Why polls may all be bogus

According to Pew only 6% of the people they call to ask survey questions of respond; that's down from 36% in 1997 and around 18% in 2008.

While Pew contends that low response rates don't impact the accuracy of polls it's unclear if that's really true.

Clearly with such a low response rate people who do engage are probably different than people who don't.  Now they may reflect vote tally's since both voting and responding to polls show a significant interest in participation but they could be way off the mark when asking questions about non-voting issues like abortion, same-sex marriage etc.

Part of the problem may be that Democrats are weaponizing society so that if someone says that they like Trump they will face punishments ranging from people refusing to talk to them to people boycotting their business.  Hence it's not unlikely that not only are the people who are willing to respond more engaged than non-respondents but they may be more leftist.

That's why when you see a poll that is supposedly showing that the American people are supporting a radical Democrat agenda you should take it with a grain of salt.