Quasi Experimental Methods I Nethra Palaniswamy Development Strategy and Governance International Food Policy Research Institute.

Slides:



Advertisements
Similar presentations
N ON -E XPERIMENTAL M ETHODS Shwetlena Sabarwal (thanks to Markus Goldstein for the slides)
Advertisements

REGRESSION, IV, MATCHING Treatment effect Boualem RABTA Center for World Food Studies (SOW-VU) Vrije Universiteit - Amsterdam.
Advantages and limitations of non- and quasi-experimental methods Module 2.2.
Review of Identifying Causal Effects Methods of Economic Investigation Lecture 13.
#ieGovern Impact Evaluation Workshop Istanbul, Turkey January 27-30, 2015 Measuring Impact 1 Non-experimental methods 2 Experiments Vincenzo Di Maro Development.
Presented by Malte Lierl (Yale University).  How do we measure program impact when random assignment is not possible ?  e.g. universal take-up  non-excludable.
Differences-in-Differences
Assessing Program Impact Chapter 8. Impact assessments answer… Does a program really work? Does a program produce desired effects over and above what.
The World Bank Human Development Network Spanish Impact Evaluation Fund.
Who are the participants? Creating a Quality Sample 47:269: Research Methods I Dr. Leonard March 22, 2010.
Impact Evaluation: The case of Bogotá’s concession schools Felipe Barrera-Osorio World Bank 1 October 2010.
Summary of Propensity Score Matching in Education
TOOLS OF POSITIVE ANALYSIS
PAI786: Urban Policy Class 2: Evaluating Social Programs.
Experimental Design The Gold Standard?.
I want to test a wound treatment or educational program but I have no funding or resources, How do I do it? Implementing & evaluating wound research conducted.
Development Impact Evaluation Initiative innovations & solutions in infrastructure, agriculture & environment naivasha, april 23-27, 2011 in collaboration.
I want to test a wound treatment or educational program in my clinical setting with patient groups that are convenient or that already exist, How do I.
Matching Methods. Matching: Overview  The ideal comparison group is selected such that matches the treatment group using either a comprehensive baseline.
TRANSLATING RESEARCH INTO ACTION What is Randomized Evaluation? Why Randomize? J-PAL South Asia, April 29, 2011.
AADAPT Workshop Latin America Brasilia, November 16-20, 2009 Non-Experimental Methods Florence Kondylis.
Global Workshop on Development Impact Evaluation in Finance and Private Sector Rio de Janeiro, June 6-10, 2011 Mattea Stein Quasi Experimental Methods.
Designing a Random Assignment Social Experiment In the U.K.; The Employment Retention and Advancement Demonstration (ERA)
Case Studies Harry Anthony Patrinos World Bank November 2009.
Causal Inference & Quasi-experimental Methods Arndt Reichert 22 June 2015 ieConnect Impact Evaluation Workshop Rio de Janeiro, Brazil June 22-25, 2015.
The World Bank Human Development Network Spanish Impact Evaluation Fund.
Impact Evaluation in Education Introduction to Monitoring and Evaluation Andrew Jenkins 23/03/14.
Beyond surveys: the research frontier moves to the use of administrative data to evaluate R&D grants Oliver Herrmann Ministry of Business, Innovation.
Africa Impact Evaluation Program on AIDS (AIM-AIDS) Cape Town, South Africa March 8 – 13, Causal Inference Nandini Krishnan Africa Impact Evaluation.
1 Experimental Research Cause + Effect Manipulation Control.
CAUSAL INFERENCE Presented by: Dan Dowhower Alysia Cohen H 615 Friday, October 4, 2013.
Rigorous Quasi-Experimental Evaluations: Design Considerations Sung-Woo Cho, Ph.D. June 11, 2015 Success from the Start: Round 4 Convening US Department.
AP Review #4: Sampling & Experimental Design. Sampling Techniques Simple Random Sample – Each combination of individuals has an equal chance of being.
AFRICA IMPACT EVALUATION INITIATIVE, AFTRL Africa Program for Education Impact Evaluation David Evans Impact Evaluation Cluster, AFTRL Slides by Paul J.
Nigeria Impact Evaluation Community of Practice Abuja, Nigeria, April 2, 2014 Measuring Program Impacts Through Randomization David Evans (World Bank)
Applying impact evaluation tools A hypothetical fertilizer project.
Non-experimental methods Markus Goldstein The World Bank DECRG & AFTPM.
What is randomization and how does it solve the causality problem? 2.3.
1.) *Experiment* 2.) Quasi-Experiment 3.) Correlation 4.) Naturalistic Observation 5.) Case Study 6.) Survey Research.
1 Module 3 Designs. 2 Family Health Project: Exercise Review Discuss the Family Health Case and these questions. Consider how gender issues influence.
Implementing an impact evaluation under constraints Emanuela Galasso (DECRG) Prem Learning Week May 2 nd, 2006.
Randomized Assignment Difference-in-Differences
Bilal Siddiqi Istanbul, May 12, 2015 Measuring Impact: Non-Experimental Methods.
1 Joint meeting of ESF Evaluation Partnership and DG REGIO Evaluation Network in Gdańsk (Poland) on 8 July 2011 The Use of Counterfactual Impact Evaluation.
Outcomes Evaluation A good evaluation is …. –Useful to its audience –practical to implement –conducted ethically –technically accurate.
How Psychologists Do Research Chapter 2. How Psychologists Do Research What makes psychological research scientific? Research Methods Descriptive studies.
Measuring causal impact 2.1. What is impact? The impact of a program is the difference in outcomes caused by the program It is the difference between.
Copyright © 2015 Inter-American Development Bank. This work is licensed under a Creative Commons IGO 3.0 Attribution-Non Commercial-No Derivatives (CC-IGO.
Innovations in investment climate reforms: an impact evaluation workshop November 12-16, 2012, Paris Non Experimental Methods Florence Kondylis.
Cross-Country Workshop for Impact Evaluations in Agriculture and Community Driven Development Addis Ababa, April 13-16, Causal Inference Nandini.
1 An introduction to Impact Evaluation (IE) for HIV/AIDS Programs March 12, 2009 Cape Town Léandre Bassolé ACTafrica, The World Bank.
The Evaluation Problem Alexander Spermann, University of Freiburg, 2007/ The Fundamental Evaluation Problem and its Solution.
Quasi Experimental Methods I
Quasi Experimental Methods I
An introduction to Impact Evaluation
Quasi-Experimental Methods
Quasi-Experimental Methods
Chapter Eight: Quantitative Methods
Matching Methods & Propensity Scores
Matching Methods & Propensity Scores
Establishing the Direction of the Relationship
Impact Evaluation Methods
1 Causal Inference Counterfactuals False Counterfactuals
Matching Methods & Propensity Scores
Impact Evaluation Methods: Difference in difference & Matching
Evaluating Impacts: An Overview of Quantitative Methods
Class 2: Evaluating Social Programs
Class 2: Evaluating Social Programs
Applying Impact Evaluation Tools: Hypothetical Fertilizer Project
Presentation transcript:

