If You Torture Them Sufficiently, Polls Will Confess to Anything
The same has been said of statistics, but one poll finds the majority of Greenland respondents support joining the United States.
This, in a headline blazoned across an article from The Hill: 1 poll finds majority of Greenland respondents support joining US.
Disclaimer, I must admit I am a huge opponent of polls, mostly the political type, but under all circumstances as well. Not only do I not trust the methodology, but strongly object to small groups forecasting what large groups will do. If landline phones are dying out, how do political polls work today? Taking the objection a step further, I feel that ‘wanting to be on the winning side’ is a huge American weakness, and that steers decisions in ways they ought not to be steered.
So, distrustful of my own ignorance on such a widely popular process, I Googled ‘do polls require land lines to register opinions?’ The answer was abstract, to give it the most charitable description:
Polling practices have changed significantly over the past two decades, especially with the decline of landline usage. A staggering 183 million Americans (7 in 10 U.S. adults) no longer use landlines and rely solely on their cellphones.
This shift has prompted pollsters to find more ways to reach voters.
For reputable organizations like Gallup, Pew and Emerson, the process is randomized. Teams randomly select registered voters from lists compiled by secretaries of state, then screen for those who are likely to cast a ballot. Most organizations have moved to mixed mode. That means a combination of phones and text-to-web, and maybe an online panel thrown in. “Once we’ve collected all the data, we go through a process of weighting the data to ensure that the demographic breakouts look like what we expect them to be on Election Day.”
So, if I have this correctly, random selections are screened, then weighted to ensure the demographics look like what we expect on election day.
Screened, weighted, and ensured certainly sounds like something I’d put my trust in.
“We’re setting expectations, the general electorate, people who are interested, can see how competitive the race is. They can have some idea of what to expect on Election Day.”
Three things wrong with that, if not more, once they are chewed over: First, a poll should never ‘set expectations,’ in what is supposed to be a well thought through, and personal exercise in democratic representation. Second, if the race is shown to be a walk in, many who might have voted, may decide to stay home, affecting the ultimate outcome. And finally, desire to be on ‘the winning side’ may actually change the vote of the weak minded. We seem to have more and more of them these days.
Returning to the subject of Greenland
Who gives a damn what Patriot Polling finds, having established confirmed results that 57.3 percent of respondents approve of Greenland becoming part of the U.S. and a mere 37.4 percent disapproved of the potential acquisition, while 5.3 percent are undecided about the move.
Greenland is an autonomous territory of the Kingdom of Denmark. The opinion of 416 random nitwits, out of a population of six million is absurd.
Patriot Polling, the instigator of all this nonsense, only receives a 1 star rating out of three.
Nonsense building upon nonsense, the Trump nonsense of acquiring Greenland has now been broadcast to the half of America who believes his random rantings.