Twenty-one sources of error and bias in transport project appraisal

Twenty-one sources of error and bias in the appraisal of transport projects are identified. These relate to objectives, definitions, data, models and evaluation conventions. Objectives may be unclear, incompletely specified or inconsistent with appraisal criteria. Definitions of study areas and scheme options for testing may bias the outcome. There are multifarious sources of data and model error. Double counting, inappropriate values, and failure to balance quantified and non-quantified items can all affect the evaluation. We suggest that there is a systematic tendency to a mega-error--that of appraisal optimism. Three antidotes to this condition are briefly suggested.