PROMOTING BAD STATISTICS.
Society
| March 01, 2001 | Best, Joel
In contemporary society, social problems must compete for attention. To the degree that one problem gains media coverage, moves to the top of politicians' agendas, or becomes the subject of public concern, others will be neglected. Advocates find it necessary to make compelling cases for the importance of particular social problems. They choose persuasive wording and point to disturbing examples, and they usually bolster their case with dramatic statistics.
Statistics have a fetish-like power in contemporary discussions about social problems. We pride ourselves on rational policy making, and expertise and evidence guide our rationality Statistics become central to the process: numbers evoke science and precision; they seem to be nodules of truth, facts that distill the simple essence of apparently complex social processes. In a culture that treats facts and opinions as dichotomous terms, numbers signify truth--what we call "hard facts." In virtually every debate about social problems, statistics trump "mere opinion."
Yet social problems statistics often involve dubious data. While critics occasionally call some number into question, it generally is not necessary for a statistic to be accurate--or even plausible--in order to achieve widespread acceptance. Advocates seeking to promote social problems often worry more about the processes by which policy makers, the press, and the public come to focus on particular problems, than about the quality of their figures. I seek here to identify some principles that govern this process. They are, if you will, guidelines for creating and disseminating dubious social problems statistics.
Although we talk about facts as though they exist independently of people, patiently awaiting discovery, someone has to produce--or construct--all that we know. Every social