Polls are horribly unreliable on every politically charged issue because most people will give the answer of the party they identify with no matter what they think personally. Add in that the way a poll is structured has a significant impact on the way another 10 percent or so of the population answers by giving two or three leading questions before asking the 'key' question to sway that percentage of the population that isn't axiomatically using their party of choice stance and you can get any poll to show a 'majority' agree with statement X.
You can do the same manipulation with Abortion or Gun Control or Renewable energy Subsidies simply by how you work the questions. Abortion is a clear and concise neutral word but if I ask you, 'Are you pro-Choice or Anti-Choice?" clearly the answer I am hoping for is you are 'pro-choice', which sounds ever so much nicer than 'pro-abortion'. By the same manipulation if I replace the words so I am asking are you 'pro-life' or 'anti-life' clearly I am hoping you are going to say 'pro-life'.
So you want the poll to say most Americans think President Trump is a Chump, Climate Change Is Real. You start the poll with a boilerplate statement that goes something like this, '97 percent of climate scientists say humans are causing global warming and it is dangerous, especially to the minority poor and children who will be most impacted. Do you think climate change is a threat?' Or you can be a bit more subtle and spread the statements out into a series of lesser statements so the conclusion is the same but the presentation is a bit less heavy handed.
This is exactly why polls about President Trump were so far wrong, the Media hates the man so every poll they took started from the POV that all right thinking people were against him. In a country as evenly divided as the USA that sways enough of that squishy 10 percent who just say what they say to get along and the skewed result looks like a landslide. It is the same effect in reverse with President Obama, the media love the guy so when they ask people if the country was heading the right direction under his leadership the result was 65 percent say NO or Heck No!, but when they ask if people approve of the President personally they say "Oh yes of course' no matter how they actually feel about him. People don't want to be called racists for not liking the First Black President, so when a poll is taken they lie in the face of the poll takers. It is the Bradley effect, not at all unknown, but if the media is dishonest they ignore the effect and just report the numbers that say what they wish were true.
The reality is, most Americans Do Not Believe Global Warming is A Threat. You can tell because their lifestyle is 180 degrees away from what a true believer would act. They buy trucks and SUV's for personal transport, fly cross country for vacations, have their A/C set to 70 or less in the summer and their heat set to 75 or higher in the winter. The examples are as endless as the number of ways Americans profligately waste fossil fuel energy. https://ballotpedia.org/Bradley_effect
The Bradley effect, sometimes called the Wilder effect, is a concept that attempts to explain discrepancies between voter opinion polls and outcomes in elections where white candidates campaign against minority candidates. Adherents of the Bradley effect believe that some voters will tell pollsters that they are undecided or likely to vote for a minority candidate but will vote against the minority candidate on Election Day. It was named for Tom Bradley, an African-American candidate who lost the 1982 California gubernatorial race despite having a lead in the polls going into the election.
A related concept is social desirability bias, which describes the tendency of individuals to "report inaccurately on sensitive topics in order to present themselves in the best possible light." According to New York University professor Patrick Egan, "Anyone who studies survey research will tell you one of the biggest problems we encounter is this notion of social desirability bias." Some researchers and pollsters theorize that a number of white voters may give inaccurate polling responses for fear that, by stating their true preference, they will open themselves to criticism of racial motivation.
Some analysts have dismissed the Bradley effect theory; others have argued that it has played a diminishing role in recent elections. One analysis of 180 Senate and gubernatorial elections between 1989 and 2006 suggested that "before 1996, the median gap [in public polling data and actual vote share] for black candidates was 3.1 percentage points, while for subsequent years it was -0.3 percentage points."