ANNUAL LECTURE Heterogeneous Agents, Social Interactions and Causal Inference This talk is based on “Heterogeneous Agents, Social Interactions, and Causal.

Slides:



Advertisements
Similar presentations
1 Chapter 4 The Designing Research Consumer. 2 High Quality Research: Evaluating Research Design High quality evaluation research uses the scientific.
Advertisements

Choosing the level of randomization
A Guide to Education Research in the Era of NCLB Brian Jacob University of Michigan December 5, 2007.
Holland on Rubin’s Model Part II. Formalizing These Intuitions. In the 1920 ’ s and 30 ’ s Jerzy Neyman, a Polish statistician, developed a mathematical.
The World Bank Human Development Network Spanish Impact Evaluation Fund.
GROUP-LEVEL DESIGNS Chapter 9.
The counterfactual logic for public policy evaluation Alberto Martini hard at first, natural later 1.
Using a Regression Discontinuity Design to Estimate the Impact of Placement Decisions in the Los Angeles Community College District Tatiana Melguizo &
PHSSR IG CyberSeminar Introductory Remarks Bryan Dowd Division of Health Policy and Management School of Public Health University of Minnesota.
Missing Data Issues in RCTs: What to Do When Data Are Missing? Analytic and Technical Support for Advancing Education Evaluations REL Directors Meeting.
Who are the participants? Creating a Quality Sample 47:269: Research Methods I Dr. Leonard March 22, 2010.
Pooled Cross Sections and Panel Data II
Robert L. Linn CRESST, University of Colorado at Boulder Paper presented at a symposium sponsored entitled “Accountability: Measurement and Value-Added.
Impact Evaluation: The case of Bogotá’s concession schools Felipe Barrera-Osorio World Bank 1 October 2010.
Analysis of Clustered and Longitudinal Data
The World Bank Human Development Network Spanish Impact Evaluation Fund.
Making all research results publically available: the cry of systematic reviewers.
Opportunities and Challenges in a Multi-Site Regression Discontinuity Design Stephen W. Raudenbush University of Chicago Presentation at the MultiLevel.
Effects of Low Doses Probability of Causation and implications for Public Policy Lecture at UC Berkeley March 2nd 2001 by Richard Wilson Mallinckrodt Research.
Causal Inference for Time-varying Instructional Treatments Stephen W. Raudenbush University of Chicago Joint Work with Guanglei Hong The research reported.
College Preparatory Curriculum for All Lessons Learned from Chicago Elaine Allensworth with TakakoNomi and Nicholas Montgomery at the Consortium on Chicago.
T tests comparing two means t tests comparing two means.
Quasi Experimental Methods I Nethra Palaniswamy Development Strategy and Governance International Food Policy Research Institute.
Designing a Random Assignment Social Experiment In the U.K.; The Employment Retention and Advancement Demonstration (ERA)
Causal Inference and Adequate Yearly Progress Derek Briggs University of Colorado at Boulder National Center for Research on Evaluation, Standards, and.
DISTRICT MANAGEMENT COUNCIL ACADEMIC RETURN ON INVESTMENT (A-ROI)
Implementation and process evaluation: developing our approach Ann Lendrum University of Manchester Neil Humphrey University of Manchester Gemma Moss Institute.
Consumer behavior studies1 CONSUMER BEHAVIOR STUDIES STATISTICAL ISSUES Ralph B. D’Agostino, Sr. Boston University Harvard Clinical Research Institute.
© Michael Lechner, 2006, p. 1 (Non-bayesian) Discussion (translation) of Principal Stratification for Causal Inference with Extended Partial Complience.
CAUSAL INFERENCE Shwetlena Sabarwal Africa Program for Education Impact Evaluation Accra, Ghana, May 2010.
