• Quick note - the problem with Youtube videos not embedding on the forum appears to have been fixed, thanks to ZiprHead. If you do still see problems let me know.

Let's make America smart again

wareyin said:
Still pretending that we need pinpoint accuracy to tell the difference between a teacup and a mountain.
Click to expand...

No, we don't need pinpoint accuracy. But we need accuracy sufficient for the claimed purpose. Your hypothetical scenario has accuracy sufficient for the purpose. Does Politifact's sampling have sufficient accuracy? How would you even know if you can't determine what that accuracy is? And if you CAN determine what that accuracy is, why don't you?

If the accuracy is good enough, then where's your argument to demonstrate that? You have none. You simply declare it to be so.
 
Argumemnon said:
Yeah, but since we can't break open his head and show his bias to the rest of the forum, do you think you could address the actual arguments he's made? I mean, questioning other posters' motives is something we do regularily, but I think you've made your point about it, already.

I understand your frustration, but we're not getting anywhere with this.
Click to expand...

His argument appears to be that if we pretend that Politifact's numbers might be inaccurate, we can ignore them. My response to that argument has been that the difference between the candidates telling of lies has been so great that quibbling about exact accuracy is pointless.
 
wareyin said:
His argument appears to be that if we pretend that Politifact's numbers might be inaccurate, we can ignore them. My response to that argument has been that the difference between the candidates telling of lies has been so great that quibbling about exact accuracy is pointless.
Click to expand...

No, what he's saying is that the sampling is not controlled and therefore politifact could have deliberately biased it in favour of Clinton or against Trump.
 
Argumemnon said:
No, what he's saying is that the sampling is not controlled and therefore politifact could have deliberately biased it in favour of Clinton or against Trump.
Click to expand...

The biasing doesn't have to be deliberate. Accidental biasing can happen too. That's part of why double blind tests are the gold standard in medical research: people can fool themselves.
 
wareyin said:
His argument appears to be that if we pretend that Politifact's numbers might be inaccurate, we can ignore them.
Click to expand...

I don't have to pretend this. It's obvious. If, that is, you understand anything about statistics.

My response to that argument has been that the difference between the candidates telling of lies has been so great that quibbling about exact accuracy is pointless.
Click to expand...

You can't even tell me what the ballpark accuracy is. You can't tell me anything about its accuracy.
 
Ziggurat said:
I don't have to pretend this. It's obvious. If, that is, you understand anything about statistics.



You can't even tell me what the ballpark accuracy is. You can't tell me anything about its accuracy.
Click to expand...

I did not have to perform any calculations about accuracy to see that a mountain is larger than a teacup. I similarly don't need to them to compare the mountain of Trump's lies to the teacup-full of Clinton's.
 
wareyin said:
I did not have to perform any calculations about accuracy to see that a mountain is larger than a teacup. I similarly don't need to them to compare the mountain of Trump's lies to the teacup-full of Clinton's.
Click to expand...

If you don't need calculations to compare them, then why are you trying to defend calculations which compare them? I already gave you an out which doesn't require calculations of any kind, and permits you to draw conclusions in favor of whomever you want to.
 
Ziggurat said:
If you don't need calculations to compare them, then why are you trying to defend calculations which compare them? I already gave you an out which doesn't require calculations of any kind, and permits you to draw conclusions in favor of whomever you want to.
Click to expand...

What mountain? The mountain Politifact produced without any sampling controls?

What teacup? The teacup Politifact produced without any sampling controls?

What data, exactly, are you looking at, to reach your conclusion?
 
theprestige said:
What mountain? The mountain Politifact produced without any sampling controls?

What teacup? The teacup Politifact produced without any sampling controls?

What data, exactly, are you looking at, to reach your conclusion?
Click to expand...

The mountain that conservatives are determined to ignore.

The teacup that conservatives are determined to pretend is larger than the mountain.

The data I look at to determine that Trump lies more than Clinton are his lies vs her lies.
 
