, infrastructure, marketing (IT systems, mobile)

Retail Chainsmarketing, operations, merchandising (POS systems, video, social, mobile)

InsuranceUsage-based premiums (telematics)

“Old”:• Cameras• Sensors• IT systems

(POS, calls,…)New:• GPS• Internet• Mobile• Social• Things

Page 7: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care

BBD on Employees

Service Providersquality control, employee performance

Electronic Performance Monitoring (EPM) systems, web surfing, e-mails sent and received, telephone use, video, location (taxis)

Page 8: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care

BBD on Citizens, Customers, Employees: Internet!• BBD now also available to small companies & organizations• Online platforms have BBD (e-commerce, gaming, search,

social networks…)• Voluntarily entered by users (UGC): personal details, photos,

comments, messages, search terms, bids in auctions, likes, payment information, connections with “friends”

• Passive footprints: duration on the website, pages browsed, sequence, referring website, Internet browser, operating system, location, IP address.

• BBD now available to individuals: Quantified Self

Page 9: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care

1. Research Opportunity

2. Understand3. Collaborate

How does your ME work relate to BBD? To Data Analytics & Social Sci?

Engineering

Social Sciences

Data Analytics

Behavioral Big Data

Page 10: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care
Page 11: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care
Page 12: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care
Page 13: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care
Page 14: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care

From theory to practice

Page 15: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care

More and more human and social activities are moving online

Most companies that have BBD were not created for the purpose of generating BBD

Two important points

Page 16: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care

Why should mechanical engineers care about BBD?

Technology is advancing in two directions

Fully automated (algorithmic) solutions

Because you are (and should be) involved in designing both!

Micro-level recording of human and social behavior

Page 17: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care

1. Research Opportunity2. Understand3. Collaborate

How does your ME work relate to BBD? To Data Analytics & Social Sci?

Engineering

Social Sciences

Data Analytics

Behavioral Big Data

Page 18: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care
Page 19: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care

the most crucial choices about the future of ordinary voters and their children are probably made not by Brussels bureaucrats or Washington lobbyists but by engineers, entrepreneurs, and scientists who are hardly aware of the implications of their decisions, and who certainly don’t represent anyone.

Page 20: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care

Brief Tour of BBD Research in the Land of Social Science & Business

Page 21: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care

Research using BBD

Duncan Watts, Microsoft Research (NY):1. Social science problems are almost always more

difficult than they seem2. The data required to address many problems of

interest to social scientists remain difficult to assemble3. Thorough exploration of complex social problems

often requires the complementary application of multiple research traditions

Page 22: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care

Academic Research Qs using BBD

Causal questions about human and social behavior

examine new phenomena

re-examine old phenomena with better data

Page 23: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care

Research Methodologies Using BBD

Quasi experiments

Randomized experiments

Observational studies

Survey studies

Naturalexperiments

Page 24: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care

Research Communities

Researchers with social science + technical backgrounds

Information Systems

Marketing

Computational Social Science

Page 25: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care

7 Examples of BBD Studies in Top Journals

Page 26: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care

Emotional Contagion in Social Networks (Kramer et al. Proc of the National Academies of Sciences, 2014)

• Can emotional states be transferred to others via emotional contagion?

• Old question, new data• Large-scale experiment run by FB,

manipulating users’ exposure level to emotional expressions in their Facebook News Feed

Anonymous Browsing in Dating Websites (Bapna et al. Management Science, 2016)

• How does anonymous browsing affect outcomes on dating sites?• New questions about human behavior due to new technologies• Large-scale experiment on N American dating website

Identifying Influential and Susceptible Members of Social Networks (Aral and Walker, Science, 2012)

• How do individuals’ attributes modulate peer influence

• Old question in new context• Experiment on social news

aggregation website where users contribute news articles, discuss them, and rate comments

Page 27: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care

Consumption in Virtual Worlds (Hinz et al. Info Sys Research, 2015)• Does conspicuous consumption increase social status? • Age-old sociology question with new BBD data• Observational BBD from 2 virtual world websites (gaming with social network)

Impact of Online Intermediaries on HIV Transmission (Ghose & Chan MIS Quarterly, 2015)

• Does entry of major online personals ad website increase HIV prevalence?

