In one of the more explosive Wikileaks Podesta emails released over the weekend, Democratic operatives appear to discussing methods to “over sample” certain population areas in order to provide more favorable polling results to their candidates.
This year, Donald Trump has brought up the politically taboo topics of a rigged polls, rigged politics and rigged elections, only to be promptly criticized for it by the media and even President Barack Obama himself. Over the past few days we have seen national polls show anywhere from Clinton +12, to Trump +2, and everything in between. Obviously with spreads that large throughout so many polls, at least some of the polling methodologies are likely to be skewed.
It seems Trump wasn’t too far off his mark in regards to “rigged polls”, as the latest Podesta emails from Wikileaks now show a discussion between Democratic operatives in 2008 regarding recommendations for oversampling of polls through one of their polling agencies.
The link to this email on Wikileaks can be found here.
The emails states, “I also want to get your Atlas folks to recommend over samples for our polling before we start in February. By market, regions, etc. I want to get this all compiled into one set of recommendations so we can maximize what we get out of our media polling.”
There is a even a handy attachment to this email that outlines precisely which oversampling methods to use in designated populations areas. For example, in Arizona the report suggest to:
– Over-sample Hispanics
– Use Spanish language interviewing. (Monolingual Spanish-speaking voters are among the lowest turnout Democratic targets)
– Over-sample the Native American population
When polling in Florida, the report suggests to “consistently monitor the sample to ensure it is not too old, and that it has enough African American and Hispanic voters to reflect the state.” Incidentally, those are exactly the voter groups that are more likely to vote Democrat in an election.
On a national scale, the report’s recommendations are more general, leaving it up to the polling agency to “ethnic over sample as needed“:
– General election benchmark, 800 sample, with potential over samples in key districts/regions
– Benchmark polling in targeted races, with ethnic over samples as needed
– Targeting tracking polls in key races, with ethnic over samples as needed
In other words, no matter what geographic location or state you are polling in, find an ethnic minority that favors Democrats and make sure to over sample them to provide higher polling numbers for your candidate than are realistic.
For a current example of exactly how this type of oversampling can create the bias polling results that a political party is after, just look at the recent ABC News national presidential poll showing Hillary Clinton in a commanding 12 point lead, yet over samples Democrats by 9%. That allows ABC to then issue such headlines as, “Clinton Vaults to Double Digit Lead” in an attempt to discourage Trump supporters.
So far, between the Wikileaks emails and James O’Keefe’s undercover videos, the Democratic party has been shown to engage in rigging politics (DNC collusion with media), rigging elections (DNC hiring organizations to send agitators to Trump rallies) and now rigging polls. Donald Trump doesn’t seem too far off his mark when he talks about a “rigged election” anymore.
President Barack Obama told Trump last week to “stop whining” in regards to the possibility of a rigged election. Well, it’s easy to make such statements when your party is the one that benefits from such underhanded election tactics, isn’t it?