Ziggurat said:
If you don't need calculations to compare them, then why are you trying to defend calculations which compare them? I already gave you an out which doesn't require calculations of any kind, and permits you to draw conclusions in favor of whomever you want to.
Click to expand...
If you understand that one can make a comparison between the 2 without calculations, why have you been demanding calculations?
 
wareyin said:
If you understand that one can make a comparison between the 2 without calculations, why have you been demanding calculations?
Click to expand...

Calculations were offered to try to support a claim, I didn't demand them. I just demand that if you want to use calculations, you do it right.
 
Argumemnon said:
If you don't making any sort of calculation, how would you arrive at an objective conclusion?
Click to expand...
I keep going back to the teacup/mountain example, because I seriously cannot understand how people are unable to look at 2 things, compare them to each other, and conclude which is larger without performing any actual measurements or calculations.
 
wareyin said:
I keep going back to the teacup/mountain example, because I seriously cannot understand how people are unable to look at 2 things, compare them to each other, and conclude which is larger without performing any actual measurements or calculations.
Click to expand...

You're assuming your conclusion, wareyin. Here's the thing: you have a number of statements by two people, and want to determine who's the biggest liar. You can't assume that one's the mountain and the other the teacup without first checking a large, well-controlled sample of these statements. Otherwise you're even more open to bias.
 
Nate Silver looks at education levels in county level election data.

I took a list of all 981 U.S. counties with 50,000 or more people and sorted it by the share of the population3 that had completed at least a four-year college degree. Hillary Clinton improved on President Obama’s 2012 performance in 48 of the country’s 50 most-well-educated counties. And on average, she improved on Obama’s margin of victory in these countries by almost 9 percentage points, even though Obama had done pretty well in them to begin with.
Click to expand...

At the other end of the scale, in the counties with the lowest percentage of college graduates:

Clinton lost ground relative to Obama in 47 of the 50 counties — she did an average of 11 percentage points worse, in fact. These are really the places that won Donald Trump the presidency, especially given that a fair number of them are in swing states such as Ohio and North Carolina.
Click to expand...


Education, Not Income, Predicted Who Would Vote For Trump
 
Argumemnon said:
You're assuming your conclusion, wareyin. Here's the thing: you have a number of statements by two people, and want to determine who's the biggest liar. You can't assume that one's the mountain and the other the teacup without first checking a large, well-controlled sample of these statements. Otherwise you're even more open to bias.
Click to expand...

Except you don't want a completely random sample of statements for this calculation. The 'biggest liar' is not simply who has the most lies, but the magnitude of the lies is important as well. Someone who tells large numbers of small utilitarian lies such as 'I'm busy' rather than 'no I don't want to come come your son's birthday' is in no way a bigger liar than one who says they have proof of treason when they don't, at least in no way that's meaningful. There is no good way to remove some trace of value judgement from this kind of analysis.

This is one reason the Politifact numbers are more than accurate enough. They rate things that they are asked to rate, and things their researchers find important. The only real way to cast serious doubt on their numbers for this use would be to find many important statements that weren't rated for either Trump or Clinton (or both).

Trumps lies run the gamut from petty to ridiculous to major. He's literally a fraudster. He lies consistently about things he's on video saying from not long before he lies about saying them! It's not as if it's about misremembering what was said or the exact wording. The importance of these ranges from trivial to major.

It's as accurate as this type of analysis can be and more than it needs to be. Trumps the bigger liar between himself and Clinton.
 
tyr_13 said:
Except you don't want a completely random sample of statements for this calculation.
Click to expand...

I didn't say "random" anywhere in my post. In fact we've already discussed taking statements from interviews and debates only.

The 'biggest liar' is not simply who has the most lies, but the magnitude of the lies is important as well.
Click to expand...

Sure, but it's hard to judge magnitude without a huge part of subjectivity. Better to take a large number of statements from the aforementioned places and hope that on average the "magnitude" is equivalent.

Trumps lies run the gamut from petty to ridiculous to major. He's literally a fraudster. He lies consistently about things he's on video saying from not long before he lies about saying them! It's not as if it's about misremembering what was said or the exact wording. The importance of these ranges from trivial to major.
Click to expand...

I know but if we want to demonstrate it objectively we need more than our impressions.
 
You must log in or register to reply here.

Back
Top Bottom