• New context• Natural experiment on Craigslist

Impact of Info Hiding on Crowdfunding(Burtch et al. Management Science, 2016)

• Does peer influence drive information hiding in crowdfunding campaigns and effect on contributions

• New online social context• Observational BBD from large online

crowdfunding platform

Page 28: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care

Forecasting Elections with Non-Representative Polls(Wang et al. Intl. Journal on Forecasting, 2014)

• Can elections be forecast using a non-representative sample?

• Old question, new data• Survey BBD from Xbox with built-in daily poll

Page 29: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care

ONE WAY MIRRORS IN ONLINE DATINGA Randomized Field Experiment

Ravi Bapna, University of MinnesotaJui Ramaprasad, Mcgill UniversityGalit Shmueli, National Tsing Hua

UniversityAkhmed Umyarov, University of

Minnesota

Page 30: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care

Online Dating

46

of the single population in the US uses online dating to find a partner (Gelles 2011)

%

Page 31: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care

Online Dating Website

Page 32: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care

Non-anonymous Browsing (Default)

Profile Visit

Recent visitor:

Page 33: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care

Anonymous Browsing

Profile Visit

Recent visitor:

NONE

Page 34: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care

Research Question (in simple words)

How does anonymous browsing affect user behavior?

… and matching?

Page 35: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care

Formal Research Question

what is the relative causal effect of social inhibitions on search preferences vs. social inhibitions of contact initiation in dating markets?

given known gender asymmetries, how does this effect differ for men vs. women?

Page 36: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care

Randomized Field Experiment on Large Online Dating Website

50,000 users receive gift of anonymous browsing

Page 37: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care

Results

Users treated with anonymity

become disinhibited view more profiles, view more same-sex and interracial mates

get less matcheslose ability to leave a weak signal - especially harmful for women!

Page 38: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care

Role of anonymity and importance of WEAK SIGNAL

in online platforms

Page 39: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care

In Academia

Purpose: Scientific inquiry

Causal Qs are most popular• Determinants of social phenomena• Impact studies

Predictive Qs (quite rare)

In Industry

Purpose: evaluate or improve products, service, operations, etc.

Mostly predictive, but also causal• Netflix Prize: recommender system• Yahoo!, LinkedIn, FB: personalized news

content to increase user engagement/clicks

• Target: pregnancy prediction• Amazon: pricing, logistics,...• Government: campaign targeting

Study Types• Observational• Experiments• Surveys

BBD-based Research: Academia vs. Industry

Page 40: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care

Getting BBD for Research

1. Open Data, Publicly Available DataData.govTwitterKaggle (UCI MR)API and web scraping

2. Partnering with a Company• Both parties interested in research question• Data purchase• Personal connections• Partnership between school and organization

(CMU Living Analytics Research Lab)

Page 41: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care

3. CrowdsourcingAMT Replacing student subjects• Experiment subjects• Survey respondents• Cleaning and tagging data

“easy access to a large, stable, and diverse subject pool, the low cost of doing experiments, and faster iteration between developing theory and executing experiments” [Mason and Suri, 2012]

Page 42: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care

Using BBD for Research: Human Subjects

Institutional Review Board (IRB)“ethics committee”

University-level committee designated to approve, monitor, and review biomedical and behavioral research involving humans.• performs benefit-risk analysis for

proposed study• guidelines: Beneficence, Justice, and

Respect for persons

Page 43: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care

• HHS propose new IRB exemption criteria for publicly available data (or even buying it)• Council for Big Data, Ethics & Society’s letter: “these criteria for exclusion focus on the

status of the dataset… not the content of the dataset nor what will be done with the dataset, which are more accurate criteria for determining the risk profile of the proposed research

Ethics: Beyond IRBFacebook experiment [Kramer et al. 2014]: IRB Exemption

“[The work] was consistent with Facebook’s Data Use Policy, to which all users agree prior to creating an account on Facebook, constituting informed consent for this research.”

• Expression of Concern by PNAS editor• Varied response from public, academia,

press, ethicists, corporates [Adar 2015]

Page 44: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care

Big Behavioral Field Experiments: 5 Challenges

Page 45: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care

Big Behavioral Field Experiments: Challenges

1. Fast-Changing EnvironmentUsers keep evolvingTechnology changes fast (Netflix)Parallel experiments run every day (Amazon)

2. Multidimensional Behavior, Context, Objectives Comp. advertising & content recommendation: 3M’s [Agarwal & Chen 2016]• Multi-response (clicks, shares, likes,…)• Multi-context (mobile, email,...)• Multiple objectives (engagement, revenue,...)

