This paper is from Session 7F: Challenges for statistics in the past and in the future
which comes under Topic 7: Statistical literacy in the wider society
Paper 7F2 (Thursday 12th, 11:00-12:30)
Statistics, Reality, Truth
Presenter
- Giovanni A Barbieri (Istat (Statistics Italy), Italy)
Abstract
Among its fundamental principles, the European statistical law requires reliability, "meaning that statistics must measure as faithfully, accurately and consistently as possible the reality that they are designed to represent”. The reference to reality is necessary if one wants to avoid statistics to be just one of the many narratives competing for attention. On the other hand, the statement exposes statistics to a risk of naif positivism, as it was in Quételet’s time. The formulation itself is ambiguous: one thing is to represent, another to measure. How can we avoid being smashed between the anvil and the hammer? Do we need an epistemology for official statistics? And – even more crucially – how can we communicate the scientific principles at the base of official figures?