polling agencies continue to snub empirical data collection rules

justoffal

Diamond Member
Jun 29, 2013
23,729
15,800
1,405
It looks like the pollsters have learned nothing from their embarrassment of the previous season of false polling and totally contradictory results.
202010061000-ftxzmd5p53.jpeg
 
Last edited:
Because polls are now designed to influence opinion rather than report on opinion
 
It looks like the pollsters have learned nothing from their embarrassment of the previous season of false polling and totally contradictory results.
View attachment 398113

It weighs against Democrats instead. Produce a poll with a bias towards Trump. The fact is that we also have the 2018 election results as well. We also have the primaries in which the turnout was much higher in 2020 than it was in 2016.

Here is another thing. The exit polls from 2016 have no context. They have 31% of voters as independent. That is not true. When you look at how people vote, there are far fewer independents. There are many independents who favor 1 party over the other. Frank Luntz says that based on voting behavior, there are no such thing as independents. I don't necessarily agree with that. The NBC/Wall Street Journal poll has 28% strong Republican vs 22% strong Republicans. 7% are essentially somewhat Democrat vs 5% somewhat Republican. 10% are independent/lean Democrat vs 9% who are

Also in 2016 a lot of people stayed home. There were a number of states where Clinton underperformed. In Ohio, she got fewer votes than Romney in 2012 and McCain in 2008. In Minnesota, it was close not because Trump got a lot more votes than Romney. but because Clinton underperformed Obama's numbers in 2012. There were also a number of states that Clinton won where she got fewer votes than Romney did in losing them. Trump got fewer votes in winning Pennsylvania than Obama did in winning them in 2012.
 

Forum List

Back
Top