The selection and vetting of statistical stories

National statistical offices (NSOs) have broad mandates to report on economic and social conditions and so potential topics for statistical stories are almost limitless. On the other hand, NSOs operate with a number of internal and external constraints that affect the selection of statistical stories, the manner in which they are told and the medium of communication. This paper outlines the factors that affect the selection and vetting of statistical stories disseminated by NSOs, with particular reference to Statistics Canada. The paper presents a model of the information flowing from NSOs, focusing on products intended for media redistribution and discusses the motivation for reaching large audiences. It then examines the story selection process - proposing a simple model, introducing the players involved and describing the principles and practices of story selection at Statistics Canada. A discussion of the processes for vetting stories and maintaining institutional values follows. The concluding discussion focuses on the applicability of these observations to other statistical offices.