Statistical methods for policy evaluation


Statistical methods for policy evaluation

Course Leader


Target Group

Official statisticians (including managers and junior statisticians), needing basic knowledge on how data can inform policy design, analysis and evaluation

Entry Qualifications

  • Sound command of English. Participants should be able to make short interventions and to actively participate in discussions
  • Recommended: statistical programming language such as R-CRAN, Python, STATA
  • Laptops with the statistical programs installed, internet connection to complete quizzes


This training introduces key analytical tools for the appraisal of projects, policies, programmes and regulations. Attention will be paid to the conceptual framework for appraisal with particular reference to methods used to value non-market impacts. Given that these tools are increasingly being used in actual policy formulation by, for example, national governments and international development agencies, a range of applications will be drawn upon.


  • Introduction to appraisal and evaluations: evidence based policy making, cost-benefit analysis, evaluation processes and approaches, ethical considerations and complementary approaches, sensitivity analysis
  • Causal inference and counterfactuals: Confounding and Causal Diagrams, Back doors and front doors, mediation, colliders and Simpson’s Paradox, Selection bias
  • Randomized Control Experiments: From Clinical Trials to Field Experiments, Outcome and performance indicators,  Clinical Significance and Statistical Power, Survival Analysis with Censoring, Tests for conditional independence
  • Quasi experiments: Difference in difference, Regression discontinuity, Interrupted time series, Instrumental variables, Matching and propensity scores, Inverse probability of treatment weighting,

Expected Outcome

By the end of the course learners will:

  • Understand the basic principles of policy evaluation and program appraisal approaches;
  • Be able to analyse a range of policy evaluation approaches of non-market impacts;
  • Interpret results of these approaches and identify their limitations

Training Methods

  • Presentations and lectures
  • Hands on lab exercises and case studies
  • Discussions, problems, quizzes

Required Reading


Suggested Reading

  • Gertler, P.J., Martinez, S., Premand, P., Rawlings, L.B. and Vermeersch, C.M., (2016). Impact evaluation in practice, Second Edition, World Bank Publications.
  • Josselin, J. M., & Le Maux, B. (2017). Statistical tools for program evaluation: Methods and applications to economic policy, public health, and education. Springer.
  • Glennerster, R. and Takaarasha, K. (2013).  Running Randomised Evaluations: A Practical Guide.  Princeton: Princeton University Press
  • Pearl, J., Glymour, M., & Jewell, N. P. (2016). Causal inference in statistics: A primer. John Wiley & Sons.

Required Preparation



Ofer ENGEL (University of Groningen)


Practical Information





Application  via National Contact Point


4 days


ICON-INSTITUT Public Sector GmbH

Deadline: 03.05.2021