Polls Are Often Wrong

Many observers of political campaigns spend an inordinate amount of time and effort "poll watching." This becomes increasingly tedious and misleading because there are sometimes wide differences between the results reported by various polling organizations. Some of this is due to statistical error, and others due to the structure and wording of the poll. I have come to believe, however, that one of the major factors often overlooked in reviewing these polls, is that individuals responding to polls often give either tentative, partial, uncertain responses, and in some cases, I believe they outright lie.

Historically, as election dates near, polling margins narrow. To a large degree, that is because a large number of individuals are still party loyal, even if they are dissatisfied with the performance of their political party. Often, initially, opposition candidates poll very well as an anti-incumbent vote. However, when voters get a chance to think about it more thoroughly, they often realize that they also do not agree with the opposition candidate, or that what he wishes to do, might be worse.

In reality, the only poll that really matters is the polling booth on Election Day. Amongst nations that have democratic elections, Americans tend to vote in far lower percentages of eligible voters, than in other nations. Many individuals previously polled, who are disgruntled with the incumbent, or disillusioned with the state of the nation or locality, do not have strong pro-opposition candidate feelings either. I believe this, to a somewhat large degree, often accounts for low voter turnouts on Election Day.

These advance polls have been shown, however, to often have an impact on voter turnout. A close advance poll generally results in a larger voter turnout, while a poll that indicates a "landslide," often reduces voter turnout. Many potential voters either feel their vote doesn't count, or that they really don't particularly like either candidate.

Another factor that often impacts turnout, especially on the west coast, is the results announced earlier from east coast races. Often, voters go with the trends, and if results from the earlier closed polls on the east coast seem to indicate a trend (or that the media interprets it that way), it often has a significant impact on later voter turnout on the west coast. In many past elections, many experts believe this behavior had a significant impact on election results.

It is important to vote, if only for the fact that if you vote, it gives you the "right to complain," while if you do not, you may have been "part of the problem." Don't pay that much attention to pre- election polling. If you believe that one candidate is better qualified, and will better represent the interests of his constituents, I urge you to go to the polls on Election Day, and vote!