A SurveyUSA poll this weekend has captured huge headlines: Republican Donald Trump now tops every Democratic contender.
While this makes for great headlines, looking at the actual study data and methodology tells a slightly different story, one that looks at two different sides of the equation.
SurveyUSA uses robopolling, a method that has become almost synonymous with Republican-leaning in the past several elections. In the 2012 election, SurveyUSA conducted 17 polls in the last 21 days of the campaign, with a 2.2 percent average error that skewed Republican. In fact, robopolling in general tended to have about a 2 percent skew toward Republicans in 2012.
Bearing this in mind, this poll is probably within its own margin of error, making any conclusions drawn from it worthless.
Demographically, the over-50 voters overwhelmingly supported Trump, while the under-50 voters moderately supported the Democratic candidates. And while the under-50 group was bigger, the biggest difference was that the under-50 demographic had a far larger share of undecided voters than the over-50 demographic.But there was a more significant problem in the poll, one of methodology that was a structural flaw in the study's design.
There has long been the debate of what happens when cell phones are/aren't included in national polling. At this point, almost all polling companies have had to accept a reality: the cell phone is most people's sole means of communication.
In polling, the Devil is always in the details, and the percentages can be skewed by any number of design tweaks that can significantly change the outcome of the entire poll.
In this poll, 62 percent of respondents were on landlines with voice questioning, while 38 percent were on electronic devices (mobile devices) with written questions.
Just the difference in delivery alone can skew the results, but in the end, two different tales were told.
In the "cellphone universe," as the pollsters put it, Clinton defeated Trump by 16 points, Sanders by 12 points, Biden by 17 points, and Gore by 17 points.
Of course the poll included the small caveat, "The more cell-phone respondents a pollster includes in its "mix" of voters, the more Democratic the poll results will be."
It seems as though this poll was purely constructed to achieve a specific outcome by manipulating the numbers of landlines/cellphones, even when there is pretty good demographic data on the percentages of cell phone users available (and it's way higher than 38 percent).
So while Donald Trump can bask in the few days of glory of being on top, reality will eventually sink in and he better hope that cellphone users don't vote in 2016.
Photo retrieved from RedState.com