Replacement Cases Framework overview Thresholds for inference and % bias to invalidate The counterfactual paradigm Internal validity example: kindergarten.
SUTVA, Assignment Mechanism STA 320 Design and Analysis of Causal Studies Dr. Kari Lock Morgan and Dr. Fan Li Department of Statistical Science Duke University.
Assumptions of value-added models for estimating school effects sean f reardon stephen w raudenbush april, 2008.
Laying the Foundation for Scaling Up During Development.
Beyond surveys: the research frontier moves to the use of administrative data to evaluate R&D grants Oliver Herrmann Ministry of Business, Innovation.
Optimal Design for Longitudinal and Multilevel Research Jessaca Spybrook July 10, 2008 *Joint work with Steve Raudenbush and Andres Martinez.
Application 2: Minnesota Domestic Violence Experiment Methods of Economic Investigation Lecture 6.
EXPERIMENTAL EPIDEMIOLOGY
Treatment Heterogeneity Cheryl Rossi VP BioRxConsult, Inc.
Generalizing Observational Study Results Applying Propensity Score Methods to Complex Surveys Megan Schuler Eva DuGoff Elizabeth Stuart National Conference.
Applying impact evaluation tools A hypothetical fertilizer project.
What is randomization and how does it solve the causality problem? 2.3.
Visualize Math: Success for Every Student From Techno COOL to Techno Tool: From Techno COOL to Techno TOOL –
Randomized controlled trials and the evaluation of development programs Chris Elbers VU University and AIID 11 November 2015.
Studying the Mean and Variation in the Effect of Program Participation in Multi-site Trials The research reported here was supported by a grant from the.
The Power of Comparison in Learning & Instruction Learning Outcomes Supported by Different Types of Comparisons Dr. Jon R. Star, Harvard University Dr.
ECON 3039 Labor Economics By Elliott Fan Economics, NTU Elliott Fan: Labor 2015 Fall Lecture 91.
Bilal Siddiqi Istanbul, May 12, 2015 Measuring Impact: Non-Experimental Methods.
Three ‘R’s for Evaluating the Memphis Striving Readers Project: Relationships, Real-World Challenges, and RCT Design Jill Feldman, RBS Director of Evaluation.
Impact Evaluation for Evidence-Based Policy Making Arianna Legovini Lead Specialist Africa Impact Evaluation Initiative.
1 The Training Benefits Program – A Methodological Exposition To: The Research Coordination Committee By: Jonathan Adam Lind Date: 04/01/16.
Copyright © 2015 Inter-American Development Bank. This work is licensed under a Creative Commons IGO 3.0 Attribution-Non Commercial-No Derivatives (CC-IGO.
Do European Social Fund labour market interventions work? Counterfactual evidence from the Czech Republic. Vladimir Kváča, Czech Ministry of Labour and.
Randomized Control Trials: What, Why, How, When, and Where Mark L. Davison MESI Conference March 11, 2016 Department of Educational Psychology.
David M. Murray, Ph.D. Associate Director for Prevention Director, Office of Disease Prevention Multilevel Intervention Research Methodology September.
Measuring College Value-Added: A Delicate Instrument
Measuring Results and Impact Evaluation: From Promises into Evidence
Threats and Analysis.
Stephen W. Raudenbush University of Chicago December 11, 2006
Quasi Experimental Methods I
An introduction to Impact Evaluation
AERA workshop April 4, 2014 (AERA on-line video – cost is $95)
Explanation of slide: Logos, to show while the audience arrive.
Impact evaluation: The quantitative methods with applications
RESEARCH METHODS Lecture 33
Explanation of slide: Logos, to show while the audience arrive.
Rerandomization to Improve Baseline Balance in Educational Experiments
Counterfactual models Time dependent confounding
RESEARCH METHODS Lecture 33
Chapter 3 Hernán & Robins Observational Studies
Presentation transcript:

