Daniel Williams/Collegian
Joel Best explains how to read statistics so as not to be duped by faulty figures.
“Every year since 1950, the number of American children gunned down has doubled,” one journal wrote in 1995. According to this statement, at the time of publication, the number of children gunned down must have been about 35 trillion.
This was one of the many absurd statistics that Joel Best, a professor of sociology from the University of Delaware, gave in a lecture Monday night.
Best is the author of Damned Lies and Statistics, a book that documents such ridiculous statistics and teaches one to think critically regarding them.
“We live in a very complicated world,” Best said. “Over the last 100 years people have realized we need statistics to understand that world, but a lot of the numbers we hear aren’t very good.”
Other examples of poor statistics include claims that 150,000 women die from anorexia each year. The truth is that of 50,000 women aged 15 to 44 that die each year, about 100 die from anorexia, Best said.
Best said his goal is to help people think critically about numbers.
“We tend to assume that a number is a little nugget of truth,” Best said. “And that’s wrong.”
Rather, Best said, people should view statistics like jewels that have to be cut and shaped by human hands. In the process, some statistics can go awry.
There are two warning signs that can help people discern between good and bad statistics, Best said.
First, packaged and rounded numbers, such as one million or one billion, should be approached with caution.
“It seems unlikely to me that our social problems were coming to us in even million slices,” Best said.
The second is common sense. People should ask themselves, “How can anybody know that?”
“Every time I hear a number and think, ‘Holy cow, I had no idea it was that bad,’ it usually isn’t,” Best said.
In general, government statistics tend to be more reliable. The U.S. census has about 98 percent accuracy, using several tests to check itself.
Because his research has been so vital to understanding statistics, Best has been interviewed by both left- and right-wing talk show hosts throughout the nation. Usually the shows want him to debunk the other side’s statistics.
“One of the things that I really like about Best’s work is that he’s really interesting no matter what your political leaning is,” said Professor of Sociology Peter Blum, who invited Best to campus.
Most students who attended Best’s lecture seemed to appreciate it.
“I like the way he gives tangible examples that a person can see easily relate to everyday life, which he also does in his book,” junior Ann Schanbacher said.
Best’s research approach takes a step back from immediate social problems to examine how society views these problems.
“The kind of research that Best does looks at what happens when we don’t think critically,” Blum said. “I think that’s extremely important. Best does the kind of research that made me want to go into sociology.”