As we move into the final seven weeks of the campaign, poll after poll are being released from every corner of the country. Some polls are local. Some are national. Most every poll comes to a different conclusion by attaining different results. One day in one poll, Romney and Obama are tied. In a separate poll on the same exact day, Obama is up eight points. How exactly is this possible? The key to understanding political polls is much more in-depth than the majority of people are willing go. That’s why, in today’s post, I’ll do the hard work for you.
First off, it is a must to recognize this fact: news agencies and professional polling companies WILL twist the numbers to get their ideal result. Just as the majority of the media has shilled for the Obama campaign, spanning back to the 2008 Presidential campaign (cue Chris Matthews – http://www.youtube.com/watch?v=no9fpKVXxCc), polls are now being used to set the narrative that Obama has a guaranteed win in November. And why, you might ask, would so many polls push this? The answer is quite clear: voter suppression. Democrats would like nothing more than for Republicans to get discouraged and just stay home on voting day – much like they did in 2008 when there was no thrill running up anyone’s leg for John McCain.
2012 is different, and the Dems know it.
Just as we did with the tax discussion, let’s talk about the barebones of political polling. First, there is the classification of the entire group being polled. Normally at this time in the election cycle, polls will choose either registered voters or likely voters as participants. This is an essential detail to take note of. Registered voters are not always likely to vote. The group includes some individuals who have registered to vote, but who may not like either candidate, may not be able to vote or just don’t vote period. Polling from a registered voters group, as examined by the Pew Research Center, normally favors Democrats by a 4% average, meaning the poll is skewed by that many percentage points toward the Democrat in a poll. Likely voters, on the other hand, are the people who actually plan to go out and vote for their candidate of choice. This group is considered the most accurate when executing polls. As an illustration of the difference between polling registered voters and likely voters, look at the following:
A poll of registered voters is taken, with a result of Obama 48% – Romney 44%. If that poll had instead been taken with a group of likely voters, the results would be a statistical tie: Obama 46% – Romney 46%.
This is just the first step in understanding whether a poll is accurate or not (Don’t run away yet, I promise this is the juicy stuff). To locate the real data behind the result, you must dig a little deeper. Each poll also samples select percentages of each ideological group: Republicans, Democrats and Independents. Here’s where things get a little tricky. Normally, the poll should reflect, at least somewhat, the expected turnout for the coming election. In 2008, the Democrats had much more enthusiasm for their candidate and the expectation was that a higher percentage of them would vote than Republicans. This was accurate, as 7% more Democrats turned out for the election. However, in 2010, the tide turned as the Tea Party and so many Americans, disturbed by the Obamacare albatross, flooded their polling places and gave the Republicans a slight edge over the Democrat turnout. In the 2012 election, Republicans are again expected to have the advantage in turnout, as disengaged, regretful Obama voters stay home (Gallup polling has shown voter enthusiasm on the Republican side by a 2 to 1 margin). So why aren’t the polls reflecting this?
Take a look at this poll: http://www.nytimes.com/interactive/2012/08/23/us/politics/23poll-docs.html. This is a poll that focuses on three key battleground states: Florida, Ohio and Wisconsin. Don’t be fooled by the large PDF – we only need to look at three of the sixteen pages. Start on page 1: they are polling likely voters. Off to a good start. Slide down to page 2:
In Florida, Obama leads Romney 49% – 46%
In Ohio, Obama leads Romney 50% – 44%
In Wisconsin, Obama leads Romney 49% – 47%
Pretty disheartening, huh? Problem is, you’re being deceived. With your newly discovered ability to decipher phony polls, slide down to page 13. Conveniently tucked away in the last couple pages of the polling data, the question is posed, “Generally speaking, do you consider yourself a Republican, a Democrat, an Independent, or what?” Here are the totals:
Florida – Republican 28%, Democrat 34%, Independent 32%
Ohio – Republican 26%, Democrat 35%, Independent 34%
Wisconsin – Republican 28%, Democrat 32%, Independent 33%
The stats don’t lie. Obama leads in Florida because the polling gives him a +6 Democrat to Republican differential. Obama leads in Ohio because the polling gives him a +9 Democrat to Republican differential. Obama leads in Wisconsin because the polling gives him a +4 Democrat to Republican differential. And these pollsters really think they are fairly portraying the electorate and the state of the Presidential race? The answer is no – it’s on purpose.
Let’s just say an equal amount of Republicans and Democrats show up on Election Day 2012 (even though the Republicans are expected to have an edge). If the numbers were altered to reflect an even electorate, here’s what it would realistically look like:
In Florida, Romney leads Obama 49% – 46%
In Ohio, Romney leads Obama 49% – 46%
In Wisconsin, Romney leads Obama 49% – 47%
This is just a microcosm of what is occurring on a daily basis – whether in the news or through the constant release of poll after poll – spewing numbers and percentages without giving you any background information. To reiterate the point: THEY DON’T WANT YOU TO KNOW! Take this information and use it, if only to calm your fears when you hear the next poll declaring “Obama leads Romney”.
If you look closely enough and often enough, you’ll see the truth: