Did you know that about 300 million Americans went without food, water and shelter at some point last year?

I am a survivor.

If you were blessed with the prodigiously creative and cunning mind of a politician, that kind of statistic -- meaningless but technically true -- could be put to good use.

In the entertaining 1954 classic "How To Lie With Statistics," Darrell Huff writes, "Misinforming people by the use of statistical material might be called statistical manipulation … (or) statisticulation."

One of the most persistent examples of modern-day statisticulation is the sufficiently true claim that 46 million (it becomes 50 million when senators really get keyed up) Americans don't have health insurance.

Set loose on the public's compassion, this number is a powerful tool in the hands of eloquent orators, such as President Barack Obama, when they're peddling government-run health care reform. And no matter how often the figure is debunked, no matter how many studies point to its inexact nature, it's just too politically inviting not to embrace.

Wherever we stand on health care policy, surely we can admit that it's just as important to understand why Americans are uninsured as it is to get a handle on how many Americans are uninsured.

It is true that the 46 million figure is based on unreliable Census Bureau data. But even the less unreliable Congressional Budget Office puts the number at about 31 million. And even that number, former CBO Director Douglas Holtz-Eakin claims, is an "incomplete and potentially misleading picture of the uninsured population."

For one reason, the uninsured figure counts all Americans (and illegal immigrants) who have been uninsured for any amount of time during a year, even if they happen to be between jobs or changing insurance plans or on family visits to Guatemala.

According to the CBO, 45 percent of the uninsured are uninsured for four months or less, which seems like a pretty positive number to me. Continued...