It's mentioned in the last graph of the article. The crux of this article is rather than 'shy trump voters' there is simply a large amount of these Trump voters who will refuse to take a survey flat out.
If 10% of people surveyed, at random, lie to the pollsters and choose the "opposite" candidate out of spite, you'll actually see a sort of reversion to the mean compared to the actual opinions.
I would bet there is a significant correlation with voting habits -- particularly in this election cycle. Why wouldn't someone who holds the view that polls are corrupt and biased have their attitude towards answering them affected by that?
If your opinion was that polls were biased against your candidate, how would that make you answer? Is it more likely to make you answer in favor of your candidate, answer in favor of the other candidate, or go out of your way to participate in polls to "fight back" against the bias?
You can invent a narrative that supports pretty much any activity. Only some of these narratives have any data behind them.
As far as I'm aware the best-supported theory by data isn't that people intentionally lie, but that they simply don't participate in polls because they narrative pushed by republican campaigns is that the polls are always wrong. This is, we'll say, half true.
What I'm interested in, now that the republican talking point is that the actual polls don't matter, since Democrats are stealing the election anyway, is if Republican turnout will similarly decrease.
The other was personal anecdata explaining that it doesn't have to be a political agenda driving the bad data.
A disproportional amount of people (after controlling for race/gender/income/education/etc) simply refusing to answer the survey has similar but conceptually different effect; one that (as the OP claims - perhaps they're mistaken) is new and had not been a significant factor before.
If half of them weren't push polls trying to disguise themselves as unbiased, I might be more likely to answer.
As it is, though? No. Go away. All of you go away. You'll find out my opinion when I vote. Until then, leave me alone.
That is different than someone who just doesn't want to take a poll but may be perfectly happy to tell you their views if you struck up a conversation. My own observation is that over the last 4 years, Trump supporters have become ever more loud and proud of that support, not more shy.
It's just become a lot more salient now, as the polls have been pretty crappy lately.
My job is basically data analyst/programmer, but I have a decent background in research of this sort (not politics, surveys and qualitative research) and occasionally have to conduct surveys. It often falls to me, or I insist on it, because practically no one knows how to write quality survey questions that aren't leading questions that bias the response.
For example, if you wanted an opinion on a new food product, someone that doesn't know how to ask a neutral question (or is purposely rigging the survey) could ask "How much do you like these french fries? 1) Fantastic 2) Very Good 3) Good 4) Okay 5) Awful. Those answers are front stacked with positive answers, only one negative answer, and no neutral answer. If someone doesn't believe they're awful, they have no choice but to answer "okay".
In any case, one common method I use to measure the presence of responder selection bias is a followup communication to non-responders 1-2 weeks later to nudge them to take the survey. The hypothesis is that people who did not response initially, but do respond when prodded again, will represent a middle point between initial responders & complete non-responders. If that middle point is not statistically significant in their responses that the original responders, I can infer that non-responders are less likely to be a group that would answer differently if had bothered to answer.
No, this is absolutely not a perfect method. More robust methods are needed to further reduce uncertainty: Ethnographic & other qualitative studies can help. Talk to responders to find out what they responded and consider the inverse. e.g., if they say "this is a really important election we have to get Trump out and I want to spread the word so it was important to share my opinion" that should make you question whether Trump supporters would feel the same way about responding.
To really get to non-responders, you need more time & more resources. Find them and pay them to talk to you. It's not uncommon to reward research participants who certainly wouldn't participate otherwise.
Again, none of these are perfect, they just inch you close to greater understanding a little bit at a time.
"... voters who will refuse to take a survey flat out" is equivalent to "... simply tak[ing] a neutral, non-committal stance on almost every hot-button issue these days."
I'd lie to pollsters because I dislike the practice. It causes irrelevant news, and feeds into strategic voting. One distracts / causes pageviews, other one lowers the integrity of our democracy. My data can skew the accuracy of pollsters, however little that is, thereby hurting their interests.
I am also open to lying to telemarketers as I despise their job. Perfect example of a Bullshit Job.