By BARRY MEIER

It is the mantra of the nation’s opponents of smoking: that sweeping changes in the way cigarettes are marketed and sold over the next decade will stop thousands of teen-agers each day from starting the habit and spare a million youngsters from untimely deaths.

President Clinton recently warned, for example, that one million people would die prematurely if Congress did not pass tobacco legislation this year. Sen. John McCain, R-Ariz., and the author of a $516 billion tobacco bill, has urged lawmakers to stop "3,000 kids a day from starting this life-threatening addiction."

But with the Senate beginning debate on Monday on tobacco legislation, many experts warn that such predictions are little more than wild estimates that are raising what may be unreasonable expectations for change in youth smoking rates.

The assertion that one million lives would be saved, for example, comes from a statement by the American Cancer Society last year that a 60 percent decrease in youth smoking in coming years could reduce early deaths from diseases like lung cancer by a million. But critics say the 60 percent figure was merely a target of anti-smoking advocates, with no analysis to back it up.

Social issues often spark unfounded claims cloaked in the reason of science. But the debate over smoking, politically packaged around the emotional subject of the health of children, is charged with hyperbole, some experts say. Politicians and policy makers have tossed out dozens of estimates about the impact of various strategies on youth smoking rates, figures that turn out to be based on projections rather than fact.

"I think this whole business of trying to prevent kids from smoking being the impetus behind legislation is great politics," said Richard Kluger, the author of "Ashes to Ashes" (Knopf, 1996), a history of America’s battle over smoking and health. "But it is nonsense in terms of anything that you can put numbers next to."

Everyone in the tobacco debate agrees that reducing youth smoking would have major benefits because nearly all long-term smokers start as teen-agers. But only a few studies have tried to analyze how steps like price increases and bans on advertising affect youth smoking. And those have often produced contradictory results.

Consider the issue of cigarette pricing. In recent congressional testimony, Lawrence Summers, the deputy Treasury secretary, cited studies saying that every 10 percent increase in the price of a pack of cigarettes would produce up to a 7 percent reduction in the number of children who smoke. Those studies argue that such a drop would occur because children are far more sensitive to price increases than adults.

"The best way to combat youth smoking is to raise the price," Summers said.

But a recent study by researchers at Cornell University came to a far different conclusion, including a finding that the types of studies cited by Summers may be based on a faulty assumption.

Donald Kenkel, an associate professor of policy analysis and management at Cornell, said that earlier studies tried to draw national patterns by correlating youth smoking rates and cigarette prices in various states at a given time.

But in the Cornell study, which looked at youth smoking rates and cigarette prices over a period of years, researchers found that price had little effect. For example, the study found that states that increased tobacco taxes did not have significantly fewer children who started smoking compared with states that raised taxes at a slower rate or not at all.

Kenkel added that he had no idea how the price increase being considered by Congress -- $1.10 per pack or more—would affect youth smoking rates because the price of cigarettes, now an average of $2 a pack, has never jumped so much in the United States. And he added that there were so few studies on youth smoking rates and price that any estimate was a guess.

"It is very difficult to do good policy analysis when the research basis is as thin and variable as this," Kenkel said.

Jonathan Gruber, a Treasury Department official, said that the Cornell study had its own methodological flaws and that the earlier findings about prices supported the department’s position. He also pointed out that Canada doubled cigarette prices from 1981 to 1991 and saw youth smoking rates fall by half.

Under the tobacco legislation being considered in the United States, cigarette prices would increase by about 50 percent. And while advocates of the legislation say that the increase would reduce youth smoking by 30 percent over the next decade, they say that an additional 30 percent reduction would come through companion measures like advertising restrictions and increased penalties for store owners who sold cigarettes to underage smokers and for youngsters who bought them.

The claim that comprehensive tobacco legislation would reduce youth smoking by 60 percent over the next decade is perhaps the most frequently cited number by advocates of such bills. But that figure first emerged last year in a different context and quickly came under attack.

The American Cancer Society, soon after the settlement plan was reached in June between the tobacco industry and 40 state attorneys general, said that one goal of that agreement—a 60 percent decline in youth smoking rates over the next decade—would spare one million children from early deaths from smoking-related diseases. The plan, which recently collapsed, would have raised cigarette prices by about 62 cents over a decade and banned certain types of tobacco advertising and promotional campaigns.

But some tobacco opponents soon found fault with the cancer society’s estimates. For one, those critics pointed out that the 60 percent figure represented only a target, and that penalties would be imposed on tobacco companies if it were not reached. And the cancer society, they added, had not performed any analysis of the June deal to determine whether it could produce a 60 percent decline in youth smoking.

"They basically made up the number and I think it was totally irresponsible of them," said Dr. Stanton Glantz, a professor of medicine at the University of California at San Francisco. "It is like assuming that by snapping our fingers we could make breast cancer go away."

In a letter to Dr. Glantz, Dr. Michael Thun, the cancer society’s vice president for epidemiology and surveillance research, acknowledged that the group’s statement was based on an "if-then" projection, rather than an analysis of whether the proposal’s programs would accomplish that goal.

"The way the number was derived has nothing to do with what will effectively get us there," Dr. Thun said in a recent interview.

The new 60 percent estimate is based on a different formulation. But it, like the cancer society statistic, also coincides with a target for reducing youth smoking that would result in industry penalties if not reached. And along with questioning the impact of price on reaching such a goal, experts are at odds over whether advertising bans and sales restrictions would produce the projected 30 percent drop in youth smoking.

In California, for example, youth smoking began to decline in the early 1990’s, soon after the state began one of the most aggressive anti-smoking campaigns in the country. But it has begun to rise again in recent years.

Dr. John Pierce, a professor of cancer prevention at the University of California at San Diego, said he thought that reversal might reflect the ability of cigarette makers to alter their promotional strategies to keep tobacco attractive to teen-agers even as regulators try to block them.

For their part, cigarette makers, whose internal documents suggest a significant impact on youth smoking from price increases, appear happy to play both sides of the statistical fence.

Last year, they estimated that the price increase in the June plan would cause sales to drop by nearly 43 percent among all smokers over a decade. But now that Congress is considering raising prices by twice that much, producers have turned around and said that higher prices would undermine, rather than help, efforts to reduce youth smoking.

Steven Duchesne, an industry spokesman, said tobacco companies thought that high cigarette prices would encourage those in the black market to target teen-agers.

"Smugglers would sell cigarettes out of the back of trucks without checking ID’s," Mr. Duchesne said.