• Welcome to OGBoards 10.0, keep in mind that we will be making LOTS of changes to smooth out the experience here and make it as close as possible functionally to the old software, but feel free to drop suggestions or requests in the Tech Support subforum!

CBS Flash Poll: 72% of Americans support Prez Trump’s immigration stance

Wakeforest22890 said:
With a population of 200 people and using a sample size of 3, your margin of error is over 50%. With a population of 300,000,000 and using a sample size of 584, your margin of error is somewhere around 5-6%.

The percentage of the total population you're sampling really doesn't matter.
Click to expand...

Well, that's just like, your opinion, man
 
Does it surprise anybody that B-Rad doesn't understand basic statistics?
 
giphy.gif
 
Yep playing the clown and changing the subject here is probably your best course of action.
 
So with a MOE of 50% (per Numbers' analysis above), the Tunnels' Left agrees that somewhere north of a double digit % of statewide elected officials in Blue states must wear (or have worn) blackface?
 
Colonel Angus said:
So with a MOE of 50% (per Numbers' analysis above), the Tunnels' Left agrees that somewhere north of a double digit % of statewide elected officials in Blue states must wear (or have worn) blackface?
Click to expand...
If those people were randomly selected then there would be a high probability that the answer to your question is yes. Basic statistics.. Learn it... Love it... Live it...
 
LOL. I've never seen someone undermine their own arguments as effectively as Brad. He just repeatedly brutalizes himself, it's incredible.
 
The irony is Brad makes a point - just not the one he thinks he’s making. If you select people not at random and who are not necessarily representative of the entire population in order to get a true sense of the total population, your final results will likely not reflect the population. Which is exactly why taking viewers of the SOTU (who opted in to watching) and their feelings on Donald does not reflect the population.

I imagine Brad would also be shocked to find out that randomly selecting people to poll typically yields different results than polls where people opt themselves in to the poll (ie taking an online poll).
 
Even if you select people totally at random, but end up being a skewed population, it makes the poll useless.
 
RJKarl said:
Even if you select people totally at random, but end up being a skewed population, it makes the poll useless.
Click to expand...

Yeah this is why margin of error is important. Also how the pollsters go about randomly sampling.
 
Wakeforest22890 said:
The irony is Brad makes a point - just not the one he thinks he’s making. If you select people not at random and who are not necessarily representative of the entire population in order to get a true sense of the total population, your final results will likely not reflect the population. Which is exactly why taking viewers of the SOTU (who opted in to watching) and their feelings on Donald does not reflect the population.

I imagine Brad would also be shocked to find out that randomly selecting people to poll typically yields different results than polls where people opt themselves in to the poll (ie taking an online poll).
Click to expand...

Yep. Usually the amount of attention we should pay to a poll is equivalent to the amount of effort the pollster put into it. “Flash Poll” is a big indication we should pay this little attention.
 
Wakeforest22890 said:
With a population of 200 people and using a sample size of 3, your margin of error is over 50%. With a population of 300,000,000 and using a sample size of 584, your margin of error is somewhere around 5-6%.

The percentage of the total population you're sampling really doesn't matter.
Click to expand...

Not to mention the results are more likely representative of the group if the sample is randomly selected.

I mean, if you're sampling the population of NC to see what might be the incidence of influenza infection and you choose your sample from those reported to have had the flu, well, that's not going to give you very useful information.



Edited to add, yeah I'm just posting as I read through. I see others have made this point.
 
Last edited:
Wakeforest22890 said:
Yeah this is why margin of error is important. Also how the pollsters go about randomly sampling.
Click to expand...

But that MOE is way out of whack when you have twice as many Republicans as Dems as this poll did. It also doesn't address who that 32% of alleged independents were.

Another hint is when poll after poll shows a 60% opposition to the wall (and even higher for family separation) and your poll shows 72% approve of it. Only the most willfully blind would try to justify the 72% as accurate.

You have to throw out the East German judge's score.
 
RJKarl said:
But that MOE is way out of whack when you have twice as many Republicans as Dems as this poll did. It also doesn't address who that 32% of alleged independents were.

Another hint is when poll after poll shows a 60% opposition to the wall (and even higher for family separation) and your poll shows 72% approve of it. Only the most willfully blind would try to justify the 72% as accurate.

You have to throw out the East German judge's score.
Click to expand...
No the margin of error accurately reflects the people who watched the State of the Union. That just doesn't reflect the population as a whole because of people who watched the State of the Union skewed Republican.
 
Right.

If they did a good job of randomly sampling the population to determine the make up of those that watched then it's fairly accurate probably, even if the results (those who watched) aren't = to the makeup of the population as a whole (those who did and didn't watch).

That's part of the point of doing a survey/poll like this.
 
OTOH, those "polls" that are just links on a website that visitors can respond to are NOT legitimate samples or polls. Except to tell you something about responses of visitors to the website. Or bots etc.
 
I may have mistakenly compared Brad to RJ. He may actually be more comparable to theReff.

I always assumed you were smart Brad. An asshole with basically no human compassion, but smart. Try and prove that assumption correct.
 
You must log in or register to reply here.
Back
Top