If you follow the Presidential campaign, you will hear a lot said about who has the lead at any one time. 

A growing number of mainly conservative voices, however, are making the claim that the poll results being widely publicized are misleading. 

Are they right?

Well, let’s lay out some basic facts about Presidential polling.

To get an accurate reading of how a large group of citizens feel about a particular subject, whether it is a television program, potato chip or candidate for President of the United States, a pollster must contact a sufficiently large number of people. If the number is too small, of course, the likelihood of error increases. 

The biggest flaw in Presidential polling cited by conservative critics is not that pollsters contact too few people, but rather that many polls simply contact “registered voters” and do not measure or take into account whether a particular voter is likely to vote in the particular election.  

History teaches us that slightly more than half of all registered voters typically turn out in Presidential elections. (More than 56 percent voted in 2008, the highest number in forty years, but there is no reason to think it will increase.)  That means that when “registered voters” are contacted and asked their preferences, nearly half of those respondents will never go bother to vote come Election Day. 

This business about whether a particular voter is likely to vote as opposed to simply being “registered” to vote is critical, some commentators say, when trying to predict results in a Presidential election.  

One pollster, Rasmussen, claims that its team measures the preferences of “likely” as opposed to simply “registered” voters. The company does this by asking questions and gathering information about whether a particular person has voted in a recent election, or plans to vote in this particular election. 

If we assume that Rasmussen’s methods are sound and its assumptions are correct, the Presidential race may be much tighter than it appears in some national polls.

In the latest Rasmussen tracking poll reaching so-called “likely voters,” published just today (Oct. 5, 2012), the President holds as two point lead. 

Many pundits claim that “likely voter” polls are more reliable. The argument goes that because conservatives and Republicans historically are more likely to vote, polls that reach “registered voters” tend to overstate the impact of liberals and Democrats.

There is another second step in the process of polling worth mentioning, and that is making sure that the group that is polled is representative of the population as a whole. This step is very tricky. For example, persons in some groups are more difficult to contact than others (the young, minorities), while persons in some other groups are easier to contact (seniors being the best example). A good pollster must strive to contact persons of different races, religions, economic circumstances and locales in order to make sure that the final sample group represents the US population as a whole. 

In Presidential polling, an emphasis is placed on contacting sufficient numbers of Democrats and Republicans. The pollster then tries to predict which percentage of voters in the upcoming election will be Democrats and which percentage of voters will be Republicans.

Critics claim that this step in the process also is being mishandled by the well-known pollsters. The critics claim that polls showing Obama with a sizeable lead are flawed for the additional reason that they are relying primarily on the 2008 results when predicting the relative turnout in 2012 of Democrats and Republicans. The critics maintain that in 2012, it is more likely than not that the ratio of Democrats and Republicans that turn out to vote will move in favor of Republicans as, historically, voters are usually more fired up to throw someone out than keep someone him. These critics also contend that the 2008 turnout was inflated by unusually high turnout among youth, many of whom voted for Obama. Some pollsters believe that the youth turnout will return closer to normal levels in 2012.

Why does any of this matter? After all, the voters will decide the matter come Election Day. If one candidate leads in the polls, he may still go down to defeat. The beauty of the election is that the voters get to decide.

The reason this matters is three-fold. 

One, as Election Day approaches, fundraising efforts intensify. It is easier for a candidate with a lead in the polls to raise money. 

Two, people love to bet on a winner, so if they are undecided, or wavering, it is possible that a significant portion of them will go with the person they believe will win, versus the person they want to win. 

Finally, there are those that believe that turnout can be suppressed if voters believe that voting is “hopeless.” If their candidate is “behind in the polls,” the thinking goes, some of the candidate’s supporters will not show.

If these critics are right, polls showing Romney trailing Obama could – all by themselves – be damaging to the Republican challenger’s campaign.

I am not sure being shown in the lead is a great thing for the President either.  If Obama supporters believe that he is a lock to win, it is possible that some of them will get complacent and not vote. As we saw in 2000, a few votes here and there can decide an election.

So here is some free advice to you, the interested voter. Follow the polls, but don’t be afraid to dig a little deeper. What is the sample size? Were “registered” or “likely” voters sampled? What assumptions underlie the polls as far as voter turnout? Be the smart one at the next dinner table discussion and raise all these issues for consideration.

On the other hand, if all of this starts to make you crazy, just wait for Election Day.