The latest poll from McClatchy Newspapers finds former Speaker of the House Newt Gingrich is the strongest Republican candidate when matched one-on-one against President Barack Obama.  The poll places him statistically neck and neck with the incumbent president, back just 2 percentage points among registered voters, with Obama 47% to Gingrich's 45%.


Looking at other candidates:

• former Massachusetts Governor Mitt Romney does next best, with Obama leading 48 % to 44 %
• Representative Ron Paul of Texas is the third best, with Obama at 49% to Paul's 41%
• Herman Cain is next, with Obama showing 49% to 39% over Cain
• Fifth best is Governor Rick Perry of Texas, coming in with Obama at 51% and Perry at40%
• And the last person tested was Representative Michele Bachmann of Minnesota, giving Obama 54% to her 35%.

Finding a strong candidate is critical for Republicans, as Obama continues to show general election strength despite 9 percent unemployment.  But he is extremely vulnerable.  In another poll, just 43 percent of registered voters approve of the way he's doing his job.  And on the economy, the top issue for most voters, 59 percent disapprove of Obama's performance, while only 36 percent approve.  Also, just 38 percent say they definitely plan to vote for Obama, while 48 percent say they definitely will vote against him.

McClatchy claims that one reason why Gingrich and Romney do better than other Republicans against Obama is that they both win independent voters, who swing back and forth. However, I have another theory, and it has to do with the nature of polls.  I believe that Gingrich and Romney simply have more name recognition.  Gingrich was Speaker of the House, and therefore in the news for many years.  Romney actually ran for President in the primaries four years ago.  Ron Paul, in third place, also ran for President.  The other three candidates are all new on the scene, and many potential voters are paying little attention to the race because it's so early.  However, when a poll confronts them with a list of names, they will gravitate to names they already know.

This leads me to a discussion about the methodology of polls and a caution to you to take anything a poll shows you with a grain of salt.  McClatchy states that this poll was a national poll of 1,026 adults.  Only 872 were registered voters.  Right there is a red flag, because only 87% of those polled even vote, so the other 13% are not really important to the question asked.  Also, we all know that on election day, less than 50% of registered voters bother to vote, which means a large number of people polled will not affect the real outcome anyway.

McClatchy also says adults 18 years of age and older residing in the continental United States were interviewed by telephone, and telephone numbers were selected to ensure that each region was represented in proportion to its population.  Why would you think regions are a valid predictor of a Presidential election, when states are given influence only through the Electoral College?  A valid poll for this purpose should be made state-by-state, with each state weighted the same as its Electoral Votes.

Finally, it's a telephone poll, and all such polls are highly flawed.  Some people will set their phones to dump unknown callers, some calls won't be answered because no one is home.  And just think about how many poll calls you get and how many you refuse to participate in!  Besides, people can and do lie in a poll, because there is no consequence.  And we also know that pollsters weight their sample for various factors, producing bias.  The most famous of which is that most polls are weighted higher in Democrats responding than Republicans.

McClatchy claims the survey has an error margin of plus or minus 3 percentage points, but if you factor in all these problems, it's really much larger.  In fact, here's what the numbers really tell us:  IF Presidential elections were a popular vote (instead of Electoral Votes), and IF people who don't vote were somehow counted, and IF there is no bias in McClatchy's sample, and IF only people who respond to telephone polls voted, THEN President Obama would win the election LAST WEEK, EXCEPT that the 3% margin of error places the actual result anywhere up to Gingrich = 48% to Obama 44%! Beware of poll results masquerading as real information.