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.

 

Reuters Poll: Black Male Approval For Trump Doubles In One Week

page: 4
40
<< 1  2  3    5  6 >>

log in

join
share:

posted on May, 3 2018 @ 02:05 PM
link   

originally posted by: dug88

originally posted by: RadioRobert
a reply to: dug88

Using confidence level of 95%, I find a confidence interval of 7.5 for that sample sjze. What did you get?


Well it's been a while since i've done statistics...but doesn't seem like you can really say anything at all either way with a confidence interval of 7.5 at 95%.

In order to even begin making the claims in the OP this poll would have needed at least 9600 participants...and that's just to get a confidence interval of 1 at 95% confidence.

Based on that poll you'd have a better chance at flipping a coin and trying to guess.



Are e believing polls again or are they still false?

I can't keep track.

Are celebrities in politics lame or cool..

Again I keep losing track.

If it's kanye it's legit. If it Beyonce it's not?

I need a memo.




posted on May, 3 2018 @ 02:23 PM
link   
It is actually a crazy jump. It will be interesting to see what it will be next week when the press quote-mines another soundbite to make him sound racist. I guess we'll see.



posted on May, 3 2018 @ 02:30 PM
link   
a reply to: LesMisanthrope

Ben Carson was racist.

So will Kanye be.




posted on May, 3 2018 @ 02:33 PM
link   
Trump disapproval rating 80% = trumptards declare victory..

Jesus f christ



posted on May, 3 2018 @ 02:35 PM
link   

originally posted by: dug88

originally posted by: RadioRobert
a reply to: dug88

Using confidence level of 95%, I find a confidence interval of 7.5 for that sample sjze. What did you get?


Well it's been a while since i've done statistics...but doesn't seem like you can really say anything at all either way with a confidence interval of 7.5 at 95%.

In order to even begin making the claims in the OP this poll would have needed at least 9600 participants...and that's just to get a confidence interval of 1 at 95% confidence.

Based on that poll you'd have a better chance at flipping a coin and trying to guess.



+/- 7.5 is the same as flipping a coin? Do you know many polls running a plus/minus 1?



posted on May, 3 2018 @ 03:06 PM
link   
There’s nothing scarier to a leftist than a free thinker...


Kanye Tweeted out the one thing Obama doesn’t want to hear… That his home city of Chicago isn’t falling for Obama’s empty words anymore. That Obama was in charge for 8 years and NOTHING changed on the violent streets of Chicago.

“ reached out to Obama for years and couldn’t get anything done, but Trump gave me a meeting,” Kanye said.

Liberals became more furious when Kanye proudly Tweeted out a picture of his autographed “Make America Great Again” hat, which President Trump retweeted!

After Kanye Announced Support For Trump, He Leaked Obama’s Worst Nightmare





posted on May, 3 2018 @ 03:19 PM
link   
a reply to: ErrorErrorError

Nice try George...



Fake News: AP Touts Anti-Trump Poll with Warped Demographics Conducted by Soros-Backed Org

The Associated Press on Monday released a news making story with a glaring headline claiming that “Most young Americans don’t see Trump as a legitimate leader.”

A closer look at the survey finds it was conducted by a group financed by billionaire George Soros whose activist arm demands reparations for slavery and “mass incarceration,” and has engaged in anti-police activism.

In addition, the survey was not representative of the racial and ethnic profiles of young adults. Instead it focused heavily on demographic samples of populations that voted overwhelmingly against Donald Trump.

Fake News: AP Touts Anti-Trump Poll with Warped Demographics Conducted by Soros-Backed Org



posted on May, 3 2018 @ 03:37 PM
link   

originally posted by: Xtrozero

originally posted by: dug88

That's a pretty statistically significant sample size you got there bud.

Polls are fairly useless ways to get information period.



You put a lot of energy into your post without really knowing statistics, but you are correct that 177 samples IS significant.


177/4200000 == 0.000004214285714285% of the target population.

In what way has it ever been possible to make broad sweeping claims based on a sample size of
0.000004214285714285% of a population?

