This is 43% (+/- 7%) an old statistician bellyaching about pollsters - so best to just skip this for the feckless rant it is.
This morning’s New York Times (yes yes, i know) has an opinion from “Dr Neil Gross” professor of sociology at Colby College (who is writing a book. hey you gotta earn a living). The headline is: “This is Why Young People Really Voted for Trump”. First and foremost: that is never answered in the piece. There’s a bit of “Well they said it was because of the economy, so likely it was that”, yet go back and squint at that “Really” in the headline. There is no @#$! “really” at all in the piece.
On to the second part of the rant. This was based on a survey done now. After it’s bloody apparent to all but the most deluded fox"news" “Innie” that trump is a disaster for the economy, for global politics, for damn near everything but a few billionaires. So you conduct a survey which amounts to: “You’re a young person, you voted for trump, why did you do that?” And not once, not in a single bloody sentence, from a doctor of sociology is it posited that maybe their motivation for voting for trump (rather than Harris) might be other than they are ‘publicly’ or even what they’re admitting to their outward facing selves, might not be truthful?? That is: when they say it was the economy then that’s it - that’s why they voted for trump. Damn it, doctor of sociology, folks are aware - even as they act on it - that misogyny, racism, and other shorted-sighted selfishness, is not something to be overtly proud of.
The ‘conclusion’ of this NYT piece is that there was all-in-all no shift toward trump; despite there being some small but clear shift for trump. @#$!
I guess we are about to find out if USSC will allow the emasculation of the judicial system or not. It’s only been 2 months, and we are already here. Come on ACB!
I no longer trust any polling. It’s just really broken right now. First of all, almost all of it is biased. Whoever paid for the poll, or is behind the polling company, is directing the questions, and the questions are almost always biased. Not only that, but when you read news articles about these polls, they almost never tell you the exact questions asked, and they really should be required to. The exact vocabulary and grammar used can influence how people respond.
Second, while I realize that statistical science says if you randomly select x number of people, that can accurately represent the views of y number of people, I’m really not sure I believe it. People are not inanimate objects or phenomena that can be objectively measured. People lie. People even lie to themselves. And they have all kinds of reasons for lying, some innocent, some not so innocent. I will just never believe that a survey of 2500 citizens can accurately reflect how 150 million people are going to vote. That math just doesn’t math for me.
And I went up there, I said, “Shrink, I want to kill. I mean, I wanna, I
wanna kill. Kill. I wanna, I wanna see, I wanna see blood and gore and
guts and veins in my teeth. Eat dead burnt bodies. I mean kill, Kill,
KILL, KILL.” And I started jumpin up and down yelling, "KILL, KILL, " and
he started jumpin up and down with me and we was both jumping up and down yelling, “KILL, KILL.” And the sargent came over, pinned a medal on me, sent me down the hall, said, “You’re our boy.”
Feb 14 (Reuters) - Days after U.S. President Donald Trump ended duty-free entry for cheap Chinese goods entering the U.S., his administration put the order on hold after more than a million packages piled up at New York’s John F. Kennedy International Airport.
It was the result of a rushed, confusing policy change that proved unworkable on short notice. Government officials are now scrambling to implement the order in a way that won’t cripple America’s hyper-efficient import system.
Before cell phones, I had more trust in polling. The people who designed them were more professional so the wording generally was not an issue. People almost always answered their phones, so they could get a representative selection of America. Pollsters factored in a certain margin of error due to potential lying by respondents.*
Now we have “pollsters” who design biased questions for specific politicians, issues, or lobbyists. We’re mostly a cell phone-based country, and there is no way that the people who answer their phones are representative of the diversity of millions Americans, even if it takes them 10,000 attempts to get 2,500 (or more likely 1,000) respondents.
*I remember a poll done in the mid-ish 80s (it was after my class in polling) and I had a question about one of the questions. The poll was so neutral that I couldn’t even guess who was behind it. I asked them to clarify the question (“Do you mean…?”) and they wouldn’t. That was one well-administered poll, even if one of the 25+ questions needed to be fixed (which they cannot do mid-polling).
Dr Joanne Liu, pediatrician and former International President of Médecins Sans Frontières/Doctors Without Borders (MSF) had Her conference at the Universityof New York canceled, as the institution fears Her talk could be considered antisemitic and offensive to Mr Trumps’ ego
I highly doubt polls can even really claim “randomly selected” in most cases. I’m not exactly the first to observe this, but I think most sample populations would better be described as “people with landlines of some sort, at home when we call, willing to talk to some random bozo for 5 to 10 minutes, and restricted to giving replies on a Likert scale for questions about concept that aren’t that easy to simplify.”
Well they do call cell phones, too. But the issue remains…who answers calls from numbers they don’t recognize? Very few from Gen X and younger, I’d bet. And the ones who do may not be representative of the whole.
I immediately thought of the relatively high number of poll calls we’ve gotten on our home number (a VOIP phone that I’ve had the same number on for almost 20 years), but I can’t recall getting on on my mobile, but realized you already explained that: I almost never answer unknown numbers on that phone!
TL;DR - yeah, they are very self-selecting and are in no way actually random.