ANNUAL LECTURE Heterogeneous Agents, Social Interactions and Causal Inference This talk is based on “Heterogeneous Agents, Social Interactions, and Causal Inference” by Guanglei Hong and Stephen W. Raudenbush, to appear in Morgan, S. (Ed.) Handbook of Causal Analysis for Social Research (Springer 2012) and draws on two examples originally reported in: Savitz-Verbitsky, N. and Raudenbush, S.W. (in press). Evaluating community policing program in Chicago: A case study of causal inference in spatial settings. To appear in Epidemiologic Methods; and Raudenbush, S.W., Reardon, S. and Nomi, T. (in press). Statistical analysis for multi-site trials using instrumental variables. To appear in Journal of Research and Educational Effectiveness. The research reported here was supported by a grant from the Spencer Foundation entitled “Improving Research on Instruction: Models Designs, and Analytic Methods;” and a grant from the W.T. Grant Foundation entitled “Building Capacity for Evaluating Group-Level Interventions.” 21 ST MARCH 2012 Stephen W. Raudenbush Lewis-Sebring Distinguished Service Professor in the Department of Sociology at the University of Chicago and Chairman of the Committee on Education

Abstract This talk will focus on two pervasive features of social interventions designed to increase human health, skills, or productivity. First, the interventions are usually delivered by human agents – physicians, teachers, case workers, therapists, police officers, or workplace managers - who tend to be ‘heterogeneous’ in the sense that they differ in their beliefs, training, and experience. These agents enact the intervention and shape its effects. Second, the participants in these interventions – patients, pupils, employees or offenders - are typically clustered in organizational settings, and social interactions among these participants influence the success of the intervention. In this presentation, Stephen will argue that causal models conventionally used in medical research are not well suited to study these interventions. Instead, he proposes a model in which the heterogeneous agents and social interactions among participants shape participants’ response to an intervention. Stephen will illustrate this model with studies of community policing and high-school curricular reform.

Outline Counter-Factual Account of Causation The “drug-trial paradigm” for causal inference An alternative paradigm for social interventions Heterogeneous agents Social interactions among participants Examples Community policing High School Curricular Reform Conclusions

Counter-factual Account of Causality In statistics (Neyman, Rubin, Rosenbaum) In economics (Haavelmo, Roy, Heckman)

Potential Outcomes in a Drug Trial Y (1) : Outcome if the patient receives Z = 1 (the “new drug”) Y (0) : Outcome if the patient receives Z = 0 (the “standard treatment”) Y (1) – Y (0) : Patient-specific causal effect E (Y (1) – Y (0) ) =  : Average causal effect

Stable Unit Treatment Value Assumption (Rubin, 1986) Each patient has two potential outcomes Implies – Only one “version” of each treatment – No “interference between units” Implies the doctor and the other patients have no effect on the potential outcomes

Formally…

Failure of SUTVA in Education Teachers enact instruction in classrooms –Multiple “versions of the treatment” Treatment assignment of one’s peers affects one’s own potential outcomes –EG Grade Retention –Hong and Raudenbush, Educational Evaluation and Policy Analysis, 2005 –Hong and Raudenbush, Journal of the American Statistical Association, 2006

Group-Randomized Trials Potential outcome Thus, each child has only two potential outcomes – if we have “intact classrooms” – if we have “no interference between classrooms”

Limitations of cluster randomized trial Mechanisms operate within clusters * Example: 4Rs teachers vary in response classroom interactions spill over We may have interference between clusters * Example: community policing

Alternative Paradigm Treatment setting (Hong, 2004): A unique local environment for each treatment composed of * a set of agents who may implement an intervention and * a set of participants who may receive it Each participant possesses a single potential outcome within each possible treatment setting Causal effects are comparisons between these potential outcomes

Example 1: Community Policing (joint work with Natalya Verbitsky-Shavitz) Let Z j =1 if Neighborhood j gets community policing Let Z j =0 if not Under SUTVA

Relaxing SUTVA Potential outcome for any unit depends on the treatment assignment of ALL units in the population, Individual Causal Effect: Population Average Causal Effect:

“All or none”

“Shall we do it in my neighborhood?”

