Plan

• Stock taking
• Challenges
• Models for a community of practice

Stock taking

We (you!) are in a very good place

• Talented people
• Institutional support
• Momentum

A moment for consolidation

Some recollections

• DDR, Sierra Leone 2003 / CDR started (2005-)
• Many early failures: Sudan 2005, Haiti 2006, Unicef Mano 2008
• Resistance, protectionism, inertia
  • HQ, field crossed wires
  • Confusion about the ability to identify causal effects
  • Resistance to cross-case inference
• On our side: Weaker methods, lower standards

Change makers

• Pioneers: 3IE, IRC, DIME
• DfID did remarkable things

Now

• Rich set of high quality studies

• Delivered by broad range of actors

• Being picked up and used

• Broad agreement on standards of transparent production, analysis, and sharing

• Struck by how little fundamental disagreement here

Some challenges

• Motivation:

  • Accountability and learning; weight shifted to learning

• Execution:

  • Execution: It’s so costly! (but Florence: not so much!)
  • The basic idea is so easy: But it’s very very hard to get this right. Seeing creative uses of partnerships.
  • Ethics: a lot of thought going into this

• Inference

  • Welcome focus on the “internal” aspect, but
  • Fundamental heterogeneity (external validity) is a great challenge

• Use

  • We already know the answers!
  • Still: now more optimistic than ever

A focus on external validity

Three approaches:

1. Theoretical structure
2. Systematic cross case variation

Both require a form of coordination

A focus on external validity

Consolidation

Light model

Continue as you are; things may continue to improve

• But gains from coordination not maximized
• Risks of loss of what has been gained

Ambitious model

Form some kind of more institutionalized community of practice for humanitarian impact evaluation (COPHIE!)