Maybe you've heard that the abortion rate has risen by 25 percent during the Presidency of pro-life George W. Bush. Or that 100,000 Iraqis have been killed during the war. Or that before long, Social Security will be $11 trillion in the red. In recent months, all these stats have turned up repeatedly in heated political debates. And all of them are either misleading or just flat wrong. Pick them apart and you'll see that solid-sounding numbers can be as slippery as a greased pig. Who's feeding us these phony figures? Hardcore partisans who twist numbers to score a point. Call it twististics.
Take the abortion rate. Last fall ethics professor Glen Harold Stassen at Fuller Theological Seminary in California wrote an op-ed article citing research he'd done supposedly showing abortions are rising again after a decade-long decline. Democrats like Howard Dean and John Kerry used the study to jab at George W. Bush, saying his pro-life policies clearly weren't working. "In fact, abortion has gone up in these last few years," Kerry said on NBC. Dean went further, coming up with a hard number: "You know that abortions have gone up 25 percent since George Bush was President."
Stassen hadn't even used the 25 percent figure. A bigger problem, though, was that his findings turned out to be seriously shaky -- based on only 16 states. The Alan Guttmacher Institute, the source for Stassen's data, has compiled its own thorough study using government data for 44 states. Its conclusion? The abortion rate is still falling.
You could be forgiven for believing politicians when they throw out specific stats. "There's a tendency to assume that a number is a little nugget of truth, that it's real in the sense that a rock is real," says Joel Best, a University of Delaware professor and author of Damned Lies and Statistics.
The politicians and special- interest partisans sure know this. Okay, they might not always realize they're using funny numbers, but the people whom they hire to provide the research sure do.
Look at how the Bush Administration has spun the training of Iraqi security forces -- a critical barometer of our success in that country. Back in March the Pentagon said that more than 140,000 Iraqi police and soldiers had been trained. That sounded impressive. But a nonpartisan government study pointed out that the figure included possibly thousands of Iraqi policemen who had gone AWOL, and thousands more with just a few weeks of basic training. The chairman of the Joint Chiefs of Staff told the Senate that the number of Iraqis who "can go anywhere in the country and take on almost any threat" was around 40,000 as of last January. Seven months later, Sen. Joe Biden was insisting we had "fully trained" fewer than 3,000 Iraqis. Whatever the right figure, we're not dealing with rounding errors.
Of course, critics of the Iraq war haven't played it straight either. A team of scholars made a splash with a 2004 study published in a British medical journal claiming the war had killed 100,000 Iraqis. That number caused an uproar, especially among antiwar liberals. But a closer look showed that the number of deaths was wildly uncertain. The same study indicated it might range from 8,000 to 194,000. When does uncertainty become a meaningless guess -- or political math?
Likewise, both sides have dirt on their hands in the debate on Social Security reform. At a public forum back in January, President Bush made much of an Administration estimate that the Social Security system faces a future shortfall of $11 trillion. That sounds terrifying -- until you realize that this mega- figure projects the future of Social Security not over a generation or so, but into infinity. The nonpartisan American Academy of Actuaries has called such projections worse than useless, and "likely to mislead anyone lacking technical expertise" into thinking the system is in worse shape than it is.
But liberals have some statistical explaining to do too. Last winter, the activist group Moveon.org ran an ad implying that Bush would cut Social Security benefits by 46 percent. The ad ignored Bush's repeated assurance that he would not cut benefits for anyone at or near retirement -- and, indeed, the 46 percent figure referred to the projected growth of benefits for retirees some 75 years down the road. It also ignored estimates that, if we do nothing to fix Social Security, by mid-century benefits will have to be cut by around a quarter anyway.
Then there was the Social Security benefit calculator posted at several Democratic Senators' websites. The calculator showed deep benefit cuts under Bush's plan, but only if you assume that the stock market -- where Bush wants to allow some Social Security funds to be invested -- will grow at just 3 percent over inflation. What the calculator fails to calculate is that, over the past century, the stock market has averaged more than twice that level of growth.
I could go on and on: misleading divorce rates, exaggerated counts of homeless veterans, inflated numbers of illegal immigrants. There's enough baloney out there to start a deli.
So what's a person to do who just wants to get the straight facts? First, assume that the numbers you're given have been through the spin cycle. If it's a hot-button political issue, they probably have. Double-check suspicious statistics by looking for multiple, nonpartisan sources. A good place to start is the Annenberg Public Policy Center's FactCheck.org.
My own assumption, as a reporter, is that at least 50 percent of the stats I see don't tell the whole story. But you'd be smart to check that number out.


From



Advertisement 



































Your Comments
See all
...
Post your commentCancel