Quasi Experimental Methods I Nethra Palaniswamy Development Strategy and Governance International Food Policy Research Institute

What we know so far Aim: We want to isolate the causal effect of our interventions on our outcomes of interest  Use rigorous evaluation methods to answer our operational questions  Randomizing the assignment to treatment is the “gold standard” methodology (simple, precise, cheap)  What if randomization is not feasible? >> Where it makes sense, resort to non-experimental methods

When does it make sense?  Can we find a plausible counterfactual?  Every non-experimental method is associated with a set of assumptions  Assumptions about plausible counterfactual  The stronger the assumptions, the more doubtful our measure of the causal effect  Question assumptions ▪ Are these assumptions valid?

Example: Funds for community infrastructure  Principal Objective ▪ Improving community infrastructure- primary schools Intervention ▪ Community grants ▪ Non-random assignment  Target group ▪ Communities with poor education infrastructure ▪ Communities with high poverty rates  Main result indicator ▪ Primary school enrolment

5 (+) Impact of the program (+) Impact of external factors Illustration: Funds for Community Infrastructure(1)

6 (+) BIASED Measure of the program impact Before-After comparisons

7 « After » Difference between participants and non-participants Before-After comparisons for participating and non-participating communities « Before» Difference between participants and non-participants >> What’s the impact of our intervention?

