Donald Trump was again underestimated in opinion polls.
On the one hand, this poll error is quite normal. We’re not talking about huge polling gaps, and polls still correctly “appointed” the election: Joe Biden won. But nonetheless, it seems as if the polls still couldn’t catch some of Trump’s support.
There are a number of possible explanations for this, and there are no definitive answers, but one thing I came across in my poll is that there is an increase in the number of Americans who are more socially excluded from society. And these voters disproportionately support Trump.
The idea that some of Trump’s supporters are more likely to break away from civilian life is hardly new. During the 2016 Republican pre-election, Yoni Appelbaum noted in the Atlantic region that Trump disproportionately supports support from those who claim to be independent of their citizenship. An analysis by Emily Ekins, a conservative-minded Cato Institute, showed that while Trump continued to strongly support white Lutheran Protestants, they were actually more positively assessed by supporters who did not participate in regular religious practice. Finally, research by David Shor, a Democrat pollster, on the 2016 election echoed what we found in our own pre-2020 survey: There was a lot of momentum for Trump among white voters with low social confidence – a group that researchers found that they were less involved in telephone surveys.
In our pre-election survey of Americans ’social networks, we found that almost one in five Americans (17 percent) reported not having anyone close to them, an increase of 9 percentage points compared to 2013. What’s more, we found that these socially isolated voters are much more likely to view Trump positively and support his re-election than those who have stronger personal networks. Biden was strongly supported by registered voters with larger social networks (53-37 percent), but Trump had an advantage among voters without close social ties (45-39 percent).
And this was especially true among white voters, even after taking into account differences in income, education, and racial attitudes. Sixty percent of white voters who have no direct social network preferred Trump, compared to less than half of white voters with stronger social ties (46 percent).
Of course, the fact that Trump was a significant attraction among those who profess to be socially connected is not convincing evidence that these people were left out of the polls. In fact, it is really First, it’s hard to catch these people: For example, many who fall into this group would not respond to a survey.
Nevertheless, there is good reason to believe that some people who do not have a strong social connection may be missing from opinion polls. Social isolation or lack of social integration has long been thought to reduce the willingness to participate in surveys. Americans who are alienated or alienated from society do not feel compelled to participate in the surveys because of their civic duty.
Researchers at the University of Nebraska-Lincoln and the University of Hiroshima have found that Americans with weaker social connections are less willing to cooperate in survey requests, and some survey estimates may be “systematically biased due to lack of participation by socially isolated people”. Our survey was not suitable for exploring why people did not participate, but we found that those with smaller social networks are much less politically committed. For example, Americans with at least four social ties are three times more likely to turn to an elected official in the past 12 months.
However, there may not be a simple solution to this problem. Providing financial incentives to strengthen cooperation and increasing the on-site duration of the survey can help. These are common research research practices to increase the representation of hard-to-reach groups. But it is not yet clear whether encouraging respondents or extending interview time would increase participation rates among socially isolated Americans. A 2000 study found that increasing the interview period from five days to eight weeks hardly changed the end result — although the Republican party’s perception was more positive in the survey, which included a longer interview plan.
The consequences of this, of course, extend far beyond our choices. As Americans become more isolated from society, they disconnect from local and national institutions and invest less in their ongoing operations. In addition, they are more prone to mistrust of political processes and believe they serve illegal purposes. And they may lose their faith in the messy and multifaceted process of democratic change.
Over the next few years, we will spend a lot of time explaining the country’s growing social, economic and political problems, but we must not forget that political reforms and economic solutions will not address the fundamental problems of loneliness and social isolation in this country. This may mean that we are less and less aware of what a growing part of the country is feeling.
