Supporting material for CIE

Slides:



Advertisements
Similar presentations
EN Regional Policy EUROPEAN COMMISSION Impact evaluation: some introductory words Daniel Mouqué Evaluation unit, DG REGIO Brussels, November 2008.
Advertisements

The World Bank Human Development Network Spanish Impact Evaluation Fund.
Challenges in evaluating social interventions: would an RCT design have been the answer to all our problems? Lyndal Bond, Kathryn Skivington, Gerry McCartney,
CHAPTER TWELVE ANALYSING DATA I: QUANTITATIVE DATA ANALYSIS.
Mywish K. Maredia Michigan State University
#ieGovern Impact Evaluation Workshop Istanbul, Turkey January 27-30, 2015 Measuring Impact 1 Non-experimental methods 2 Experiments Vincenzo Di Maro Development.
Estimating net impacts of the European Social Fund in England Paul Ainsworth Department for Work and Pensions July 2011
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.
The World Bank Human Development Network Spanish Impact Evaluation Fund.
Making Impact Evaluations Happen World Bank Operational Experience 6 th European Conference on Evaluation of Cohesion Policy 30 November 2009 Warsaw Joost.
How Policy Evaluations and Performance Management are Used Maureen Pirog Rudy Professor of Public and Environmental Affairs, Indiana University Affiliated.
Agenda: Block Watch: Random Assignment, Outcomes, and indicators Issues in Impact and Random Assignment: Youth Transition Demonstration –Who is randomized?
Chapter 9 Flashcards. measurement method that uses uniform procedures to collect, score, interpret, and report numerical results; usually has norms and.
PAI786: Urban Policy Class 2: Evaluating Social Programs.
Non Experimental Design in Education Ummul Ruthbah.
NIGERIA Impact of IT Training on Youth Employment Mr. Y.S. Labaran, Mr. F.O. Bajowa, Mr. Hamza Bello, Mrs. Yemisi Joel-Osebor.
1 The Need for Control: Learning what ESF achieves Robert Walker.
Evaluating Job Training Programs: What have we learned? Haeil Jung and Maureen Pirog School of Public and Environmental Affairs Indiana University Bloomington.
Evaluation of an ESF funded training program to firms: The Latvian case 1 Andrea Morescalchi Ministry of Finance, Riga (LV) March 2015 L. Elia, A.
Lecture PowerPoint Slides Basic Practice of Statistics 7 th Edition.
Public Policy Analysis ECON 3386 Anant Nyshadham.
AFRICA IMPACT EVALUATION INITIATIVE, AFTRL Africa Program for Education Impact Evaluation David Evans Impact Evaluation Cluster, AFTRL Slides by Paul J.
Applying impact evaluation tools A hypothetical fertilizer project.
Non-experimental methods Markus Goldstein The World Bank DECRG & AFTPM.
Randomized Assignment Difference-in-Differences
Bilal Siddiqi Istanbul, May 12, 2015 Measuring Impact: Non-Experimental Methods.
Prof. (FH) Dr. Alexandra Caspari Rigorous Impact Evaluation What It Is About and How It Can Be.
Do European Social Fund labour market interventions work? Counterfactual evidence from the Czech Republic. Vladimir Kváča, Czech Ministry of Labour and.
The Evaluation Problem Alexander Spermann, University of Freiburg 1 The Fundamental Evaluation Problem and its Solution SS 2009.
Patricia Gonzalez, OSEP June 14, The purpose of annual performance reporting is to demonstrate that IDEA funds are being used to improve or benefit.
Looking for statistical twins
Introduction to Labor Economics
JOB EVALUATION MAGNETIC CONTACTORS 1/26/2018.
Theme (i): New and emerging methods
Measuring Results and Impact Evaluation: From Promises into Evidence
Competence Centre on Microeconomic Evaluation (CC-ME)
L. Elia, A. Morescalchi, G. Santangelo
Unit Two Unemployment.
BUS 308 Competitive Success-- snaptutorial.com
BUS 308 Education for Service-- snaptutorial.com
BUS 308 Teaching Effectively-- snaptutorial.com
Experimental Design.
Evaluation Partnership Meeting March 2015
Explain the Impact of Poor Cost Information
Document E4/URBAN/2001/6_EN
Presentation at the African Economic Conference
Module 8 Statistical Reasoning in Everyday Life
Matching Methods & Propensity Scores
Centre for Research on Impact Evaluation
Matching Methods & Propensity Scores
CRIE activities in 2017 ESF Partnership Meeting 15 March 2017
Guide for Terms of Reference A checklist European Commission
Tabulations and Statistics
Gathering and Organizing Data
Impact Evaluation Methods
1 Causal Inference Counterfactuals False Counterfactuals
Matching Methods & Propensity Scores
Data collection, Data access and Data merging
Community of Practice on CIE
Learning Seminar - Targeting employment policies
III. Practical Considerations in preparing a CIE
Impact Evaluation Methods: Difference in difference & Matching
Guidance on Evaluation of Youth Employment Initiative
Lithuanian Experience of Counterfactual Impact Evaluation
Evaluating Impacts: An Overview of Quantitative Methods
Class 2: Evaluating Social Programs
Class 2: Evaluating Social Programs
Positive analysis in public finance
Feedback from Peer Review on 'Counterfactual Impact Evaluation'
Estimating net impacts of the European Social Fund in England
Presentation transcript:

