Approval Rating Polls, and How Tremendously Different They All Are

[JimRaynor]

Five Thirty Eight has an aggregate of all the polls on Trumps job performance which when every single poll is combined the aggregate score currently is 40% approval but take a look at how wildly different each poll is.

https://projects.fivethirtyeight.com...roval-ratings/

Rasmussen Poll, which is right leaning has him at 50% approval. Politico and Morning Consult have him at 47% approval, Monmouth University which is one of the most credible polls has him at 45% Approval.

Then you have left leaning Gallup and CNN polls which have him a 37% approval, and then some less known polls like American Research Group which puts him at 36% approval, while another lesser known poll Harris Interactive puts him at 45% approval.



Question is, how are these polls being conducted that they're all getting such vastly different results. Comparing to Rasmussens high of 50% to say Gallups 37%, thats a 13 point swing, who and in which states are they asking these questions.

[Thraxox]

Wasn't there a study that approval ratings somewhat depended on the question being asked?

[Scipio2009]

Polls come down to who you ask

Five Thirty Eight has an aggregate of all the polls on Trumps job performance which when every single poll is combined the aggregate score currently is 40% approval but take a look at how wildly different each poll is.

https://projects.fivethirtyeight.com...roval-ratings/

Rasmussen Poll, which is right leaning has him at 50% approval. Politico and Morning Consult have him at 47% approval, Monmouth University which is one of the most credible polls has him at 45% Approval.

Then you have left leaning Gallup and CNN polls which have him a 37% approval, and then some less known polls like American Research Group which puts him at 36% approval, while another lesser known poll Harris Interactive puts him at 45% approval.



Question is, how are these polls being conducted that they're all getting such vastly different results. Comparing to Rasmussens high of 50% to say Gallups 37%, thats a 13 point swing, who and in which states are they asking these questions.

Rasmussen polls the folks that they poll, but my understanding is that they only poll folks by landline, and by automated machine; the population of homes in the country that have landlines are generally older/more rural.

Gallup generally has polls consisting of 70% of cellphone users/30% of landline, with live interviewers asking the question; you end up with a generally younger, generally more urban, and generally more politically conscious polling audience.

You add that most of the polls have 1500 polled speaking for a national audience, and it's sensible that a range opens up.

538, in aggregating everything, gives a better population picture, in a general sense; the sample size goes from 1500 to 15000, you end up getting a good blend of landline and cell phones (better representation of the homes in the US), with a mix of automated and live interviews.

[Motorcity Cobra]

Five Thirty Eight has an aggregate of all the polls on Trumps job performance which when every single poll is combined the aggregate score currently is 40% approval but take a look at how wildly different each poll is.

https://projects.fivethirtyeight.com...roval-ratings/

Rasmussen Poll, which is right leaning has him at 50% approval. Politico and Morning Consult have him at 47% approval, Monmouth University which is one of the most credible polls has him at 45% Approval.

Then you have left leaning Gallup and CNN polls which have him a 37% approval, and then some less known polls like American Research Group which puts him at 36% approval, while another lesser known poll Harris Interactive puts him at 45% approval.



Question is, how are these polls being conducted that they're all getting such vastly different results. Comparing to Rasmussens high of 50% to say Gallups 37%, thats a 13 point swing, who and in which states are they asking these questions.

Did you get this inbox from a drunk. -MEGA- last night telling you half your post should be in the Trump master thread or was it just me?

[-MEGA-]Not that getting banned is that big a deal. But you need to take it down a couple notches.

I’d also appreciate it if you stop spamming the lounge with your bull crap. Half that stuff can go into the trump master thread. Just cause you have a thought doesn’t mean it’s worthy of a thread.

Many of your threads are cool and envoke thought but some are just outright lame and racist. That’s right, racist. Grow up.

Why should you be allowed to get away with insulting white people in a manner that if somebody else said the same thing but about black people, they would get banned?

I’d appreciate if you kept that in mind when posting. If it don’t work both ways then you probably shouldn’t say it.

All that aside, you’re probably a pretty cool dude and contribute to this site. I’ve overlooked stuff time and time again because unlike others, you are actually a boxing fan.

This is your first formal warning and the last before you take a banning. Be smart, or not. It’s up to you.

I looked thru my posts and that only ones I could find was the Mueller, which has been up for months, investigation thread or maybe the guns in schools thread. Maybe he sent that to that wrong person or maybe he's flat out lying about me again

[GhostofDempsey]

We all saw just how pathetic polls are during the campaign. They were all showing Hillary as a strong favorite to win. Pols can be constructed in such a way to ensure a predetermined outcome on the poll. Some of them will obviously we inflated to reflect viewership. A CNN poll will not be favorable towards Trump, but a Fox poll more likely will.

[Bobby Deez]

Lot of shy Trump voters. Aslong as you have most of the media throwing around whatever ist a Trump supporter is supposed to be you'll have these crazy polls.

[JimRaynor]

Here is how CNN skews/manipulates its polls.

"CNN polled a total of 1,016 adults from Feb. 20-23 nationwide and had respondents self-identify their political affiliations. Thirty-three percent of respondents identified as Democrats, 44 percent as Independent or members of another party and 23 percent as Republicans."


http://dailycaller.com/2018/02/26/cn...-polling-data/

The percentage of Democrats and Independents that they poll are both significantly higher than Republicans.