We always hear that four out of five dentists recommend chewing a particular brand of gum or using a certain toothpaste. Yet there’s always one dentist who doesn’t buy into the group-think. Why is there never unanimity? And if one in five don’t recommend a particular brand of mouth wash… well that means there are thousands of dental professionals who wouldn’t recommend it! Well, the truth is that these polls don’t actually represent what they want you think they represent. It’s actually a time-honored obfuscation of truth in American marketing, and it has been used since at least the 1950s (at one point four out of five doctors preferred Camels).
Advertisement is regulated by the Federal Trade Commission but is subject to some legal grey areas that exist where government regulation of commerce meets common law and the First Amendment’s Free Speech Clause. The federal Truth in Advertising Act imposes restrictions designed to ensure that scientific statements are backed by at least some kind of study. However, the law also recognizes certain statements as ‘puffery,’ assertions no sane person would believe to be fact. If you hear words like ‘clinical study’ or ‘national survey’ you have a reasonable belief that some kind of work was done (though that’s no assurance that the work was reliable). The ‘four out of five’ trope is broadly phrased ad-speak that aims to create the illusion of statistical analysis without over-promising much in the way of evidence.
It is often important to pay attention to what exactly is being said in the commercials. For example, if the ad states that four out of five professionals recommend a certain product, they could have only asked five select people, chosen in advance or cherry picked. The ad could also claim that four out of these five dentists recommend flossing with picks, and there are five men in white lab coats on the screen. That could mean four of the dentists pictured recommend it. Or maybe the commercial tells you that nine out of ten recommend brushing your teeth with brand X. That may be true, but they could also recommend brand Y or Z. You have to imagine any dentist would recommend brushing your teeth with most brands of toothpaste.
So if these are all, in one way or another, misleading, why not make everyone agree? Probably because it’s more believable to bluff small, and a majority is good enough. While these claims sound like factual statements, the devil is usually in the details. The words are vague enough or obtusely true in a way that As Mark Twain said, there are lies, damn lies, and statistics. They can often be manipulated to reflect any number of partial truths.