The Hidden Ways Manipulated Science Harms Our Health, From Measles To Organics
Lynne Peeples, Huffington Post, Feb 11, 2015
"The fact is that, despite its mathematical base, statistics is as much an art as it is a science," the author Darrell Huff wrote in his influential 1954 book How to Lie with Statistics. "A great many manipulations and even distortions are possible within the bounds of propriety."
Perhaps no one understands this power better than corporations seeking financial gain.
"Industry is about five light years ahead of the scientists in their ability to use data to manipulate," said Moon. "They spend a lot of time and money figuring out how to do it."
Steering public perception and policy by distorting data was a common tactic in Big Tobacco's playbook, for example. That move has since been borrowed by some chemical manufacturers intent on prolonging their products' lives on the market. As industry leaders know, there's a degree of complexity and creativity inherent in all research methods, and often it's not difficult to come up with numbers that say what you want them to say. The impact of such misconduct can linger, even long after the tainted research has been debunked.
One strategy often used to obscure inconvenient truths is to water down the data. The 16 Cities Study of secondhand cigarette smoke, a federal government project that began in 1996, is a classic case. The study's authors concluded that smoking workplaces posed negligible exposures to non-smokers. Researchers who later re-evaluated the study, however, came to a different conclusion. In their follow-up, they noticed that the study's definition of a "smoking workplace" included buildings where smoking was restricted to designated areas, or where no smoking was actually observed. Reorganizing the data to account for the actual amount of smoking in the workplaces, the reviewing authors found that smoke-free workplaces would, in fact, significantly reduce secondhand smoke exposures for non-smokers.
The reviewers also noted the undisclosed involvement of the R.J. Reynolds Tobacco Company and the tobacco industry's Center for Indoor Air Research in the 16 Cities Study. The project, the reviewing authors concluded, had been "specifically conceived and designed to forestall regulation of workplace smoking."
Once again, the same data can tell very different stories depending on how the statistics are presented. "The numbers have no way of speaking for themselves," the author and statistician Nate Silver wrote in his 2012 book The Signal and the Noise. "We speak for them. We imbue them with meaning."
Several studies, including surveys of research on food products and secondhand smoke, suggest that when a third party has a hand in funding scientific research, it tends to influence the ending of that story -- whether the scientists involved are conscious of it or not.
Devra Davis, president and founder of the consumer advocacy group Environmental Health Trust, has suggested that past research concerning cell phone radiation involved similarly manipulative methods. In one 2011 study, financially supported by the cell phone industry, European researchers determined that kids who averaged one or more weekly cell phone calls over a period of at least six months were not at an increased risk of developing a brain tumor, compared to peers who were non-users. But Davis and other experts have argued that no one could reasonably expect to find a link, given such limited cell phone use and such a short time frame. Brain cancer can take decades to develop.
"You end up with results inconclusive by design," Davis told HuffPost.