I understand statistics..I mean i've only taken several statistics classes and used it every day for three years as a wildlife researcher and all...it's really easy to make any set of data look like what you want it to. But I do know as far as scientific sampling is concerned...trying to publish something like that would get you laughed at.
edit on 3/5/2018 by dug88 because: (no reason given)

edit on 3/5/2018 by dug88 because: (no reason given)



posted on May, 3 2018 @ 03:50 PM
link   

originally posted by: dug88

originally posted by: Xtrozero

originally posted by: dug88

That's a pretty statistically significant sample size you got there bud.

Polls are fairly useless ways to get information period.



You put a lot of energy into your post without really knowing statistics, but you are correct that 177 samples IS significant.


177/4200000 == 0.000004214285714285% of the target population.

In what way has it ever been possible to make broad sweeping claims based on a sample size of
0.000004214285714285% of a population?

I understand statistics..I mean i've only taken several statistics classes and used it every day for three years as a wildlife researcher and all...it's really easy to make any set of data look like what you want it to. But I do know as far as scientific sampling is concerned...trying to publish something like that would get you laughed at.



The Dems cannot put the Kanye genie back in the bottle no matter how hard anyone tries.



posted on May, 3 2018 @ 03:58 PM
link   
You would think we could have declared war on him. I mean with congressional approval even. Send the marines at him. Seize his accounts. Why is this guy allowed to devastate and destroy peoples lives and economies? Sheesh man, Assange had to hide from windows because hillary wanted to drone him. Can we vote to drone this guy? Is a strategic nuke considered acceptable?
a reply to: Murgatroid



posted on May, 3 2018 @ 04:02 PM
link   
a reply to: UKTruth

Looks like a of those many spikes that occur during the marathon of a Presidential term. We shall see it this turns into a trend instead of a 'spike'. Time will tell.



posted on May, 3 2018 @ 04:17 PM
link   

originally posted by: dug88


177/4200000 == 0.000004214285714285% of the target population.

In what way has it ever been possible to make broad sweeping claims based on a sample size of
0.000004214285714285% of a population?

I understand statistics..I mean i've only taken several statistics classes and used it every day for three years as a wildlife researcher and all...it's really easy to make any set of data look like what you want it to. But I do know as far as scientific sampling is concerned...trying to publish something like that would get you laughed at.



Your answer here tells me you know very little about statistics. The size of the pool means little, its the sample size that counts, so to say out of 20,000, 20 million or 200 million doesn't really matter.

I'm not an expert on this, but a sample size less than 30 is considered too small. To help you understands better I'll give you real numbers below based on a 95% confidence level for your 42 million.

Population of 420 would mean you need a sample size of 201 to get that 95% confidence level
4200 needs 353
42000 needs 381
420000 needs 384
4200000 needs 385
42000000 needs 385
4200000000000000000000000 needs 385


Even with a pool of 4200 you are now reaching the point of all possibilities, so after that all pool sizes have about the same number of possibilities no matter the size.

177 out of 42 million is about 82% confidence level and I do agree they should have done more sampling to get a 95%.

Get the picture..



edit on 3-5-2018 by Xtrozero because: (no reason given)



posted on May, 3 2018 @ 04:24 PM
link   
a reply to: Xtrozero

I took Statistics in college.

Was the only class that actually made my eyes bleed.



posted on May, 3 2018 @ 04:26 PM
link   

originally posted by: DBCowboy
a reply to: Xtrozero

I took Statistics in college.

Was the only class that actually made my eyes bleed.



Did my post make them start bleeding again...lol

eye test...

x = Z(c/100)2r(100-r)
n = N x/((N-1)E2 + x)
E = Sqrt[(N - n)x/n(N-1)]

edit on 3-5-2018 by Xtrozero because: (no reason given)



posted on May, 3 2018 @ 04:42 PM
link   
a reply to: TinfoilTP

Since polls are about as trustworthy as Hillary Clinton, I'll take this result with a grain of salt. I see that the usual suspects have already been here to determine that the poll is fake because 'conservative' - these same folks believed a poll that said Hillary had a 95% chance of winning the election - thats how stupid and gullible they are.

But now, as predictable as it is, now the polls are fake because it paints Trump in a positive light.

LOL, pathetic low informationers are pathetic.

