Survey Says....Who Cares?

I read an interesting column in the paper over the weekend, by the New York Times' John Tierney (which, thankfully, was reprinted by the Dallas Morning News so that you don't have to be a paid NYT Web subscriber to read it). The subject: Polls. Here's a sample:

The voters have spoken. If only they made sense.

You may have heard that American voters are disappointed. They are disappointed with Dennis Hastert and the rest of Congress. They are very disappointed with the war in Iraq. They are very, very disappointed with President Bush.

I share their unhappiness, but I must confess to one further regret. I am disappointed with the voters – or at least the ones who show up in public opinion polls. They keep complaining that Washington doesn't understand what they want, but who on earth could? Early in the Iraq war, Americans told pollsters they favored it and considered it a major part of the war on terrorism. Then they decided the war was a mistake and didn't reduce the risk of terrorism. Yet as they got angrier and angrier at Republicans for making a mess of Iraq, they kept telling pollsters that they didn't trust the Democrats to do a better job of dealing with terrorism.

Here's my take on the subject: Sure, the voters may be confused, but the real confusion here is why anyone would put stock in a poll, much less use said poll to decide for whom to vote or, if one is an elected official, what action to take on a specific policy or piece of legislation.

I'm sorry, but I never have bought into the whole concept of polls, unless they're being used for entertainment purposes (and to me, their entertainment value is on the level of, say, one's daily horoscope, but so is their accuracy). I don't accept that using the responses of a sample can predict the behavior of a population, no matter how scientific the sample appears to be. (This is one of the big reasons I abandoned my ill-chosen Ph.D in music education; the other one was, obviously, that I had become such a jazz and performance guy by the time that I left North Texas that I had no desire to put my horn down for a year and go live on the fourth floor of Willis Library doing research. But high up on the list of reasons had to be that I don't see the point in using research methods to create these artificial subgroups of the population who may or may not do things exactly as their peers do.)

I might be really stubborn on this, but perhaps one of the big reasons I don't believe in polls is because I have never participated in one. This is partially of my own doing; even back when I used to answer my home phone, a solicitor would never get past his/her inevitable mispronunciation of my last name ("Good evening, Mr. McNeary, how are you today?") before I quickly ended the call. (Indeed, an article I read a few years ago noted that this is becoming a problem for pollsters; those who have ditched their land lines for cell phones don't get to participate in polling because solicitors aren't allowed--thankfully--call cell customers, so an entire demographic of the population is automatically excluded from their surveys.) In the same way, I've never had a Nielsen box on my TV set, so I've never put too much stock in those ratings, since I've never been given a vote in that process.

I fully realize that there's no way that a poll could be 100% accurate because of the sheer impracticality of getting everyone's opinion on so many issues (i.e. if the Nielsen boxes were hooked up to every TV set in America). I'm not saying that polls shoudn't exist, but it disappoints me that so many people who make important decisions would place as much value on polls as they do, and it really annoys me to see their results be used as the basis for so many stories in the media.

So why don't I think that samples can reflect a population? It's simple: People act more as individuals than they do as part of a group (and those who actually do subscribe to identity politics are, in my humble opinion, on the wrong path). Just because four or five 37-year-old suburban women with three kids who don't work outside the home think that Political Party X would do a better job with the economy than Political Party Y doesn't mean that all people sharing that demographic feel the same way, and I sure don't think that anyone else who shares my demographic automatically speaks for me, nor should they be allowed to do so.

I doubt that the polls are going to go away anytime soon (unless everyone gives up their land line for a cell phone, at which point the polls will suddenly take the form of spammy emails, some of which will never be read by their intended targets due to filtering sofware...you get the idea), but right now, I for one am growing tired of hearing about them every day. If a wise politiican wanted to get a better idea of the "will of the people," perhaps talking to more of those people in person would be a better idea than relying on pseudo-science for the answers. In the meantime, I'll probably still read some of these poll-driven articles, just as I read my horoscope every day for fun....and, as I said, I'll put equal stock in both.

I'm using the letter W in this blog post; I hope they don't sue me too: The University of Wisconsin has threatened legal action against an Iowa school district because the district's "W" logo looks too much like the one used by the university.