Cherry Picking Statistics

There has likely been no time in history when it is easier to fact check a claim than our current age, thanks to the Google. That being the case, I'm really shocked at how often people attempt to skew statistical data that is easy to reference and double-check.

Think Progress has an interesting article discussing Sen. James Inhofe's recent declaration that defeating Obama's health care plan would be a 'huge gain' for the GOP for the 2010 election cycle. I want to zero in on an offhand talking point Inhofe delivered during his appearance on Janet Parshall's radio show:

But every day, they [Democrats] lose votes, because people find out what it is [the health plan], what it’s going to do, and what it’s not going to do. When you tell people that the
mortality rate in Canada is 25% higher for breast cancer, 18% higher for
prostate cancer, you know, they say why in the world would we emulate a system
like that?

Curious, I Google'd, "are mortality rates for cancer higher in Canada than USA," and the fourth listing was a CTV article from the CTA.ca News Staff that include this incredible factoid:
In a report on worldwide cancer survival rates, Canada ranked near the top of the 31 countries studied with an estimate five-year survival rate of 82.5 per cent.

For breast cancer, Cuba had the highest survival rates -- another country with
free health care. The United States was second, and Canada was third, with 82
per cent of women surviving at least five years.

Emphasis mine. So, Inhofe used the USA's excellent survival rates for breast cancer as a way to demean the quality of health care in Canada and universal health care (or socialized medicine) in general. He fails to mention that not only is Canada ranked third in the world in breast cancer mortality rates, but the highest ranking country is communist Cuba.

We again must ask ourselves if Inhofe is simply clueless about the state of affairs regarding cancer mortality rates, or if he is a shameless liar who cherry-picked a statistic to forward an argument that the greater corpus of the statistical data absolutely refutes.