SoylentNews is people
SoylentNews
Cliff Zukin writes in the NY Times that those paying close attention to the 2016 election should exercise caution as they read the polls because election polling is in near crisis as statisticians say polls are becoming less reliable. According to Zukin, two trends are driving the increasing unreliability of election and other polling in the United States: the growth of cellphones and the decline in people willing to answer surveys. Coupled, they have made high-quality research much more expensive to do, so there is less of it. This has opened the door for less scientifically based, less well-tested techniques. To top it off, a perennial election polling problem, how to identify “likely voters,” has become even thornier. Today, a majority of people are difficult or impossible to reach on landline phones. One problem is that the 1991 Telephone Consumer Protection Act has been interpreted by the Federal Communications Commission to prohibit the calling of cellphones through automatic dialers, in which calls are passed to live interviewers only after a person picks up the phone. To complete a 1,000-person survey, it’s not unusual to have to dial more than 20,000 random numbers, most of which do not go to actual working telephone numbers.
The second unsettling trend are rapidly declining response rates, reaching levels once considered unimaginable. In the late 1970s, pollsters considered an 80 percent response rate acceptable but by 2014 the response rate has fallen to 8 percent. "Our old paradigm has broken down, and we haven’t figured out how to replace it," concludes Zukin. "In short, polls and pollsters are going to be less reliable. We may not even know when we’re off base. What this means for 2016 is anybody’s guess."
Original Submission
(Score: 3, Insightful) by tibman on Monday June 22 2015, @09:38PM
Not sure why people are hating on you. Land-line based phones would be a terrible way to get a cross-section of voter's intentions. It was probably useful ten years ago but not now.
SN won't survive on lurkers alone. Write comments.
(Score: 2) by aristarchus on Tuesday June 23 2015, @08:50AM
Not sure why people are hating on you.
Not hating, just pointing out an idiot when one raises its ugly head. If the so egregoiusly down-modded poster had the slighted understanding of sampling, statistics, inductive reasoning, and just plain SCIENCE, they would not have said such a stupid thing. And now you, you have stepped right in with your lack of understanding of what is being discussed. The poster being so properly chastised said nothing about landlines, that is actually something that is in the FA! And now you defend them by making it seem like the post may have been on topic? Fey, fey, my dear sir (or madam), these are dangerous times, when we must choose our terms carefully, lest they become fodder for such idiots as above.
Now about sampling (and I am quite sure that professional pollsters are well aware of their difficulties), everyone should be aware of the US presidential election of 1948, where the Chicago https://en.wikipedia.org/wiki/Dewey_Defeats_Truman [wikipedia.org] Daily Tribune announce that Dewey won. Remember President Dewey? So what happened? First ever telephone poll. In 1948, who owned telephones, a rather new fangled technology? That's right, wealthier people. And who do stupid rich people vote for? Right! Republicans! Even in 1948 (some things never change). Classic example of a prediction based on a biased sample. (This is also why, nowadays, when rightwing newspapers publish poll results, they are not trying to report on the election, they are trying to influence it, just like Karl Rove on the Fox Armaggeddon Channel, when he had the "real maths".)
So the only point being made in the Fine Article is 1) landlines are no longer a reliable way to sample public opinion, and 2) polling may no longer be a reliable way to sample public opinion. Now knowing both of these facts, I am sure that professional pollsters will have no problem producing product for us in the forseeable future. It will just be a question of whether it is pull polling, or push polling, or whether advertising is no longer a viable business model for the internet. It's like the guy in the dingy in the hold of the Exxon Valdez at the end of the not-so-great movie "Waterworld", when the flare is dropping down, and he says, "Oh, Thank god."
(Score: 0) by Anonymous Coward on Tuesday June 23 2015, @03:38PM
s/stupid/ignorant/
I'm not sure whether you're genuinely stupid or just ignorant on the difference between those words, though.
(Score: 0) by Anonymous Coward on Tuesday June 23 2015, @03:53PM
Not hating, just pointing out an idiot when one raises its ugly head. If the so egregoiusly down-modded poster had the slighted understanding of sampling, statistics, inductive reasoning, and just plain SCIENCE, they would not have said such a stupid thing.
So it's scientific to simply assume you're right that people's answers are correct. This has nothing to do with denying statistics, sampling, inductive reasoning, or science; it's simply denying bad science built on too may assumptions.
(Score: 2) by aristarchus on Tuesday June 23 2015, @05:15PM
So it's scientific to simply assume you're right that people's answers are correct.
My Gawd, the stupid just keeps on coming! Who said any such thing? I never did, honestly, would I lie to you?
This has nothing to do with denying statistics, sampling, inductive reasoning, or science; it's simply denying bad science built on too may assumptions.
OK, I see there are some cognitive liabilities involved here. The idiot frist poster said any sampling was wrong,
Asking a few thousand people a few questions
Or were we to take the veracity the be the main issue? If we asked more people more questions, would it counteract the mendacity of the source? Or do we have to reduce the dishonesty and then the numbers won't matter? It's all terribly confusing, involving sampling and statistics and correct inductive reasoning, and margins of error. And guess what! Do you not think that pollsters take honesty into account? It is not that hard to confirm your assumptions in election polling, you can just compare your prediction to the official results! If you are wrong, there is something wrong with your methods and/or sources. This is what the Fine Article is saying, they know this already.
To just say, "Grumble grumble Small Sample size! Grumble grumble, People could lie! So Grumble grumble, Polls suck!" sounds just like, exactly like, and nearly indistinguishable from, a climage-change denier. Begin Handwaving, . . . NOW!
(Score: 0) by Anonymous Coward on Tuesday June 23 2015, @05:42PM
The idiot frist poster said any sampling was wrong,
Nope. Notice the "and" in the sentence and you will understand your mistake.
Do you not think that pollsters take honesty into account?
The polls could be manipulating the election results, and the pollsters themselves could be dishonest.
Not only that, but the polls are a useless waste of time.
(Score: 2) by aristarchus on Tuesday June 23 2015, @07:05PM
The idiot frist poster said any sampling was wrong,
Nope. Notice the "and" in the sentence and you will understand your mistake.
Getting rather tiring, but nope on the nope. "And" is a logical conjunction, which means that both claims are asserted to be true. I was only asking for clarification, such that the contrary could be true if one of the conjuncts was not, as in "if the sample size were large enough, would polls be accurate despite respondent mendacity?". Evidently I will not be getting an answer, since
Not only that, but the polls are a useless waste of time.
Why are polls a waste of time? What are they for? Why do some people pay so much money to have them done, and who are these people? Evidently polls are reliable enough for them! So are we going with Karl Rove's "real math", or Nixon's "Silent Majority"? Everyone knows that every agrees with me, so we can ignore polls. Yeah, that's the ticket!
(Score: 0) by Anonymous Coward on Tuesday June 23 2015, @08:17PM
Getting rather tiring, but nope on the nope. "And" is a logical conjunction, which means that both claims are asserted to be true.
It's describing a specific scenario. Learn the way humans actually use English rather than acting like a robot.
Why are polls a waste of time? What are they for? Why do some people pay so much money to have them done, and who are these people? Evidently polls are reliable enough for them! So are we going with Karl Rove's "real math", or Nixon's "Silent Majority"? Everyone knows that every agrees with me, so we can ignore polls. Yeah, that's the ticket!
You're ranting about random nonsense.
Why exactly are polls not a waste of time? Is there some pressing need to know what the election results will likely be before the election that doesn't involve just manipulating voters into voting a certain way? Probably not, or at least not for me. If you want to pay attention to stupid popularity polls, have at it.