This is a great graph! Very encouraging! But it's not going to fly with the apparently nonmathematical people who analyze polls for, say, the @nytimes.com
I haven’t had time to digest all of the coverage yet, but I can make some general statements about the times polling methodology. Short, I think they do phone only polls as best as they possibly can be done.
I personally know, and have a great deal of respect for the team running the surveys. And their track record is born out not just in the polling performance, but also in their gold standard level of methodological disclosure.
Yes some of my former colleagues work there and they are attuned with development of survey methods advances. But we also know, long line of research shows people have partisan lens and if you have red glass, that is all you see.
You'd need pages of nested charts to convey layers of complexity of polling discourse bc polls themselves are several things & discourses proliferate w/ those. Among other things, polls are an infotainment industry. They've also become a Veblen Good for consumers who invest huge amounts of their ...
personal capital in fashioning an identity as savvy masters of this arcane and (largely) unproductive shiny object. The polls themselves vary vastly in quality, w/ many cheap knockoffs & increasing numbers of partisan fakes to take advantage of consumer demand. The many vectors involved are hard ...
to isolate much less measure. It's comparable, say, to trying to chart the many strange qualities of vaccine discourse. There too faux-expertise (about a highly valuable thing) has become a prize possession of people who dominate some discourses in order to whip up fear in themselves & others.
The discourses about polls and vaccines are parallel in several ways, fundamentally because for most of those involved the goal is to persuade others to invert perception of the actual utility of the shiny thing. Vax yakkers aim to nullify the extremely high actual value of vaccines. Poll yakkers...
by contrast try to inflate the importance of evanescent, inexact, narrowly framed estimates of what a small sliver of the public wishes to say about their thinking - as tools to impose their own wishes on society or to predict election outcomes a short time before the votes are counted.