It looks like you're using an Ad Blocker.

Please white-list or disable AboveTopSecret.com in your ad-blocking tool.

Thank you.

 

Some features of ATS will be disabled while you continue to use an ad-blocker.

 

How can 1,500 out of millions be an accurate poll sample???

page: 2
4
<< 1   >>

log in

join
share:

posted on Oct, 14 2012 @ 06:33 PM
link   
reply to post by inverslyproportional
 


I would say there is no credibility at all in the election system in america, and this is pathetically ironic because americans like to export their extremely flawed version of democracy to everyone else.



How the hell can people have a well rounded view of the parties and candidates come ballot time, when ONLY TWO CANDIDATES FROM THE SAME TWO PARTIES are allowed to debate each other every single time?

Remember one of the stipulations of making the debate is having at least 15% of the popular support in each of the officially recognised polls. The problem is that the poll mafia will not even list the other parties and their candidates so that means people have to do their own research ahead of time.

It is a vicious circle of lies and fraud. Gary Johnson filed an antitrust lawsuit against the debate commission but I have the feeling it was more for publicity than anything else. Ralp Nader was denied entrance into the debate hall back in 2008.



posted on Oct, 14 2012 @ 06:49 PM
link   
reply to post by EarthCitizen07
 

There is an online sample size calculator here. There's about 250 million adults of voting age in the US. To accurately model that population with a 4% margin of error and a 95% confidence interval, you need to sample 600 subjects. That means you can be 95% confident the poll will be accurate to within +/- 4%. (Or, it will be accurate to with +/- 4%, 95% of the time.) If you want to increase the confidence level to 99%, you need 1,040 samples. That's just math, of course. If you want meaningful results, you have to design a good survey and execute it well. If you have a bad survey instrument or a bad sampling method, it doesn't matter how many people you sample, you're going to get a biased result.



posted on Oct, 14 2012 @ 06:53 PM
link   

Originally posted by BrokenCircles
reply to post by EarthCitizen07
 



How can 1,500 out of millions be an accurate poll sample?

That's why it is called 'A Sample'.


Sample

1: a representative part or a single item from a larger whole or group especially when presented for inspection or shown as evidence of quality: [A specimen]

2: a finite part of a statistical population whose properties are studied to gain information about the whole




Is it an accurate sample? I don't know. Probably is though.
Is it accurate for the still yet to come definitive conclusive results? Nope.
Why not? because it is just a sample.









edit on 10/14/12 by BrokenCircles because: (no reason given)


Um, no -- it's impossible to be an accurate sample.

1,500 can never be representative of 311 million.

311,591,917
1,500

Do you see why yet?

Also, where you gather the sample matters. When talking about something that is as variable as political alignment, you need a much bigger sample rate than 0.0001%.

1% of 300 million is 3 million.

1,500 is not even 1% of a million. 10,000 would be 1% of a million.

If you want to have an accurate sample, you'd have to at least have a sample of about 30%.

In conclusion, these samples are propaganda. They'd like you to believe that since they are the experts that they must be giving you accurate data. It couldn't be anymore phony baloney.

And to prove it works, You even said "Is it accurate, Dunno but it probably is."

It's not even CLOSE.

If what you are testing is NOT variable, or has a very small variable range, small samples provide an incredible amount of evidence.

When what you are testing is extremely variable, a sample size of even 1% is waaaaaaaaaaaaaaaaaaaaaay to low to get any kind of useful data.

That's just the way it is.
edit on 14-10-2012 by Laykilla because: (no reason given)



posted on Oct, 14 2012 @ 07:07 PM
link   

Originally posted by FurvusRexCaeli
reply to post by EarthCitizen07
 

There is an online sample size calculator here. There's about 250 million adults of voting age in the US. To accurately model that population with a 4% margin of error and a 95% confidence interval, you need to sample 600 subjects. That means you can be 95% confident the poll will be accurate to within +/- 4%. (Or, it will be accurate to with +/- 4%, 95% of the time.) If you want to increase the confidence level to 99%, you need 1,040 samples. That's just math, of course. If you want meaningful results, you have to design a good survey and execute it well. If you have a bad survey instrument or a bad sampling method, it doesn't matter how many people you sample, you're going to get a biased result.



It doesn't work that way.

