Question regarding poll methodology. For most of the polls the MSM and Real Clear Politics claim are showing Clinton in the lead do they call people or do it via internet? If its a call might that not lead to people just saying Clinton because its the socially acceptable choice, or saying Johnston or Stein because well, its just a poll, in greater numbers than they will when pressure is on?
Is there a way to conduct objective anonymous polls and is anyone doing this?
Other urls found in this thread:
sharylattkisson.com
youtube.com
twitter.com
OVER SAMPLE DEMOCRATS
this
here OP -
Trump performs better among independents–Reuters/lpsos poll
sharylattkisson.com
I wasn't getting at that I was saying that when I took psychology class, I forget the jargon name but if you confront people to state an unpopular opinion they might lie or omit stuff so why isn't that at play here? We hear a lot in the MSM about people who "quietly support Trump" never about people who "quietly support Clinton" so if polling is anonymous on Election Day won't people just vote who they want for rather than who they think others want them to?
If I voted for Jill Stein in 2012 and I'm planning to vote Trump this year do I count as an independent? Or what, definitionally, makes someone an independent?
Also, if its a phone poll (and I think it is) I remember hearing years ago that a big issue is that only certain people answer the phone for pollsters and they tend to be older. So might that not be providing a representative sample?
The only time in my life I've ever been polled was when Obama wanted to know more than how many people lived at my address for the 2010 census so he could build his race database and destroy "segregated" areas.
Real Clear sold out last year.
Don't ever bring this up again.
Should I never poll you or your wife's sons again as well?
>Or what, definitionally, makes someone an independent?
You can specifically register as an independent or you can not register as democrat/republican. You can also vote for anyone you want, regardless of party, no matter which party you're registered with.
I'm registered as green party. So do I count?
That just means in most states that you have special status like being able to vote in the green primary.
You are not answering if that means I count as an independent or not.
As far as the state is concerned? Obviously not.
t. registered independent, which means jack shit other than that I can't vote in any primaries in my state
I'm registered green party because its ALSO the DC Statehood party, and I support DC becoming a state.
Assuming no one is lying, polling can be very accurate.
Now if people are lying, then things get dicey:
Pollster lying - oversampling democrats for example will, though i really don't think it's as exaggerated as trumptards claim and is 1% tops
People lying or deliberately dodging polling for fear of being known for supporting a controversial candidate - okay this is bad, it's impossible to detect until it's too late and impossible to correct (in UK it's called the shy-tory factor)
OK, good for you then. I almost re-registered as a republican this year, which I've never done before, just to vote Trump in the primary.
Putting that aside, can you give me your elevator speech on why you think DC should be a state?
Not lying but rather lets say you hold an unpopular opinion and someone asks you in a personal way, that is on the phone ("what do you think?"), you are more likely to lie than if its just an anonymous way of asking.
I think we should have Congressional representation. I get that yes, it would give votes to Dems and Blacks and whatnot but D.C. is majority white now and getting moreso and that would give us power to regular our own affairs.
I don't know OP, but I know better than to trust CNN's fucking polls.
Fuck CNN, this year they're giving RT a run for their money for the "most partisan" news organization medal.
Well what I'm asking is, would there be a way to conduct a representative sampling of the registered voting population by an anonymous means without a risk of fraud? And why can't anyone answer the simple question of, are these polls conducted by phone? And what is the average age of the poll respondents compared to voters?
>I think we should have Congressional representation
Isn't Congress pretty much already your city council? Why would you want to have like Puerto Rico levels of representation in Congress instead of just having your own municipal government?
That is called 'push polls', and implies pollster lying
There are people genuinely afraid that they'll have their support of a controversial candidate harm them in the future. Such as by the government or through data leaks (to weird groups such as antifa and BLM?). And are, hence, motivated to lie regardless of what you do.
No we don't have Congress and due to the Home Rule Act they can overrule any of our laws we don't like. I want to have a vote, a say in the affairs of my country by electing Congressmen and Senators. I don't think that's unreasonable. I want to be able to pass laws here that outsiders from other states can't unilaterally overrule, as is the case now.
Right, but wouldn't it be more effective to have your own local government like all other cities rather than relying on the federal government that will always be controlled disproportionately by people from everywhere other than in your city?