Page 46: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care

4. Spillover EffectsTreatment can affect control group (social networks)How to randomly assign on a social network?Dependence among units (data analysis) [Fienberg, 2015]

3. Knowledge of Allocation; Gift Effect (≈ clinical trials) • Allocation knowledge can affect outcome• Blinding? placebo?• Online users discover their allocation via online forums• “Gift” or preferential treatment can affect outcome

BB Field Experiments: More Challenges

Page 47: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care

5. Ethical and Moral IssuesEase of running a large scale experiment quickly and at low cost -> danger of harming many people quickly

small scale pilot study?

Experiment platforms: Fair treatment & payment

BB Field Experiments: Even More Challenges

Page 48: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care
Page 49: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care

Big Behavioral Quasi-Experiments & Observational Studies: 5 Methodological Issues

Page 50: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care

Quasi-Experiments and Observational BBD: Methodological Challenges

1. Data Size & DimensionScaling of statistical inference: p-values, multiple testing“Too Big to Fail: Large Samples and the p-Value Problem” (Lin, Lucas & Shmueli ISR 2013)

Data DredgingCan detect lots of tiny & complex effectsRole of theory vs data discovery

Role of Prediction“Predictive Analytics in Information Systems Research” (Shmueli & Koppius MISQ 2011)

Page 51: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care

2. Self-Selection BiasUsers choose treatment/control groupScaling of stat/econ methods to big data

“A Tree-Based Approach for Addressing Self-selection in Impact Studies with Big Data” (Yahav, Shmueli & Mani, MIS Quarterly 2016)

Page 52: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care

More challenges (in search of causal explanations)

3. Simpson’s ParadoxCausal direction reverses when data are disaggregated

Big data: lots of possible breakdowns

“The Forest or the Trees? Tackling Simpson’s Paradox with Classification Trees” (Shmueli & Yahav, 2016)

Does a dataset display a paradox?

Page 53: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care

And finally…

5. Data Contaminated by Experiments+ some of the randomized experiments issues (fast-changing environment, etc.)

Page 54: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care

Using Observational Data: Ethical & Moral Issues

1. Web data collection by researchers

2. Data protection, data sharing, and reproducible research (Privacy - Netflix)

3. Data tagging by AMT – fair payment (+quality issues)

Page 55: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care

Large Scale Surveys

Data quality issues at large scale• duplicate responses• insincere responses

Online surveys: cheap, easy, fastLarge pool of available “workers”Supplement experimental/observational studies

The promise of para dataData on how the survey was accessed/answered (OECD Survey of Adult Skills)• time stamps of opening invitation email, survey access,…• duration for answering each question

Page 56: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care

The real gorilla in large scale surveys: Generalization

Sampling and non-sampling errors“The central issue is whether conditional effects in the sample… may be transported to desired target populations. Success depends on compatibility of causal structures in study and target populations, and will require subject matter considerations in each concrete case.” - Keiding and Louis, JRSS 2016

Statistical generalization & scientific generalization

Who do the Turkers represent?

Information Quality: The Potential of Data & Analytics to Generate Knowledge, Kenett & Shmueli, Wiley 2016“Clarifying the terminology that describes scientific reproducibility” (Kenett & Shmueli, Nature Methods 2015)

Page 57: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care

Summary

Technical ChallengesData accessAnalysis scalabilityQuick-changing environment

BBD = lots of behavioral dataWho has it?How is it analyzed?For what purpose?

Methodological ChallengesSelection biasGeneralizationData contaminated by other experimentsSpillover effectsLack of methodical lifecycle

Legal, Ethical, Moral ChallengesPrivacy violation (Netflix; networks)Risks to human subjectsCompany vs. Researcher ObjectivesGains of company at expense of individuals, communities, societies, & science

Page 58: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care

Why should mechanical engineers care about BBD?Technology is advancing in two directions

Fully automated (algorithmic) solutions

Micro-level recording of human and social behavior

Page 59: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care

Going Forward…

Convergence of Social Sciences & Engineering

Things now collect BBD (intentionally or not)

Page 60: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care

1. Research Opportunity2. Understand3. Collaborate

How does your ME work relate to BBD? To Data Analytics & Social Sci?

Engineering

Social Sciences

Data Analytics

Behavioral Big Data

Page 61: Research Using Behavioral Big Data: A Tour and Why Mechanical Engineers Should Care

Galit Shmueli 徐茉莉Institute of Service Science