Ever since the rise of the Internet's accessibility, politicians and transportation planners have touted telecommuting as an antidote to crowded highways, poor air quality and even low productivity.

Their efforts were rewarded, as statistics showed a significant rise in the number of people working from home. By 2001, more than 15 percent of all Washington workers, nearly 400,000 in all, were telecommuting at least one day a week. Success was at hand.

Only thing was, the numbers were inflated -- by nearly 100,000, or one out of every four telecommuters. That's because, until now, any worker who did not spend a full day at his primary office qualified as a telecommuter, regardless of how much time was spent driving.

The definition meant that sales and repair people who drove around the region all day were counted as telecommuters, as were lawyers, consultants and others who worked at client sites and employees who spent half a day at home and then drove in to the office for the other half.

A new definition of telecommuting, announced yesterday by the Metropolitan Washington Council of Governments, restricts those counted to people who spend minimal time on roads because they "work at home or at a telework or satellite center during an entire workday."

"It's disappointing that that's the case," Rep. Frank R. Wolf (R-Va.) said of the revision downward. Some of those workers "should have been counted earlier, but no longer," he added. "That's just common knowledge."

The revised numbers aren't a blow just to politicians. They also hurt the region's efforts to meet air quality standards next year because telecommuters are counted as a credit against pollutants.

Environmental groups alerted the regional planning body to problems with their telecommuting numbers years ago. In a letter sent to the region's COG dated Sept. 10, 2001, Michael Replogle of Environmental Defense wrote that the criteria "appears to overstate the effectiveness of telecommuting programs."

Replogle said the false numbers are "enough to make a difference in whether we meet federal air quality standards for ozone as required in 2005." He also said that overstating the number of telecommuters cut down on efforts to encourage transit, walking, biking and smart growth strategies.

Ron Kirby, transportation planning director for COG, acknowledged that the revision "will have an immediate effect on our estimate for emissions for 2005" but said that he didn't think the difference would be enough to make the region fall short of the requirements.

Kirby said that the loose standard for telecommuting was used for years because, "frankly, people just didn't think about it. It was something that kind of evolved over time."

After accounting for the definition change, the number of telecommuters in a 2001 survey of commuting patterns was reduced from 386,650, or 15 percent of all commuters, to 290,319, or 11 percent of commuters. The number of telecommuters rose to 318,130 in 2004, or nearly 13 percent of commuters, in a spring survey of 7,200 households.

The commuting study is done every three years by COG; other results will be released in the fall, Kirby said.

Regional leaders were pleased that the percentage of federal teleworkers jumped from 6.9 percent to 11.8 percent between 2001 and 2004. The federal government, the region's largest employer, has lagged well behind its goal that all employees deemed eligible to telework be able to do so by the end of next year.

The revised telecommuting figures come at a time when some are again questioning whether the practice is the solution to traffic and pollution problems that political leaders have been proclaiming it to be.

Few dispute that keeping people at home and off the roads is a good thing, but some also say that telecommuting serves as a major incentive for people to live farther from their jobs. Workers who would be turned off by an hour-long commute might find it palatable only three times a week. That can increase their time and distance on roads when they do go to the office and aggravate suburban sprawl conditions.

"On one hand, it can lead to fewer auto trips on the road," said Stewart Schwartz, executive director of the Coalition for Smarter Growth. "On the other hand, telecommuting allows you to live on a mountain in Shenandoah and commute in one day a week and contribute to sprawl."

The report showed that those who lived farthest from their jobs telecommuted most. Seventeen percent of those who travel 30 miles or more telecommute, compared with 14 percent who travel 10 to 29 miles and 10 percent who go fewer than 10 miles.

Others questioned the benefits of telecommuting by noting that many telecommuters run additional errands in their cars on days that they work from home, crowding local streets.

That's not today's worry, said Gerald E. Connolly (D), chairman of the Fairfax County Board of Supervisors and a supporter of telecommuting. He said the immediate concern is increasing the number of those who telecommute.

"It's disappointing," Connolly said. "But the numbers are the numbers, and we have to deal with the reality we face."