As long as Trump keeps winning and folks keep calling out the tards in the democratic party and MSM and FBI and DOJ, then things really aren't that bad, in fact, one would think that the tards would be sick of being emotionally triggered on a daily basis, but alas, just like crack-whore junkies, they too are addicted to their daily (even multiple times per per) spoon-fed outrage instructions from MSM.



posted on May, 3 2018 @ 04:51 PM
link   

originally posted by: Xtrozero

originally posted by: dug88


177/4200000 == 0.000004214285714285% of the target population.

In what way has it ever been possible to make broad sweeping claims based on a sample size of
0.000004214285714285% of a population?

I understand statistics..I mean i've only taken several statistics classes and used it every day for three years as a wildlife researcher and all...it's really easy to make any set of data look like what you want it to. But I do know as far as scientific sampling is concerned...trying to publish something like that would get you laughed at.


So your answer here tells me you know very little about statistics. The size of the pool means little, its the sample size that counts. So to say out of 20,000, 20 million or 200 million doesn't really matter.

I'm not an expert on this but a sample size less than 30 is considered too small, so below I'll give you real numbers based on a 95% confidence level to help you understand.

Population of 420 would mean you need a sample size of 201 to get that 95% confidence level
4200 needs 353
42000 needs 381
420000 needs 384
4200000 needs 385
42000000 needs 385
4200000000000000000000000 needs 385


Even with a pool of 4200 you are now reaching the point of all possibilities, so after that all pool sizes have about the same number of possibilities no matter the size.

177 out of 42 million is about 82% confidence level and I do agree they should have done more sampling to get a 95%.

Get the picture..




Ya that's my point 82% confidence level is not an acceptable level to be making claims about an entire population.

Would you take a drug that was tested and approved with that level of confidence? Say for effectiveness and I dunno risk of death?
edit on 3/5/2018 by dug88 because: (no reason given)



posted on May, 3 2018 @ 04:52 PM
link   

originally posted by: Sublimecraft
a reply to: TinfoilTP

Since polls are about as trustworthy as Hillary Clinton, I'll take this result with a grain of salt. I see that the usual suspects have already been here to determine that the poll is fake because 'conservative' - these same folks believed a poll that said Hillary had a 95% chance of winning the election - thats how stupid and gullible they are.




The problematic thing with polls is having the correct sample size and polling correctly. In this case I pointed out that sample size of 101 and 177 is not really that good to prove a point, it should have been 385 for a 95% confidence level and if they really wanted to get wild they could have gone for a 99% confidence level with a 664 sample. So that is one concern on how accurate it is.

The other part is they need to do a proper sample spread based on the demographics of the whole pool. In this case they would need to sample black men across all of America, AND they should show that sampling in their results. To say Hillary is 95% likely to win, and she lost her ass to everyone's surprise, we never saw the sample statistics. Did they only sample from blue states to get their 95%? who knows maybe only LA...lol

Polling can be extremely hard with a huge amount of variables involved. One such variable is the time of polling. If they polled black men across America but only from 0800 to 1600 then they most likely skewed their results with black men who do not have a job.



edit on 3-5-2018 by Xtrozero because: (no reason given)



posted on May, 3 2018 @ 04:54 PM
link   
I've noticed a lot of black men and women returning to their conservative roots

what the left doesn't want people to know

Booker T Washington
Martin Luther King
Fredrick Douglas
Thomas Sowell

Black Americans who have been a part of civil rights movements, education advancement and others who have played key rolls in the civil rights struggle in our history.

Christian conservatives.



posted on May, 3 2018 @ 04:59 PM
link   

originally posted by: dug88



Ya that's my point 82% confidence level is not an acceptable level to be making claims about an entire population.

Would you take a drug that was tested and approved with that level of confidence? Say for effectiveness and I dunno risk of death?


Well that wasn't your point until I pointed that out, but I will accept it anyways..

BTW so 177 sample is not acceptable, but would 385 be OK with you for either this or your drug test?
edit on 3-5-2018 by Xtrozero because: (no reason given)



posted on May, 3 2018 @ 05:50 PM
link   
Kanye is tweeting Thomas Sowell quotes?
Well that would do it...




top topics



 
40
<< 1  2  3    5  6 >>

log in

join