Past Recipients‎ > ‎

The use and abuse of statistics in the media

posted Jul 8, 2010, 9:04 PM by Glenna Buford

Rebecca Goldin

George Mason University

AWM-MAA ETTA Z. Falconer Lecture
August 2008

Abstract.

News increasingly depends on a careful dissection of numbers. Statistics are everywhere, from how many people are not covered by health insurance to whether Vitamin E is good for you or not. Yet for being so prevalent, statistics are awfully badly understood by the general public.

In this talk, I'll illustrate how the press can misuse and even abuse statistics. Since news sources are the main avenue by which the public understands many public health issues, these misguided representations of science can actually shape public policy, legislation, and individual choices. We will see why it is so important that media writers understand basic concepts from statistics, epidemiology and even toxicology. I will also show how powerful the work can be when the press goes beyond politics and morality to get the science right.

These examples come from my experience as the research director for Statistical Assessment Service (STATS), a nonprofit media education and watchdog group affiliated with George Mason University, where I am a professor of mathematics as well. STATS takes critical aim at the poor use of statistics to justify false claims or to back-up ideological agendas, while serving as a resource for journalists and producers who want to engage in high-level responsible reporting that takes into account what the science says, what it doesn't, and what it can't.

Mark Twain popularized the quote that "There are three kinds of lies: lies, damn lies, and statistics." While this quote suggests the scary idea that statistics can be manipulated to say anything, I will argue that statistics can tell us lots of useful things when used appropriately, and that the more the media does this for us, the more educated we can be as news consumers, and the better we will be at truly evaluating risk for ourselves and others.

Comments