What do you know about "The Simpsons" and the First Amendment?

[Airscrew]

Originally posted by Dead Man Flying
Airscrew, random sample surveys or variations thereof are actually very accurate if performed scientifically and without bias.

-- Todd/Leviathn

The trick there is scientifically and without bias.  This particular survey just doesnt seem very accurate to me.  the survey was only run for 2-3 days and only included 1,000 out of almost 300 million people.   Usually its not the survey data thats suspect anyway, its the interptation of the data and how the data is used that leads me to distrust surveys.

I think that before I made a statement like this "It's disappointing that Americans continue to be ignorant of First Amendment freedoms,, I would try to survey a larger sample, 10,000 or more

Originally posted by Mickey1992
The Simpsons is on every day at 5:00 and 5:30. Maybe if someone started showing The First Amendment on TV twice a day people would know more about it.

Remember ABCs After School Rock ?  I'm Just a Bill and I'm sitting on Capital Hill,  Or Conjuction Junction whats your function.   Those spots ran about a minute or so and managed to teach something and entertained.
But Mickey's right, we're bombarded with commerical slogans and TV shows that are empty of any meaningful content.

[Dead Man Flying]

Originally posted by Airscrew
The trick there is scientifically and without bias.  This particular survey just doesnt seem very accurate to me.

[/b]

Why does it not "seem" accurate?

Usually its not the survey data thats suspect anyway, its the interptation of the data and how the data is used that leads me to distrust surveys.

[/b]

The real issue that should worry you about survey data is not so much the interpretation as the question wording and question order.  You can draw a perfectly random and representative sample of the population, but if you ask poorly-worded questions or lead people on through question order, you can bias the results.  I suppose that's possible in this survey, and I didn't read the text of survey itself (and note that all good survey articles should include a link to the actual survey questions), but asking fact-based questions like naming the rights in the First Amendment or name the Simpsons doesn'tseem prima facie biased to me.

I think that before I made a statement like this "It's disappointing that Americans continue to be ignorant of First Amendment freedoms," I would try to survey a larger sample, 10,000 or more

[/b]

Again, why?  You may achieve a 3-4% reduction in your margin of error by doing that, but you do so at a tenfold increase in the administrative costs of the survey.  Thus while the survey, if unbiased in every other way, contained 10,000 respondents instead of 1,000, the estimators would in fact be more accurate but not substantially so.  We would achieve 100% accuracy with a total population census, but would it be worth the increase in cost from 10,000 respondents?  We're talking tens of billions of dollars in administrative costs now.

The fact is that, unlikely as it seems, a survey drawn from 1,000 respondents may be representative of the general population if done properly and scientifically.

-- Todd/Leviathn

[nirvana]

Exit polls aren't random?  They have certain people they will ask at certain polling places?  Alright you got me, what is your idea of a "random sampling".  Let me pull out my text book and write the exact definition so you know what i'm learning.


Random sampling: The key technique employed by sophisticated survey researchers, which operates on the principle that everyone should have an equal probability of being selected in the sample

Exit poll: Public opinion surveys used by major media pollsters to predict electoral winners with speed and precision

Where do you get off saying that more=less?  Are you saying that if they actually took a sample of everyone they would be more wrong then a sample of 1000?  Think again jack, more people equals a more reliable survey because you don't have 1 person representing 300.

[Dead Man Flying]

Originally posted by nirvana
Exit polls aren't random?  They have certain people they will ask at certain polling places?  Alright you got me, what is your idea of a "random sampling".  Let me pull out my text book and write the exact definition so you know what i'm learning.

[/b]

Exit polls are typically self-selected samples and thus not necessarily random.  

Where do you get off saying that more=less?  Are you saying that if they actually took a sample of everyone they would be more wrong then a sample of 1000?  Think again jack, more people equals a more reliable survey because you don't have 1 person representing 300.

Obviously you have not fully read and understood what I posted in the previous message.  Read it again please.

-- Todd/Leviathn

[Sandman]

Someone arguing statistics with DMF.

This should be good.

[nirvana]

Look when I posted last and when you post last, 8:49AM on both...hmmm.

1000 people can give a very good result if done properly and scientifically, had this been done?  Hard to say.  Keep in mind this survey was done by a museum "helping visitors understand the First Amendment when it opens in April."  Could they not just, I dunno, fake the results to show that "You are dumb so you should come to our museum and enlighten yourself"?

I would also like to see what the questions asked were.  You're right saying that the wording can give a biased result.

[Dead Man Flying]

I'd definitely be more worried about misleading or poorly-worded questions in surveys than about whether the survey sample represents the general population.  If questions intentionally or unintentionally bias the findings, then no amount of representativeness validates the survey results.

