Now, I don't usually venture into political posts (I'm in the same boat as Jim at Parkway Rest Stop, who discusses this very subject in a recent post), but one cannot live on smartass-ey quips alone, so indulge me while we discusss a bit of serious content, won't you? I promise: we'll soon return to the kind of posts you've come to know and love: penile humor, potty jokes, banality, that sort of thing.

Mark Twain is credited with one of my favorite quotes: "There are three kinds of lies in this world: lies, damned lies, and statistics." When one looks at all the polls, surveys and other attempts to portray the American populace's attitude on nearly any subject, it's a good quote to keep in mind--especially when you realize polls and surveys say more about the polling organization/research company than they do about the public's opinion. I say this for three reasons.

Reason #1: You can skew a poll's results with simple wording. In my evil marketing guise, I've participated in plenty of surveys, polls, focus groups, and the like, and I'll break some news to you: you can make a poll or survey say anything you want, before you ever ask a question. It all comes down to your wording. For instance, let's say you want to determine if most Americans are in favor of banning smoking in public places. Great. If you want to show that most Americans support a smoking ban, you might word your question something like "Knowing that second-hand smoke has been proven to cause cancer and other health problems, do you support a ban on smoking in public places?" If, on the other hand, you want to show most people oppose such a ban, you might say something like "Do you believe the government should infringe upon citizens' rights by banning smoking in public?"

Admittedly, this is a broad and extreme example. But even subtle differences in wording can have a dramatic impact on results. Ask potential voters: "Have you kept yourself informed on all the issues on this year's ballot?" You'll get a lot of "yes" answers. Now ask the same people: "Has your local election department kept you informed on all the issues on this year's ballot?" I guarantee you'll get a lot more "no" responses.

Point Number Two: Opinions change--and sometimes, quite quickly. Just this past weekend, CBS News released a poll showing that a Kerry/Edwards ticket would beat a Bush/Cheney ticket. A lot of folks have jumped on this, saying this spells trouble for President Bush's re-election efforts. Certainly, President Bush is facing plenty of obstacles, but polls like these aren't among them, because such polls mean nothing--and I mean nothing--right now. If you don't believe me, consider this: In July of 1988, the respected Gallup organization released a poll matching Michael Dukakis against George Bush Senior. The results of the poll? People said they'd vote for Michael Dukakis over George Bush, 54% to 37%. That's a 17-point spread.

Ask Michael Dukakis if poll numbers can change.

Point Number Three: What people say, and what people do, are two entirely different things. It's just human nature: when you ask someone a question, they will usually say what they think you want to hear. Even worse, they'll say what they wish was the answer. Ask: "Do you exercise three times a week?" I'll bet you more than half of the respondents say "yes." But tell me: does anyone actually believe more than half of us exercise three times a week? No way. People want to be part of the "good" crowd who exercises regularly, so they naturally say they do so.

So please, don't put too much stock in surveys, polls and other instruments that measure public opinion--especially if you're trying to form an opinion about something yourself. They're never impartial, even when they try to be.

And you can certainly trust what I say. After all, I work in marketing.