social proof

Those numbers sound incredibly fishy.

93% of high school students believe this? More than nine out of ten? Really? Even in states like California, where smoking almost everywhere is banned?

Could nine out of ten high school students really be so blind to basic statistics AND personal anecdotal experience?

If you ask the average individual to catalog all the smokers vs non-smokers they know, it is doubtful they would peg more than half as smokers. Maybe so for a small percentage of families, but more than half? Come on.

The extrapolation doesn't make sense in terms of teen psychology either. Why would smoking gain attractiveness just because people from all walks of life do it? How would it matter to a teen that the bus driver and the mailman are likely to smoke?

If teens really believed that 'everyone' smokes, that would arguably lessen the desire to smoke by reducing its counter-culture appeal.

The fact that certain "cool" minorities of society appear to be smoker-biased is a different story. Actors, musicians, fashion models and other candidates for adolescent worship no doubt have an outsized influence. (And in fact, the adolescent influence might work both ways; the drive to be 'cool' might encourage actors, musicians etc to smoke in the first place.)

As for overall results and conclusions, though, I would bet the compilers of that study are smoking something. The odds of the study being flawed are higher than the odds of such backward conclusions being correct.

One more thing re: "On average, they believe over half (56 percent) of Americans are smokers."

If 56 percent is the average guess, that implies a distribution of guesses above and below... which in turn implies that a significant portion of teenagers believe that far MORE than half the population smokes.

To get an average guess as high as 56 percent, I would imagine 30 or 40% of the teenagers surveyed would be throwing in ridiculously high guesses... on the order of 70 or 80% of the population. Which, when you take out people under the age of 16, pretty much covers all breathing adults.

Where did they take this frickin' survey? Russia?

Excellent!

So, based on figure 2, a statistically significant percentage of teenagers believe that 100% of the population smokes.

Meanwhile, a smaller (but still statistically significant) group believes that 0% of the population smokes.

Right. And such answers give no clue that the data is laughably suspect?

This study should be fodder for another study: the tendency for bored teenagers to give researchers dumb answers. Or how about "sample bias gone wild."

p.s. I don't have any smoking-related vendettas. I just find these off-the-wall research gaffes amusing.

Are we sure they didn't get this backwards? As an ex-smoker I sure I would have overestimated the percentage of smokers to make myself feel better ("hell. everyone does it). Do smokers naturally overestimate the number of smokers or does overestimation make you more likely to smoke?

I suspect this data is more coincidental than anything. Studies have shown kids who smoke are more likely to smoke themselves.

If you're a kid in a house with parents who smoke chances are you're around people who smoke (family, parent's friends etc.) - this would easily skew a kid to believe more people smoke then on average as , in general, the percentage of people around them that they smoke would be higher.

The opposite would likely be true for kid's of non-smoking houses.

As for the difference between % of students who smoke and % of americans who smoke - from what I remember, in high school the delineation between smokers and non-smokers was quite clear. Kids who smoked were at the curb between classes/at lunch and the non-smokers weren't. So smoking kids would get a pretty good picture that a smaller percentage smoked. Unlike in the rest of life where smokers and non-smokers are "mixed" more often.

Kids are the future, fortunately their present is theirs alone. Unless of course you market to them...