Supporting material for CIE Video tutorials standard introductory e-learning tools to learn about CIE: outsourcing, application, use of results, choice of data to add subtitles to share them with the consultants

Tips and tricks on how to plan intervention design in view of future counterfactual impact evaluations European Commission Centre for Research on Impact Evaluation

A What’s inside the magic box? Tricks Some practical precautions that can make CIE easer to be carried on. Tips Hints on the most suitable CIE method to be applied in each situation. Typical situation For each trick, the situation in which it best applies.

Main message of the video B Always keep in mind the importance of creating a proper comparison group!

Example 1

Example 1 Launch of “National Supported Work Demonstration” by Manpower Demonstration Research Corporation. Aim of the intervention: to offer job training programs to particular worker categories in need. To participate, people need to apply.

Example 1 Treated group: random selection among applicants. Control group: not selected applicants. Intervention: 9 to 18 months of payed supported work. To disentangle the average treatment effect (ATE) : compare average employment rate of the treated and non-treated group 6 months after the end of the training.

Example 1 Trick Randomise treatment: randomisation creates two similar groups, except for one of them being treated. Tip The average treatment effect is given by the difference between the average employment rate of the treated and non treated. Typical situation In case you have more applications than available slots, allocate them trough randomisation.

Example 3

Example 3 As in example 1, the intervention consists in offering job training programs. Priority is given to poorer applicants. Participation (treatment) will be decided upon a threshold calculated with respect to a yearly income indicator.

Example 3 Treated group: all individuals with income indicator below the threshold level of 250 €. Control group: all individuals with income indicator above the threshold level (excluded from the intervention). The effect of the intervention will be computed by comparing the average employment rate of treated people just below the income threshold level with that of untreated workers just above the same threshold.

Example 3 Trick Allocate treatment on the basis of a score (forcing variable) with a threshold. The threshold must fall in the middle of the distribution of the forcing variable: ensure having both treated units and controls! Need to collect data both for the treated and the controls. The RDD works as a local randomisation around the threshold.

Example 3 Tips Regression Discontinuity Design (RDD). Typical situation Treatment assignment is a function of some variable (income, age).

Example 2 Example 4 Example 5 Example 6 Example 7

Summing-up Take-home message Use foreseen (and unforeseen) characteristics of the policy intervention to create an appropriate comparison group!

Summing-up Several typical situations imply specific class of CIE methods:

Summing-up Not all methods are applicable in a single intervention. Different methods estimate different kinds of impact: RDD estimates the impact for individuals at the threshold PSM estimates ATT DiD estimates ATE IV estimates LATE  Choose the most appropriate impact for your intervention

Summing-up When designing the intervention, pave the way for the application of 2 or 3 methods Save money Have reliable results CIE methods can be combined and refined to fit special situations and purposes  ask to a CIE expert (https:\\crie.jrc.ec.europa.eu/?q=content/cie- market)!

Stay in touch CRIE website: crie.jrc.ec.europa.eu email: jrc-crie@ec.europa.eu YouTube: JRC CRIE Yammer: https://www.yammer.com/counterfactualimpactevaluationnetworkcie-net/#/home