Election Day 2006 is almost upon us. Based on the polls, Democrats seem to be popping champagne believing they will be taking control of the House of Representatives and possibly the Senate.
Conversely, Republicans claim the Democrats are sipping their bubbly prematurely and will have to drown their sorrows after Nov. 7. And so the spin goes on and on as pundits from both sides of the political aisle analyze their crystal balls and the polls in hopes of predicting the future.
Over the past months, pollsters have been working hard to gauge the direction of this year’s election. It’s to be expected that political campaigns hire pollsters to measure the impact of a candidate’s message, image and progress with voters.
But news organizations also contract for polling and use the data to literally manufacture news. All in the hopes of being the first media outlet to provide answers about each race or issue. The problem is that although these answers may apply the day the poll is taken, they can change a multitude of times before the election actually takes place.
This campaign season must be a banner year for the polling business. For months we’ve been hearing the projections about whether or not the Democrats will take control of the House of Representatives and the Senate, each poll breaking down specific political races so commentators can reconfigure the “what if” scenarios of wins and losses.
Obviously we won’t know the real results until Election Day on Nov. 7, but that hasn’t stopped news organizations from creating buzz about this topic for the last five or so months. By hiring pollsters to research various races, the news media is able to create a story, and “break” the news about the results. Then this story can be used and analyzed for at least the next 24-hour news cycle.
Finally, that information can be recycled in the proceeding months when additional polls are done on the same topics and the trends are analyzed. All of this is done in the name of providing the public with information.
But is polling done only in the name of providing information? Or is it used by spindoctors to sway public opinion? According to “Politicians Don’t Pander: Political Manipulation and the Loss of Democratic Responsiveness” by Lawrence R. Jacobs and Robert Y. Shapiro, “The conventional wisdom that politicians habitually respond to public opinion when making major policy decisions is wrong.”
Instead, Jacobs and Shapiro claim that, “politicians pursue a strategy of ‘crafted talk’ to change public opinion in order to offset the potential costs of not following the preferences of average voters. Politicians track public opinion not to make policy but rather to determine how to craft their public presentations and win public support for the policies they and their supporters favor.” This leads to the old adage of, “buyer beware.”
How reliable is polling? A poll is only as good as the questions and the word choice used in those questions, which is why it’s imperative to analyze the questions when reviewing a poll. Polling is based on probability and statistics. In most cases, phone numbers are chosen at random for a sampling of the polling area.
According to John Zogby of Zogby International, which has been tracking public opinion since 1984, the sampling size determines the margin of error (MOE). A sample of 400 will have MOE of +/- 5%, 600 is +/- 4% and 1,000 reduces the MOE to +/- 3%. The margin of error is the percentage range, plus or minus, that the results will vary if the same poll is repeated.
In an election scenario, however, it’s difficult for pollsters to predict voter turnout, which is critical to the accuracy of the results. And as Michael Barone reports on Real Clear Politics, “Americans have fewer landline phones than they used to, and the random digit dialing most pollsters use does not include cell-phone numbers.” Also, a growing percentage of call recipients are refusing to be interviewed.
The exit polling for the 2004 presidential election proved that these results can be far from accurate. According to the National Election Archive Project, President Bush officially won the 2004 presidential election by 2.5 percent, but the exit polls showed John Kerry winning the race by 3 percent. This is a dramatic discrepancy.
But after all the money is spent on polls, and all the analysis and pontification has evaporated like the handful of air that it is, only one thing remains sure in this and all elections. It’s not the polling numbers or the voices of the pundits that matter, it’s your voice that counts. And only one poll matters, the one that’s tallied at the end of Election Day. So, go to the polls this Nov. 7, vote and have your voice be heard.
Kathleen Antrim is a regular contributor to The San Francisco and Washington Examiners, author of “Capital Offense,” and frequently writes for NewsMax Magazine.