Difference-in-Differences Identification Strategy (1) Counterfactual: 2 Formulations that say the same thing 1. Non-participants’ enrolments after the intervention, accounting for the “before” difference between participants/nonparticipants (the initial gap between groups) 2. Participants’ enrolments before the intervention, accounting for the “before/after” difference for nonparticipants (the influence of external factors)  1 and 2 are equivalent

Difference-in-Differences Identification Strategy (2) Underlying assumption: Without the intervention, enrolments for participants and non participants’ would have followed the same trend >> Participating communities and non- partipating communities would have behaved in the same way on average, in the absence of the intervention

Data -- Example 1

NP NP 2007 =10.8 Impact = (P P 2007 ) -(NP NP 2007 ) = 10.6 – 10.8 = -0.2 Impact = (P P 2007 ) -(NP NP 2007 ) = 10.6 – 10.8 = -0.2 P P 2007 =10.6

P-NP 2008 =0.5 Impact = (P-NP) (P-NP) 2007 = = -0.2 Impact = (P-NP) (P-NP) 2007 = = -0.2 P-NP 2007 =0.7

Summary  Negative Impact:  Very counter-intuitive: Funding for building primary schools should not decrease enrolment rates once external factors are accounted for!  Assumption of same trend very strong  2 sets of communities groups had, in 2007, different pre- existing characteristics and different paths  Non-participating communities would have had slower increases in enrolment in the absence of funds for building primary schools ➤ Question the underlying assumption of same trend! ➤ When possible, test assumption of same trend with data from previous years

Questioning the Assumption of same trend: Use pre-pr0gram data >> Reject counterfactual assumption of same trends !

Data – Example 2

NP 08 -NP 07 =0.2 Impact = (P P 2007 ) -(NP NP 2007 ) = 0.6 – 0.2 = Impact = (P P 2007 ) -(NP NP 2007 ) = 0.6 – 0.2 = + 0.4

Impact = (P P 2007 ) -(NP NP 2007 ) = 0.6 – 0.2 = Impact = (P P 2007 ) -(NP NP 2007 ) = 0.6 – 0.2 = + 0.4

Conclusion  Positive Impact:  More intuitive  Is the assumption of same trend reasonable? ➤ Still need to question the counterfactual assumption of same trends ! ➤ Use data from previous years

Questioning the Assumption of same trend: Use pre-pr0gram data >>Seems reasonable to accept counterfactual assumption of same trend ?!

Caveats (1)  Assuming same trend is often problematic  No data to test the assumption  Even if trends are similar the previous year… ▪ Where they always similar (or are we lucky)? ▪ More importantly, will they always be similar? ▪ Example: Other project intervenes in our nonparticipating communities…

Caveats (2)  What to do? >> Check similarity in observable characteristics ▪ If not similar along observables, chances are trends will differ in unpredictable ways >> Still, we cannot check what we cannot see… And unobservable characteristics might matter more than observable (social cohesion, community participation)

Matching Method + Difference-in- Differences (1) Match participants with non-participants on the basis of observable characteristics Counterfactual:  Matched comparison group  Each program participant is paired with one or more similar non-participant(s) based on observable characteristics >> On average, participants and nonparticipants share the same observable characteristics (by construction)  Estimate the effect of our intervention by using difference-in-differences

Matching Method (2) Underlying counterfactual assumptions  After matching, there are no differences between participants and nonparticipants in terms of unobservable characteristics AND/OR  Unobservable characteristics do not affect the assignment to the treatment, nor the outcomes of interest

How do we do it?  Design a control group by establishing close matches in terms of observable characteristics  Carefully select variables along which to match participants to their control group  So that we only retain ▪ Treatment Group: Participants that could find a match ▪ Comparison Group: Non-participants similar enough to the participants >> We trim out a portion of our treatment group!

Implications  In most cases, we cannot match everyone  Need to understand who is left out  Example Score Nonparticipants Participants Matched Individuals Average incomes Portion of treatment group trimmed out

Conclusion (1)  Advantage of the matching method  Does not require randomization

Conclusion (2)  Disadvantages:  Underlying counterfactual assumption is not plausible in all contexts, hard to test ▪ Use common sense, be descriptive  Requires very high quality data: ▪ Need to control for all factors that influence program placement/outcome of choice  Requires significantly large sample size to generate comparison group  Cannot always match everyone…

Summary  Randomized-Controlled-Trials require minimal assumptions and procure intuitive estimates (sample means!)  Non-experimental methods require assumptions that must be carefully tested  More data-intensive  Not always testable  Get creative:  Mix-and-match types of methods!  Address relevant questions with relevant techniques