Political polling games

Even establishment sources are poo-pooing the reliability of political polls these days.   I’ve talked with a couple of political pros here in North Carolina — one from each side of the aisle — who tell me they think of the current state of political polling as an expensive way to obtain some highly questionable data about your race.  \

Following the national conventions, the drivebys are touting some polls featuring some questionable methodology that, surprise, boosts their heroine: 

”[P]olitical strategist Pat Caddell outlined his charge that Reuters tampered with its own daily tracking poll to manufacture a sudden surge for Hillary Clinton.

“They not only changed their formula, to put Hillary ahead. They went back and changed the results, for a week of results where Trump was ahead, and then they turned those into Hillary leads,” said Caddell. “They also erased all the former polling off the site. They didn’t tweak their procedure – they cooked it.”

“Never in my life have I seen a news organization, and a supposedly reputable poll, do something so dishonest,” Caddell continued. “What they have done is, they decided the people who said, ‘oh, I’m never for someone’ – oh, those must be Hillary votes. They used to be Trump voters.”[…] Hillary-Clinton-hereicome

For those of you who don’t know, Caddell is a long-time Democrat pollster.  He made his name as Jimmy Carter’s numbers guy.  So, he’s not a right-wing mouthpiece. He’s a guy that calls IT when he sees IT.  MORE: 

[…] “They made a switch, as much as nine points, in their results from the beginning of last week, the 25th and 26th. It is, beyond doubt, the most outrageous thing,” he declared, noting that results in three- and four-way polls that include independent candidates Gary Johnson and Jill Stein were also skewed.

“This is what the media is willing to do, to try to elect her,” Caddell said. “This poll is nothing but a part of a media offensive. In the 45 years since I was a child, in top-level presidential campaigns, I have never seen the media on such a jihad, and so involved in hiding facts, and not following up. This is a crisis of democracy, what the press is now doing.”

[…]

The Reuters poll he is blasting showed Hillary with a six point lead following the Democrat National Convention.  Now, we have this ridiculous McClatchy-marist poll sholiberal-media-biaswing a FIFTEEN POINT lead for Hillary.  In a two-way matchup, Hillary led 48-33.  In a four-way matchup featuring Trump, Clinton, Libertarian Gary Johnson the Green Party’s Jill Stein, you had a 45-31-10-6 split. Get a load of the methodology behind the McClatchy poll: 

[…] This survey of 1,132 adults was conducted August 1, sponsored and funded in partnership with McClatchy News Service. Adults 18 years of age and older residing in the contiguous United States were contacted on landline or mobile numbers and interviewed in English by telephone using live interviewers. Landline telephone numbers were randomly selected based upon a list of telephone exchanges from throughout the nation from ASDE Survey Sampler, Inc. The exchanges were selected to ensure that each region was represented in proportion to its population. Respondents in the household were randomly selected by first asking for the youngest male. This landline sample was combined with respondents reached through random dialing of cell phone numbers from Survey Sampling International. After the interviews were completed, the two samples were combined and balanced to reflect the 2013 American Community Survey 1-year estimates for age, gender, income, race, and region. Results are statistically significant within 2.9 percentage points. There are 983 registered voters. The results for this subset are statistically significant within 3.1 percentage points. The error margin was not adjusted for sample weights and increases for cross-tabulations.

First, are they seriously polling a sample of 1,132 adults that includes only 983 registered voters ???  What value is provided to readers by polling people about an election who are not even registered to VOTE? 

binker1Second, everyone knows a survey of LIKELY voters is much more reliable than a  survey of registered voters.  Registered voters are on paper, but it’s a real question as to whether they will show up on election day. Likely voters have shown up on the last two or three election days, and are likely to show up for this coming one. 

Third, all of that talk about how the results were “combined and balanced”?  That’s a fancy way of saying the results were cooked to show a desired result just like the Reuters poll. 

Electoral votes — awarded on a state by state basis — are more important in picking the president.  Polls of individual states would be more meaningful. George Bush lost the popular vote in 2000, but won with electoral votes.

I remember my first foray into big-time politics — the 1988 presidential race.  Following the Democrat convention, the drivebys were peddling polls that showed Mike Dukakis with a 17-point lead over George Bush.  Everybody remember President Dukakis? 

Don’t let stunts like this drag down your morale.  THAT is what the lefties and the drivebys want.