For example... ask anyone to rank Clinton as a person on a scale of 1 to 10.  Next, ask someone else to rank Clinton on a scale of 1 to 10 after being asked to rank Hitler on a scale of 1 to 10.  I think Clinton would rank significantly higher in the latter survey than in the former.  I'd be curious if the survey in question here follows the same sort of path.

-- Todd/Leviathn

[nirvana]

Good you brought that up as a local teacher has been given paid leave for comparing Bush to Hitler.  Sorry, off topic

Anyway, there is no way to give an unbiased survey, we all have our own biases, we just might not notice them.  Still, I want to see the questions.

[lazs2]

I have no problem with the polling... it is probly accurate and is an incement of our PUBLIC school system and our government leaders..

I have allways wondered why the constitution and bill of rights is not carved in stone in the entrance of every federal building....  seems that they are trying to ignore em to me.   They control the schools and they control the federal buildings soooo.... who is to blame?

I would bet that there would have been much different results if they would have asked home schooled or christian schooled people.

Yet.... the ignorant continue to defend overpaid and worthless teachers and the public school system.... probly cause they are the products of it and unable to think.

lazs

[Airscrew]

Originally posted by Dead Man Flying
Why does it not "seem" accurate?

I dont see it possible this survey could be accurate because they're attempting to measure someone's knowledge of a subject.   I would compare this type of survey to trying to access the knowledge of a high school class.  
This is not an opinion survey which to me measures how someone "feels" about an issue.  

Lets say I have a class of 30 students.   Instead of quizing all of them, I randomly select 3.  If all 3 students fail the quiz does that mean the whole class failed?  the more students I include in the quiz the better picture I have.  I totally agree that trying to survey the entire population would be cost prohibitive but I still think the survey should have had a larger sample.  I would like to see the entire survey, questions and results and also how they determined their random sample.   Don't you have to make a set of rules or guidelines on how the survey will be administered? does that make it random?

I have handled surveys before, although not professionally done, we made up our own surveys in military hospital to gauge customer service and to see if changes were needed in the hours and services.   Surveys are difficult to write correctly.  The results of customer service surveys can be biased because usually the only people that turn in surveys are the ones that had a complaint or poor service

The fact is that, unlikely as it seems, a survey drawn from 1,000 respondents may be representative of the general population if done properly and scientifically.

You are probably right

[USHilDvl]

OOoo...Oooo..ooo.ooooo...I know! I know!

The Simpsons wrote the First Amendment!  

[Dead Man Flying]

Originally posted by Airscrew

I dont see it possible this survey could be accurate because they're attempting to measure someone's knowledge of a subject.

[/b]

That would seem to me the most straightforward thing about which to survey.  Apparently all they did was ask people to list things.  They either got them right or got them wrong.

Lets say I have a class of 30 students.   Instead of quizing all of them, I randomly select 3.  If all 3 students fail the quiz does that mean the whole class failed?  the more students I include in the quiz the better picture I have.

[/b]

A sample of 3 in any large population will fail to yield reliable results.  Even minor variations in score among those three lead to major fluctuations in your findings for the whole population.  Surveys have a minimum threshhold for reliability, and 3 out of 30 does not cut it.  In addition, in the example you gave it is not prohibitive to take a census of the scores in the class since you only have 30 students.  When you're talking 300,000,000 people, cost becomes a major consideration.  And while you'll never achieve the sort of precision you get from drawing a census of the total population in public opinion polling, the increase in the margin of error by polling a substantially lower number often outweighs the costs of obtaining far larger samples.  I suspect if you had tripled the number of people surveyed here, you would have obtained similar findings at triple the cost.  

Don't you have to make a set of rules or guidelines on how the survey will be administered? does that make it random?

[/b]

Any good survey organization or newspaper reporting the survey will explain their sampling method and provide a link to the survey questions.  If they do not do so, you may consider the survey suspect at best and totally biased and unscientific at worst.

The results of customer service surveys can be biased because usually the only people that turn in surveys are the ones that had a complaint or poor service

[/b]

That is a non-random, self-selected sample.  And truthfully, there is nothing wrong with that in that situation if you understand what it means.  If those who have a complaint are overwhelmingly those who respond to such a survey, then at least you have a good sense of why people with complaints have them, and they they feel strongly enough to respond to surveys about them.  It won't give you a good sense of how customers in general view your organization however.

-- Todd/Leviathn

[Terror]

Originally posted by Airscrew
I thought '42' was the question.

no no no.. 42 is the answer.  the question takes a bit more time...