This PowerPoint presentation is an extensive guide to understanding causes in an evaluation and the difference its proper diagnosis has on the effectiveness and accuracy of an evaluation.
This excerpt is taken from slide 5 of Davidson (2009) Causal inference:Nuts and bolts
- "If an “outcome” is not caused by the intervention, it is NOT an outcome; it’s merely a coincidence
- Coincidences cannot be documented as intervention outcomes
- Therefore, causal inference is a crucial part of linking inputs to outcomes
- You do NOT necessarily need a randomised experimental design to infer causation! (although they can be a good option)'
- Causation, outcomes and coincidences
- Two applied examples: 1)Leadership development programme, 2)Performance appraisal & bonus system
- 8 strategies for inferring causation
- Hands-on practice
- Certainty about causation – how certain do we need to be?
- Guidelines for inferring causation
- Summary and further references