I'm guessing you have to pick your sample group individually based on the right criteria, and make it a very large group.
No. We have no CHOICE but to submit to federal oversight at this point due to the 1978 (IIRC) Home Rule Act so a senator from any state can decide any of our laws are bunk and prevent us from enacting them.
Well that doesn't tell us anything about WHAT THEY ARE ACTUALLY DOING HOLY FUCK WHY DOES NOBODY KNOW HOW POLLS ARE DONE
Assume I have The Brady Bunch over here of all shades and varied retardation. Doesn't make what I just told you any less true.
Do an email poll with one key one vote.
Don't use a list of emails that know they are going to be polled.
Well why does nobody know if these polls are conducted by phone or what the average age of those polled is? If this were a clinical trial the FDA would have a shitfit.
Because that's not the point of the polls. Somebody already gave you the key word "push polls." Look it up.
I assume they are done via phone I just wanted to confirm. My point is that many young people don't answer unknown numbers on their cell phones and its WAY easier to block an unknown caller now then 20 years ago with a landline.
Oh and I meant to add that means it oversamples old people because young people have cell phones and just don't answer. I'm not saying I think pollsters are being insidious just methodologically not rigorous in terms of getting a representative sample.
Last I heard it's not kosher to cold call cell phones, so that's bias right away. I sure as hell don't pick up my landline if I don't recognize the number.
However, now that my memory has been jogged, I did get something in the snail mail asking me to respond to a poll. It was all shady and undocumented, so I took a few minutes to determine that it was from a small time grad program a couple of hundred miles away. I emailed the student running it, but he didn't respond; I was too lazy to ask his advisor.
OK so they are oversampling old people and how does that skew the population?
>tfw no qt poli sci grad student bf
I'm not going to speculate. If I need proper statistics, I outsource it to people who concern themselves with it. Usually I just make up my own metrics to get a quick reading on what I'm interested in so I can make effective day-to-day decisions. I'm more of a tactician than a strategist.
I'm saying we should be able to see transparent methodology if MSM is going to trumpet TRUMP BTFO IN POLLING CANT WIN everyday. If they are just calling landlines and that isn't most people what's the point of the polls.
You're being too rational. This isn't peer-reviewed published science. This is psychological warfare.
I doubt its either. Its probably lazy middle aged yuppies using accepted outdated methodology. It isn't a representative sample and if Trump wins in a landslide because they didn't ask people on the internet or cell phones what they thought and only got grandpa's landline's opinions that's their fucking deal and shouldn't be touted be respected press as verifiable fact. Also Nate Silver is balding.
Dudebro the entire media is psychological warfare. They wouldn't plaster polls all over the place if they weren't trying to push a message. Why do you think Trump made such a show of talking about the polls in all of his rallies early on? He's not an idiot.
Here, watch this. It's educational.
youtube.com
I'm fapping but maybe in a bit. That does make me sad to know there's zero objectivity or science to it.
they over sample democrats on most polls
Retardus +6.9% more democrats
Economist/yougov
77. Generally speaking, do you think of yourself as a ...?
Democrat . . . . . . . . . . . . . . . . . . . . . . . 33%
Republican . . . . . . . . . . . . . . . . . . . . . . 24%
Independent . . . . . . . . . . . . . . . . . . . . 40%
Other . . . . . . . . . . . . . . . . . . . . . . . . . . . 3%
NBC no sample data
Bloomberg No sample data
Retardus +6.9% more democrats
Economist/yougov
77. Generally speaking, do you think of yourself as a ...?
Democrat . . . . . . . . . . . . . . . . . . . . . . . 33%
Republican . . . . . . . . . . . . . . . . . . . . . . 24%
Independent . . . . . . . . . . . . . . . . . . . . 40%
Other . . . . . . . . . . . . . . . . . . . . . . . . . . . 3%
NBC no sample data
Bloomberg No sample data
yeah you are factually wrong according to the polls own data.
If Trump wins or barely loses, this will be worse than the Brexit vote for polling companies.
There are more democrats than republicans. You don't just pick an even number of republicans, democrats and independents to get an accurate poll. These companies base their methodology on who came out to vote in previous elections.