Saving More Gas than a Prius

Hybrids are popular, hybrids are hip. A Toyota Prius can get an impressive 50 m.p.g., prompting California to give them special driving exemptions: they can use the carpool lane with nothing more than the driver.

But that's not the road to big savings in gasoline.

It sounds like it - 50 mpg is double that of most passenger cars - but the measurement itself is misleading, as you can see from this little quiz:

     Which saves more gas?

• Replacing a 10 mpg S.U.V. with a 20 mpg station wagon?

or

• Replacing a 20 mpg sedan with a 50 mpg hybrid?

Surprise! It's the first option.

The S.U.V. uses 10 gallons to go 100 miles, while the station wagon (and sedan) use 5. The hybrid uses just 2, but that's a smaller savings than replacing the S.U.V.. If you drive a 200 miles a week, moving from the S.U.V. to the station wagon saves you 10 gallons, while moving from the sedan to the hybrid saves you 6 gallons.

It's the low mileage vehicles that add up. Moving from 10 mpg to 20 mpg saves a lot more than from 20 mpg to 40 mpg or even 50 mpg.

This would be easier to see, argue authors Richard Larrick and Jack Soll, if we replaced mpg with "gallons per mile," a measure that better shows the cost and savings of improved efficiency. You can read more about their idea or try a mini-quiz on the topic.

Next week's blog will address Microsoft's newly announced strategy for Windows 7: Microsoft Hobbles Your Fun.

What Science Offers Politics

Can science help us make decisions?

While science does not tell us what we should do, it can help us gauge the impact of our choices by looking at existing data. But it can do so only if we understand how science works, and too few Americans do.

Sadly, author Michael Crichton is not one of them. After his death earlier this month,  the Wall Street Journal offered an excerpt of his speech("Aliens Cause Global Warming") about the "hoax" of global warming, the premise of one of his final books.

In it, Crichton fails to understand how science deals with uncertainty. To him, any admission of uncertainty is tantamount to "mere guesswork," and research filled with uncertainty, like the search for extraterrestrial life, is "unquestionably religion."

He may be a doctor, but he's no scientist.

The kind of "proof" he seeks isn't available to science at all. Proof is reserved for the rarefied world of mathematics, and the best we can ask for in science is evidence.

Is there evidence for global warming? Sure, but that's not the point. If he wanted to argue against the evidence, fine. He could say, for instance, that deep-sea temperature data isn't reliable, or we need to include additional data from such-and-such a source. But to say that because we can't prove a link between cause and effect is to misunderstand the nature of science. Toss it away, and we lose an invaluable tool.

This affects our world in important ways, every day. Here are a couple of rulings that affect what you can buy and the land you live on.

On December 21, 2004, the U.S.  Department of Health and Human Services recommended against allowing the import of prescription drugs from Canada because it  “could not be sure” that the imported drugs would be safe.

The next day, the Forest Service eliminated its policy for preparing Environmental Impact Statements, as well as the requirement for logging to protect all viable species in the National Forests, since such statements could not say "with certainty" what sort of harm would follow logging and development.

The above observations, from a brilliant  paper by Freudenberg, Gramling, and Davidson,  reflect opposing uses of scientific uncertainty for political ends. In the first case, uncertainty prohibited an action that couldn't be proven safe and on the other one that couldn't be proven unsafe.

Science doesn't offer proof. Neither does politics, and to ignore the science altogether is to suggest that politicians and interest groups can make decisions with certainty even when scientists cannot.

What science offers us is evidence, and it can help us answer questions like "If the earth were warming significantly, what would we see?" and "Given our current understanding, which stars in the sky are most likely to have planets in orbit, and which planets are best suited to retaining an atmosphere?" Science can tell us, "These drugs can be easily checked for safety, and these cannot. These drugs are easy to counterfeit and these are not." Science can tell us if you allow x amount of pollutant to a river, y fish are likely to survive.

Science won't decide what is right or wrong. But it can ensure that when we make our choices, we aren't doing it on the basis of "mere guesswork."