It would work that way if you are talking about something other than opinions.

That math doesn't account for the huge variable of opinion. That works if you are dealing with static values, not unknown variables.

That's SCIENCE, not math.
edit on 14-10-2012 by Laykilla because: (no reason given)



posted on Oct, 14 2012 @ 07:37 PM
link   
reply to post by EarthCitizen07
 


Forget about the polls. How about not allowing them to debate in the presidential debates!



posted on Oct, 14 2012 @ 09:30 PM
link   

Originally posted by Laykilla

In conclusion, these samples are propaganda. They'd like you to believe that since they are the experts that they must be giving you accurate data. It couldn't be anymore phony baloney.

And to prove it works, You even said "Is it accurate, Dunno but it probably is."

It's not even CLOSE.


So just me and you noticed?

Everyone else is too dumb?


Or maybe the typical "I don't give a # about anything" folks....just give me a bud light and lots of bootey. Helll yeaaahhh. Yeeeehhhhhaaaawwww bammmm bammmm (wink, wink) republicans



posted on Oct, 15 2012 @ 03:10 PM
link   
reply to post by EarthCitizen07
 


I live in Ohio, and the phone rings all day long here. It's all pollsters and everyone knows who it is so very, very few of us answer it when it rings without screening first. No one that's under 60 yrs old - unless they're driven by politics - ever picks up if it's a polling call. Frankly, I have no idea why they think that they've got a sample of normal people with any of these polls. Normal people - at least in this swing state - are completely done with this crap.

I'll say one thing, even my 76 yr old bedrock conservative mother-in-law is sick of seeing Mitt Romney's face on her TV screen, and she wishes that the spooky-sounding voice-over lady in the American Crossroads Super PAC commercials would go away and never darken her living room ever again. If she was ever entertaining a Mitt Romney vote, the last month of back-to-back ads has killed that notion off forever.



posted on Oct, 15 2012 @ 03:21 PM
link   
reply to post by Laykilla
 

Originally posted by Laykilla

Um, no -- it's impossible to be an accurate sample.
To the contrary, it is impossible for it not to be an accurate sample.
After all, it is only a sample.




Originally posted by Laykilla

1,500 can never be representative of 311 million.
I never claimed that it was. You and the Op are both arguing about just how accurate/inaccurate this small sample size could potentially be, in reference to the eventual end results.

Obviously, a larger sample will always produce more reliable results, but a sample is not meant to be an entirely accurate representation of the end results as a whole. The purpose of a sample is to get some sort of an idea, but a sample can never be completely conclusive.



Originally posted by Laykilla

When what you are testing is extremely variable, a sample size of even 1% is waaaaaaaaaaaaaaaaaaaaaay to low to get any kind of useful data.
1% is better than 0.




I know that it is basically just semantics (which is why I am done here), but you guys need to say it right. lol. A sample is still a sample, regardless of whether or not it eventually proves to have been an accurate representation of the whole.



posted on Nov, 26 2012 @ 12:07 AM
link   
It is a cover for election fraud.

Corporations and private interests buy the election.

People vote, see that their candidate lost, and say "oh well, they did poorly in the polls anyway."



posted on Nov, 28 2012 @ 08:13 PM
link   
The problem with poll accuracy is this. The statistical model is accurate only if all of the following are true.

1. Everyone understood the question and answered truthfully.

2. The surveyors accurately gave the surveys and didn't overtly or inadvertently prejudice the respondents answers.

3. The questionnaire was neutral and didn't elicit the responses it was seeking.

4. Everyone that was polled answered the question.

The problem is 1-4 are never entirely true. Real response rates are about 70%. In addition, questionnaires have leading questions and formats that prejudice answers. Surveyors fake data to do less work or ask questions too quickly or skip questions if they're paid by the survey which many are. As to point 1, some people don't even know who the current President is. There are also mistakes made coding the data into the computer.

Sampling methods that are used are far more complex that just taking a random sample of Americans. They use stratified cluster sampling methods so they can make inferences about certain parts of the country but there are dangerous in using these complex survey designs.

With an unlimited budget you could do the perfect poll but no one ever does that so were left with differences between polls and survey reports that are frequently very incomplete and only give us the poll results they want us to see.




top topics



 
4
<< 1   >>

log in

join