Why do polls underestimate Trump support? Could be his base is disproportionately socially isolated

[essentialsaltes]

Why do the Trump columns ad up to more than 100 % and the Biden columns to less than 100 %?

Because you're adding up the wrong pieces. For 'social people' about 45% each voted for Biden and Trump (adding up to ~90). For 'isolated people' it's more like 25 Biden 60 Trump (adding up to ~85)

 

[RestoreTheJoy]

Well, I don't know about civic groups. The OP talks about social isolation, and the article refers to this study, which shows evidence that "social network isolation is negatively associated with unit response. [i.e. responding to surveys]". The study provides some data that confirms a widely held hypothesis about social isolation and willingness to take surveys.

Yes, this doesn't mean "joins organizations". Many people have a social support system - someone who cares - but do not or cannot, due to immobility - join groups.

 

[Landon Caeli]

I love it when we get to play the guessing game on why Trump voters are at fault for failed polls.

 

[Nithavela]

Because you're adding up the wrong pieces. For 'social people' about 45% each voted for Biden and Trump (adding up to ~90). For 'isolated people' it's more like 25 Biden 60 Trump (adding up to ~85)

Doesn't say anything about the percentage of Trumps base that is socially isolated, then. That's what your thread title implies.

 

[Mark Quayle]

Not trying to be difficult, but it isn't obvious to me that a person who is less involved in civic groups is going to be less willing to participate in a poll.

Not necessarily, but I can tell you I'm not the only one who will hang up the phone or tell the guy who stops me on the street, or knocks on my door to ask me slanted questions to get lost.

 

[essentialsaltes]

Doesn't say anything about the percentage of Trumps base that is socially isolated, then. That's what your thread title implies.

It's definitely true that isolated people are disproportionately Trump voters. I think it has to follow that Trump voters are disproportionately isolated. It doesn't mean that they are majority isolated. Just that the socially isolated make up a bigger portion of Trump voters than of Biden voters.
(And this could be a reason why polls underestimated Trump's support.)

 

[Landon Caeli]

It's definitely true that isolated people are disproportionately Trump voters. I think it has to follow that Trump voters are disproportionately isolated. It doesn't mean that they are majority isolated. Just that the socially isolated make up a bigger portion of Trump voters than of Biden voters.

It would still be the pollsters fault for failing to know how to poll properly.

...But I have a hunch this is more about showing how deplorable the deplorables are.

Eh..?

 

[essentialsaltes]

It would still be the pollsters fault for failing to know how to poll properly.

That's the right wing kneejerk I asked people to avoid.

...But I have a hunch this is more about showing how deplorable the deplorables are.

That's the left wing kneejerk I asked people to avoid.

Instead, we might think about the relationship between people who are disconnected from society and distrustful of institutions, and a Trumpian figure who is dismissive of current institutions and norms.

 

[Nithavela]

It's definitely true that isolated people are disproportionately Trump voters. I think it has to follow that Trump voters are disproportionately isolated. It doesn't mean that they are majority isolated. Just that the socially isolated make up a bigger portion of Trump voters than of Biden voters.
(And this could be a reason why polls underestimated Trump's support.)

Without the percentage of people who are isolated, the statistic is pretty meaningless. It could just be 2% of all people polled, that would be statistically insignificant.

 

[cloudyday2]

Well, I don't know about civic groups. The OP talks about social isolation, and the article refers to this study, which shows evidence that "social network isolation is negatively associated with unit response. [i.e. responding to surveys]". The study provides some data that confirms a widely held hypothesis about social isolation and willingness to take surveys.

That's interesting. I would have assumed that if there is any relationship at all it would be the opposite. In other words, if I'm isolated at home and the surveyor calls then I might be so bored that I would actually answer those 100 questions. But if I'm socializing at the Elk's Lodge or something then I might be too busy talking to my friends to spend all that time on the phone with a surveyor.

It would be interesting to know how these political polls work today. I helped make calls in 1988, and almost nobody would answer my questions. There must have been a lot of educated guesswork even in those days, and now it has got to be far worse.

 

[essentialsaltes]

Without the percentage of people who are isolated, the statistic is pretty meaningless. It could just be 2% of all people polled, that would be statistically insignificant.

"In our pre-election survey on the strength of Americans’ social networks, we found that nearly one in five Americans (17 percent) reported having no one they were close with, marking a 9 percentage point increase from 2013."

 

[Nithavela]

"In our pre-election survey on the strength of Americans’ social networks, we found that nearly one in five Americans (17 percent) reported having no one they were close with, marking a 9 percentage point increase from 2013."

Now that's just sad.

 

[FireDragon76]

Personally I think the discrepancy between Trump's poles and actual Trump voter turn out are the 'closet' Trumpsters. They are the ones who are embarrassed to admit they voted for him.

I think it's likely that @essentialsaltes is on to something. Loneliness, substance abuse, and mental illness are more common in many rural areas, especially in the western US. Those types of things tend to be associated with less social engagement, and those same areas voted overwhelmingly for Trump.

 

[FireDragon76]

I love it when we get to play the guessing game on why Trump voters are at fault for failed polls.

View attachment 289382

It's not about fault, just trying to understand why the poling didn't reflect actual election results.

 

[Larniavc]

Id like to hear more. How is discussing important things with people in anyway related to political correctness or directing their thinking? I discuss important stuff with my wife all the time, and it has nothing to do with either.

I remember reading somewhere that people with right wing views tend to be over represented in people who exhibit a strong tendency to be self reliant and focussed on being able to live independently from or beholden to other people. It was called a ‘frontier mentality’ where practicality and personal ability were over represented.

So I can quite see how independent people who value self sufficiency and personal autonomy would live in more isolated areas (likely having a smaller social network) may tend right wing in their voting habits.