From the Editor: No Joke

The American Association for Public Opinion Research (AAPOR) held its annual conference there, always a memorable experience. At my first AAPOR conference, also in Florida, one of the leading lights of survey research wandered over from the swimming pool to my booth in the exhibit hall wearing nothing but a Speedo. And dripped on my brochures.

My therapist bought a Porsche that year.

But, seriously, folks... I'm just being silly, and there is nothing very silly about an AAPOR conference.

Each May, an impressive number of survey researchers (around a thousand this year) gathers to present their work to each other, exchange ideas, and discuss the issues of the day surrounding the art and the science of public opinion research.

Less than a decade ago, I did not know the AAPOR conference existed. I barely knew public opinion polling existed. I had no motivation whatsoever to go anywhere near it, either—I just wanted to get away from Philadelphia, and, well, public opinion was the field in which the first available job presented itself.

It is not hard at all for me to remember those days, and the almost total blankness that prevailed in the part of my brain where public opinion and polling would later take up residence. It was a complete revelation to learn about the practice and the history of polling, the statistical principles behind it, its applications and its uses and abuses, and slowly, over the subsequent years, to make the acquaintance of an entire community of people who make survey research their livelihood.

I can still easily relate to the many Americans who are now where I was back then—only vaguely aware of polling, suspicious of the little I did know, inclined to equate random sample telephone polls with the product surveys that appeared in bulk-mailed coupon packets and the nosy questions asked by people with clipboards who waylaid me in shopping malls, and in complete ignorance of the provenance of the data that occasionally punctuated news reports under names like Gallup, Pew, Kaiser, Harris, and the news organizations themselves.

A few years ago, a columnist in my local paper wrote, "I've often wondered why we need political polls in the public domain... Do voters really need to know who's ahead and who's behind when they cast their ballots?" The number of levels on which this statement demonstrated the widespread lack of comprehension about opinion research was rather breathtaking.

For one thing, there is a lot more to polling than trying to predict the winners of elections. Polls encompass hundreds of topics and explore the opinions and attitudes of the American public in many areas of politics, as well as society, science and culture, and numerous aspects of everyday life. Done properly, they are a rich source of insights into the minds, values, and condition of the people as a whole that can, at present, be gained reliably in no other way.

Polls also play an important role in safeguarding the democratic process. Before exit pollsters began asking voters why they voted as they did, for instance, there was nothing to prevent winning candidates or parties from claiming a mandate from the people on issues such as taxes, Social Security, or waging war when their victory might not have been based on any such thing.

Are polls perfect? No, far from it. As any professional pollster will readily agree, there's a lot of garbage out there, ranging from push polls to junk polls to polls that are blatantly partisan and taken for no other reason than to manipulate the system. Even good polls are no good if they are not used with the utmost care. No matter how scrupulously designed they may be, tools that measure anything having to do with human beings will always be flawed. I would wager that few, if any, medical tests are right a hundred percent of the time. Does that mean that all medical testing should be stopped? Of course not.

It means that doctors should use more than one test whenever indicated, consult with other doctors of excellent reputation who might offer different interpretations of the results, keep up on their background reading, and always, always work toward developing better tests.

This, in a nutshell, is what goes on at AAPOR conferences, and whenever professional pollsters come together to talk about their trade.

People should know this; and, to take the medical analogy one step further, the patients should also take some responsibility, by becoming informed and involved in their own care. When journalists make the effort to educate themselves about polling, they are able to distinguish the legitimate polls from the dross and not be purveyors of the rubbish some of them so rightly condemn. When citizens stay attuned to politics and current events, take the time to listen carefully to points of view other than their own, and apply critical thinking to what their leaders and fellow citizens tell them to believe, they will become more capable of understanding the role of public opinion in a democratic society, and recognize polling as the imperfect but vitally important tool that it is.

Public opinion polls should be a blessing, not a curse, to the health of the body politic, and they will be as long as we—pollsters, press, and public alike—recognize their limitations, do our homework, and eschew the cheap and easy way when we put them to use.