Do it only in high-crime areas: effect on those areas 1, HC 0, LC 1, HC 0, HC 0, LC 0, HC

Do it only in high-crime areas: effect on low-crime areas 1, HC 0, LC 1, HC 0, HC 0, LC 0, HC 0, LC

Spatial Causal Assumptions (1) Functional Form: 1, #3 1, #1 0, #4 0, #5 1, #2

Longitudinal Design: 25 districts, 279 “beats” No community policing Community policing

Results Having community policing was especially good if your surrounding neighbors had it Not having community policing was especially bad if your neighbors had it *** So targetting only high crime areas may fail***

Example 2: Double-dose Algebra Requires 9 th -graders to take Double-dose Algebra if they scored below 50 percentile on 8 th -grade math test 1200 students in 60 Chicago high schools

Double-dose Algebra enrollment rate by math percentile scores (city wide) Enrollment Rates ITBS percentile scores

Conventional Mediation Model (T, M,Y model) Cut off (T) Double-Dose Algebra (M) Algebra Learning (Y) Assume no direct effect of T on Y (exclusion restriction) Δ= Effect of double dose on the “compliers” Δ Γ= Effect of assignment to double dose (“ITT” effect) Nomi, T., & Allensworth, E. (2009) Γ Δ

Effects of Double-dose Algebra: District-wide average Effect of cutoff on taking DD (average compliance rate): Increase prob by.72 District-wide average ITT effect on Y: Average effect≈0.15 District-wide average Complier-Average Treatment Effect Average ≈0.21 double-dose algebra effects varied across schools

But the policy changed classroom composition!!

Classroom average skill levels by math percentile scores Pre-policy ( and cohorts) Post-policy ( and cohorts)

Implementation varied across schools in--- Complying with the policy Inducing classroom segregation

Exclusion Restriction Revised T-M-C-Y model Cut off (T) Double-Dose Algebra (M) Algebra score (Y) Classroom Peer ability (C)

Research Questions 1)What is the average effect of assignment to DD? (“ITT effect”) 2)What is the average effect of taking double-dose algebra? (effect “on the compliers”). 3)How much do these effects vary across schools? 4)What is the effect of taking double-dose Algebra, holding constant classroom peer ability? 5) What is the effect of classroom peer ability, holding constant taking double-dose Algebra?

Results Degree of sorting LowAverageHigh ITT Complier effects School N19 22 The effect of double-dose algebra on algebra scores by the degree of sorting We now estimate the effect of taking DDA and classroom peer composition

Statistical Models Stage 1: the effect of Cut-off on Double Dose and Peer Ability Stage 2: the effect of M and C on Y 31

Stage 1 Results: the average effect T on M and C Double-dose algebra enrollmentPeer composition Coeff 0.72 *** *** SE 0.03 The effect of the cutoff score (T) on double-dose algebra enrollment (M) and peer composition (C) Note: *** p<.001, ** p<.01, * p<.05 32

Context specific effects: The effects of cutoff score on double-dose algebra enrollment and peer ability The effect of cut off score on peer ability The effect of cut off score on double-dose algebra enrollment

Stage 2 results: The effect of M and C on Y The average effect of taking double-dose algebra (M) and peer ability (C) on Algebra test scores Double-dose algebra enrollment Classroom Peer composition Coeff 0.30 *** 0.40 *** SE

5. Conclusions The reform enhanced math instruction for low- skill students, and that helped a lot The reform also intensified tracking and that hurt On balance the effect was positive, but much more so in schools that implemented double dose with minimal tracking

Final Thoughts Conventional causal paradigm: * a single potential outcome per participant under each treatment Alternative paradigm * a single potential outcome per participant in each treatment setting - aims to avoid bias -open up new questions Policy implications are potentially large