Aggressive, jerky driving slurps extra fuel—up to 33% more on the highway and 5% more in town. By Jessica L. Anderson, Associate Editor June 1, 2012 On a recent road trip, the 2012 Hyundai Elantra I was testing averaged 34 miles per gallon in mostly highway driving. That's pretty good—but not as good as you might expect. The Environmental Protection Agency's highway-mileage figure for the Elantra is 40 mpg. A few months earlier, Hyundai had come under fire from Consumer Watchdog, which disputed the Elantra's fuel-economy numbers after Hyundai ran a high-profile ad campaign touting them. Hyundai says it stands behind its numbers, and the EPA has since verified them.SEE OUR SLIDE SHOW: 14 Most Fuel-Efficient Cars How the tests work. This dispute over fuel economy is only one example of how the EPA's mileage numbers often don’t mirror many drivers' results. The main reason for the discrepancies, as Hyundai Motor America CEO John Krafcik points out, is driver behavior. Jeff Bartlett, of the Consumer Reports auto team, says the EPA numbers are a good guide, but "only the guy in the lab is going to get those exact numbers." Here's how the tests work. Cars are placed on a dynamometer (a treadmill of sorts) and put through the paces of five testing cycles. In years past, the EPA's tests measured mileage under standard city and highway conditions. But the agency revised its tests for the 2008 model year. So now there are tests that reflect faster speeds (up to 80 miles per hour) and acceleration, air-conditioner use, and colder outside temperatures. The test cycles cover 44 total miles altogether and take a little over an hour and a half. "This is more than sufficient to determine fuel-economy estimates," says an EPA spokeswoman. Manufacturers test their own vehicles and report their results to the EPA, which reviews the reports and verifies 10% to 15% of them. Because the EPA tests are simulated, real-world fuel economy can be better or worse than the agency's numbers. That is supported by the EPA's Web site, FuelEconomy.gov, which allows drivers to report their numbers. Advertisement Driving better results. Where and how you drive are the keys to your fuel economy. The EPA's annual fuel-cost numbers assume you drive 15,000 miles a year, 55% of them in the city and 45% on the highway. If you live in a rural area, you probably cover more miles at highway speeds; in an urban area, much less. Although my road trip in the Elantra was mostly on the highway, it takes some city driving to get out of the Washington, D.C., area. Focusing solely on the highway figures, Popular Mechanics ran its own tests on the Elantra and the Ford Focus SFE, which also claims 40 mpg. In both cases, the cars "easily cleared 40 mpg, and astonishingly approached 50," it stated in an online story in February. Terrain and weather also affect mileage. If you live in the South, for example, you are likely to get better mileage on the region's flat roads, but you may also use your air conditioner more often. Running the A/C on the highest setting can reduce fuel economy by 5% to 25%. Likewise, drivers in the Northeast and Northwest have to contend with cold winters and hilly roads. Your driving style makes a difference, too. Aggressive, jerky driving slurps extra fuel—up to 33% more on the highway and 5% more in town. If you typically emulate Speed Racer, consider that every five miles an hour over 60 mph is like paying 31 cents more for gas. To maximize fuel economy, drive smoothly and don't speed. Advertisement Automakers often hype highway mileage in their advertising, and window stickers boldly list separate city and highway numbers—but minimize the combined fuel-economy number you are likely to get in mixed city and highway driving. For the 2013 model year, the EPA is introducing new window stickers that will emphasize the combined mileage number and give a range of fuel-economy figures for categories of vehicles. That way you can compare apples to apples. Follow Jessica on Twitter and Facebook.