Again, I'm very familiar with this particular poll. It's been ongoing for decades. It's legit.
Naaaaaah, not if Gallup and all the rest are now wrong, wrong, wrong, wrong, and wow, were they ever. Disastrously wrong should be evidence of a problem, but a lot of people can't face that polls are no good anymore, they so badly want future prediction by some apparent "authority."
You can't declare a blanket "wrong" without evidence to the contrary. That's just naysaying. It's also blanket generalization. Present your evidence that "all" polls are wrong wrong wrong, and we'll talk turkey.
After Election Night, very shocked (I voted for Trump but had been doing speadsheets all summer on the totally false basis, it turned out, of the polling data at RealClearPolitics, so I never for a moment supposed he would win!), shocked because I had BELIEVED the polls, more fool me, I vowed never to read another poll. If I see the word "poll" as a supposed fact point in an article, I stop reading immediately, because I know it's all lies and propaganda.
False comparison I'm afraid. Election polls are by definition acutely dynamic. What respondents plan to do today changes, often radically, from what they planned to do yesterday at the same time. The instant case is about the respondent's own GENERAL tilt/identification, which evolves at more of a glacial pace. So what you have here is apples and oranges. A poll on something as continuously changing as an election has an exponentially shorter shelf life. All polls, legitimate ones, may be created equal but their
subjects absolutely are not.
As regards your example above, again a national poll can indicate what the whole thinks, but "the whole" is not how we elect Presidents. Four years prior "the whole" indicated that Clinton would get the most votes, and she did.
Polling is exactly like the weathermen: people want to know the future so badly, they'll pay for phony claims. I'm also training myself to disregard the future as a subject of interest. More difficult than polls, but I'm getting there.
Again,
this poll isn't at all about "the future". It's about "the present". There is nothing in this poll that purports to predict, or imply, anything in the future. This is an "IS" poll, not a "WILL BE" poll --- as an election poll is.
Analogically --- we can crunch stats (or fan opinions) and
predict that the Dodgers will crush the Pirates in tomorrow's game; that's a very different thing from counting "how many self-identified Dodgers and Pirates fans there are" in the present.
I've actually been trained in that science so I know whereof I speak. It's a diligent process that sweats over literally every word so as to avoid any suggestion one way or another. Phrasings are rotated so that if there's a multiple choice, the first choice constantly changes so that that's not an implied suggestion, and so on. Value judgment adjectives are verboten. Entirely neutral.
That's not to say it can't be abused and deliberately slanted by those so inclined of course, but again that's why each poll, valid ones anyway, show you their methodology, so we can see for ourselves what the approach was. If there's any deliberate slanting going on it will show up there.
Yeah, I've taken statistics courses, too. Two. I'm perhaps a little more mathy than you think I am. And I say again: polls being disastrously wrong, again and again, after major societal changes like no Republicans EVER taking any polls, doesn't exactly inactivate our long-ago statistics learning, but it does relegate it to the status of stats on angels, dancing, pins. No longer useful, because OBE: overtaken by events.
I wouldn't know how mathy you are and given my own dearth of mathiness wouldn't challenge anybody, but this is not a math thing, it's a
word thing. Phrasing. Adjectives. That's right up my alley. The math is simple addition, not complex. But the
phrasing... aye, there's the rub. You can easily massage the math with the words, and what I'm saying is that legitimate polls bend over backward to avoid doing that.
Trust me, if the science of polling were about math it would put me right to sleep. It's all in the words.
"Push polls", the fake ones designed to push some agenda disguised as a poll, will usually NOT show their methodology, because their shenanigans would be thusly exposed.
Example: when South Carolina voters got robocalled asking, "would you be less likely to vote for John McCain if you knew he had fathered an illegitimate black child?", that's a fake poll. Its object is not to ascertain a yes or no, it's to plant the idea "John McCain fathered an illegitimate black child", which is a sleazy way of telling a lie while hiding behind the technicality that it was a question, not a declarative statement.
Yes!! I'm delighted that you remember all that, too! I well remember all the fuss over that McCain thing. I followed it for awhile and it has some fascinating dimensions, but George Bush the Younger's campaign certainly did screw up McCain's attempt to get the nomination with that push poll.
And it was a push poll around then that pushed me right out of poll-taking, forever. I'm in Maryland, so it wasn't that one, but I thought poll-taking was like a patriotic duty --- remember when people thought stupid things like that? And that I "had" to answer the phone. I haven't had the ringer on for years; the government let telephones just become a conduit for thieves with weird accents from everywhere foreign.
So this push poll started off pretty normally with routine voting questions, and then he questioned whether I "knew" something a little blue about the candidate they were trying to destroy -- and then a little bluer, and third time's a charm, the caller hit me with "Did I know" a huge obscenity, really bad!! A man saying it, too, to a woman! I can't remember what was the obscenity (or I'd probably tell you: 30 years later, we can say anything online, I suppose) or who the smeared candidate was. I just know I hung up like a wasp had stung me and never took another political poll, and never intend to.
What you've described here is another example of what I described. As I said setting it up, it's possible, and it happens, that those with agendas will push that agenda with what looks like a poll, BUT ISN'T. Polls can be faked, just as news can be faked, photos can be faked, even a singer's ability to stay on key can be faked. Dressed up to look legitimate but they're not. That's ---again --- why I led off with the point that a legitimate one publishes its
methodology. If the South Cackalackee BushPush poll had published its methodology, its blatant abuse would be screaming its head off.
But that in no way means that a
legitimate poll is illegitimate just because somebody else did a push poll. The fact that you can photoshop Robert Byrd in a Klan robe doesn't make
all photography illegitimate. Nor does it mean that because some wackadoo ran a fake poll, the rest of us should stop participating in legitimate ones. And not to be unkind but your declining to participate doesn't matter to the poll because it will just field someone else with the same profile to reach the base it needs.
I'm hardly alone on the right about this and I don't think pollsters can get past this and re-establish trust in polls. Besides, polls don't deserve trust, they are plainly leftwing slanted, so they'll be wrong, wrong, wrong for the foreseeable future. I don't read polls anymore so I don't care. I don't watch the weather reports either; I'm not into wasting my time with fake future-telling.
I've already explained how polls (when I say "polls" I mean legitimate ones unless otherwise qualified) go to minute detail to specifically
avoid any slant. You're again comparing real polls with fake ones.
To wade further into this muck, polls are not by any means always political. Industries for instance use polls to determine what consumers think of their product versus what they think about their competition. It would serve them no use at all to slant a poll into what the company wanted to hear; that would be a complete waste of time and money.
