Carmen Herrero
Research project
Opportunity bias : an alternative approach to fairness
Project summary
One of the most important questions in dealing with policy evaluation nowadays is about the selection of methodologies that should be used to evaluate governmental policy and societal condition in the light of well-being, going beyond GDP. Indicators of well-being appear in categorical terms, so that we face the problem of evaluating distributions of categorical data. The assessment of the degree of unfairness of an outcome distribution in the vein of opportunities is solved by measuring how different are the outcome distributions of population subgroups that have similar circumstances. We can think of the opportunity bias viewpoint as an application of a basic non-discrimination principle by which we try to ensure that any new member of society will have access to its average chances, no matter which social group (type) she ends up in. We propose to measure of opportunity bias by means of the balanced worth, which provides a consistent cardinal way of comparing distributions of categorical data (see Herrero & Villar (2016)). There are several examples of applications that can be addressed by this methodology, from opportunity bias in education, to health, life satisfaction, etc.
