In 2016, President Trump won several large states (such as Florida, Pennsylvania and Wisconsin) by very narrow margins, gaining all their electoral votes in the process, even as Senator Clinton claimed other large states (such as California, Illinois and New York) by much wider margins. This cemented his path to victory in 2016.
In 2020, going by the data from Real Clear Politics, President Trump draws about par with his chances in 2016. Most polling outlets vary around this number but not by a significant amount from one another. Only two agencies had shown President Trump leading over Joe Biden until recently: Rasmussen Reports and Emerson. Even these two outliers have been indicating a Biden victory in recent times.
But what about the margin of error? The key to adding context to this lies in the difference between polled populations and the voting population. As of 2016, there were 250,056,000 individuals of voting age in the U.S., of whom 138,847,000 (55.5%) turned out to vote. Polls typically survey about 1,000 to 5,000 participants either telephonically, via survey forms or via online panels. Courtney Kennedy, a Director of Survey Research at Pew, outlines the following takeaway points about polling:
- A poll may label itself “nationally representative,” but that’s not a guarantee that its methodology is solid.
- The real margin of error is often about double the one reported.
- Failing to adjust for survey respondents’ education level is a disqualifying shortfall in present-day battleground and national polls.
The third point adds a substantial nuance with regard to polling. Since people with higher levels of formal education are more likely to participate in surveys and self-identify as Democrats, the potential exists for polls to over-represent Democrats. There is a clear call to adjust the weighing of polls to deliver a more accurate verdict than the one reported in 2016. At this moment, there is no clear indication that this has been implemented by polling agencies for the 2020 elections.
How important are white voters? According to Pew, as of 2018, white voters accounted for 67% of the electorate, Hispanic and Black voters account for 13% each while Asians and “Others” account for 4% and 3% respectively. However, the greatest growth in eligible voter headcount has been among Hispanic, Black and Asian voters.