Methods of Evaluating Programme Impacts

9.30am - 5.00pm, Monday 26 and Tuesday 27 March 2012 at the University of Canberra

Presented by Professor John Micklewright, Professor of Economics and Social Statistics and head of the Department of Quantitative Social Science at the Institute of Education, University of London.


This two day course considered methods for quantifying the causal impacts of social and economic programmes in both the government and non-government sectors. How many more people are in work because of a policy to help those on the margins of the labour market? Does including a free pen in a charity mail shot increase the probability of response? Does more education for girls in developing countries promote child outcomes in the next generation? The course covers a range of experimental and quasi-experimental methods of impact evaluation that can be used to answer such questions, in and outside government:

  •  randomised control trials (‘field experiments’)
  • ‘before and after’ and interrupted time-series designs
  • ‘difference in differences’
  •  matching methods (including propensity score matching)
  •  instrumental variables in regression modelling
  •  regression discontinuity design

Exercises (not computer based) and group discussions complemented presentation of the material.


More information


Participants should be familiar with the basics of statistical inference (simple use of confidence intervals and hypothesis testing) and linear regression analysis. No prior knowledge of evaluation methods will be assumed.


A detailed guide to material will be given out in the course. Participants wishing to familiarise themselves before the course with some examples of impact evaluation might consult:

Purdon S, Lessof C, Woodfield K and Bryson C (2001) ‘Research Methods for Policy Evaluation’, DWP Social Research Division Working Paper 2
Ravallion M (2001), 'The Mystery of the Vanishing Benefits: An Introduction to Impact Evaluation', World Bank Economic Review, 15 (1): 115-40, also World Bank Policy Research Working Paper 2153;
Gertler, P, Martinez, S, Premand, P, Rawlings, L, and Vermeersch C (2011) Impact Evaluation in Practice, Washington DC: The World Bank (free download – type title into Google).


The course will be taught by John Micklewright. John is visiting Downing Fellow at Melbourne Institute of Applied Economic and Social Research, and Professor of Quantative Social Science at the Institute of Education, University of London. He previously held chairs in social statistics at the University of Southampton and in economics at the European University Institute, Florence, and Queen Mary, University of London. He also worked for several years for the UN in the UNICEF Innocenti Research Centre.

General Enquiries

Ms Penny Hope
Functions Manager, Melbourne Institute
Ph: +61 3 8344 2151