4 ms·

I don't know... there are such things as facts. In 2014, "Vulnerable Populations" made up 80% of the poor in the USA. That essentially means children and the e

I don't know... there are such things as facts. In 2014, "Vulnerable Populations" made up 80% of the poor in the USA. That essentially means children and the elderly. On top of that, it's like another 7% of people that are in poverty that are fully employed! http://www.demos.org/blog/10/23/15/2014-vulnerable-populations-made-80-poor http://www.demos.org/blog/10/23/15/2014-vulnerable-populatio...

haberman 9y ago
Nov 7, 2016 "fact": Hillary Clinton has a 98.2% chance of winning the presidency: http://www.huffingtonpost.com/entry/polls-hillary-clinton-win_us_5821074ce4b0e80b02cc2a94 http://www.huffingtonpost.com/entry/polls-hillary-clinton-wi... There are facts, but it does them no service to overstate what the evidence actually says.
- Al-Khwarizmi 9y ago
  You know that events with probability less than 1.8% are happening all the time, right? The probability of dying in a car crash is way less than 1% and still some people are unlucky enough to die in this way every day. It's a matter of sample size. In the same way, the vast majority of the time a candidate with a very slim chance won't make it to presidency, but it happened that Trump did.
- ubernostrum 9y ago
  It was factually true that their method of weighting and averaging polling data produced that result.
- mark_edward 9y ago
  This is an excellent non sequitur. What does that have to do with most of the poor being elderly or children? Are you saying they aren't?
  - haberman 9y ago
    Grandparent's statement "there are such things as facts" seems to support the idea that the evidence provides unambiguous support for the author's thesis. I think people discussing complex phenomena (like poverty) tend to overstate the power of their evidence; they overstate its ability to support the conclusions they want to draw from it. This was illustrated in dramatic fashion when many/most of the most educated, "enlightened" people in the US had a reality check that their model was off last November. If there hadn't been an actual vote, people would probably still be insisting that Hillary Clinton would have won, had there been an actual vote on Nov 8. And that was the simple question of "how many people will show up to a polling station and mark a box for HC vs DT"! And the evidence consisted of pollsters who were asking hundreds or thousands of people that precise question every day! That should have been much easier to model and predict than poverty! It didn't even get into the causes of why people think or vote in certain ways. And yet still we have people looking at complex systems and problems like poverty and saying "it's not because of X, it's because of Y" with a feeling of certainty. And thinking that they are justified in their certainty because they claim this is what the "facts" say. I think, first of all, that grandparent hasn't even accurately summarized their own link, which says "Children and eldery make up slightly more than half of the market poor." Their statement that "essentially" 80% of the poor are children and elderly has already caused a significant amount of the underlying data to be "lost in translation". I think, second of all, that bucketing elderly, children, the disabled, etc into "vulnerable populations" and saying they comprise 80% of the poor doesn't really support the author's thesis at all. No one blames a child for being poor. It is not a child's responsibility to support themselves. So saying children are poor doesn't really speak to whether poverty is a personal choice. Children are the responsibility of their parents, so it is the parents choices that would be relevant here. Elderly people, on the other hand, generally spend their lives realizing they will be old someday, and therefore poverty of an elderly person does not necessarily say anything about whether poverty is a personal choice. It could be a lack of effort put into saving during their working years. I want to be clear that I'm not against safety nets, or in favor of doling out blame and judgement towards poor people. What I am against is the (to me) crazy amount of overconfidence many people (on the left especially) have in the ability of their data to "prove" their political positions and arguments. And I'm not even against people's right to hold convictions that the evidence doesn't fully justify. What I am especially against is when people use those convictions to pass blame and judgment on others. The clear subtext of this article is that we are all guilty, as a society, for making poor people poor.
- rndmize 9y ago
  What part of this is a fact? I'd argue this is literally the opposite of a fact - it's a model and a prediction based on it. Those are things that may be based on facts, but that in no way makes the theorized conclusion a fact. https://xkcd.com/605/ https://xkcd.com/605/
  - haberman 9y ago
    Nearly every statistical fact anyone ever cites does exactly the same thing. When a fact says something like "13% of people are poor," they don't investigate every single person and determine whether they are poor. They create what is intended to be a representative sample, then they create a model that extrapolates the sample set to the entire population. They predict that their sample is representative and that a measure of the whole population would match the sample. The only difference here is that there's no "poverty election day" where the entire population goes to register their poverty level. So we never actually get to see if the prediction was correct!
    - metasean 9y ago
      There's a day every decade where the U.S. Census does their best to get an accurate count of every human in the U.S. [0] Combine that with the data from the correspondinc Apr 15th, and we should have something pretty close to a "poverty election day". [0]: cell phone and spotty network made it too hard to look up a good source - please feel free to add them if you can)
- Dylan16807 9y ago
  Sure, if you want to ignore that 538 put the odds as a toss-up. But the types of statistics here are fundamentally different. If you look at the huffington post numbers as how many people will vote for a candidate, then even their prediction is pretty close to correct. A few million votes make the difference between 50% and 99%, but a few million votes is a reasonably tight accuracy range for "how many". Similarly I would expect something like the 80% number to be pretty close to correct, at least based on the definitions chosen.
  - haberman 9y ago
    I don't especially doubt the 80% statistic. I do think it's far overstating the evidence to say that it proves the author's thesis. I think it is far harder to prove the causes of poverty than it is to prove the incidence of poverty. I probably could have been more clear about this. The fault I find with the HuffPo prediction is overconfidence and under-skepticism about the ability of their data to say what they claimed it did.
- function_seven 9y ago
  Different pundits had different odds on that race. No matter which one you go with, what makes you think those odds weren’t true facts? Remember, n=1 in this scenario.
  - haberman 9y ago
    You can look at: 1. the specific circumstances of the outcome. Hillary Clinton didn't lose just barely, she lost by a lot. And she lost states like Wisconsin, which wasn't even supposed to be in play. 2. the fact that predictions like HuffPo's were far off from more credible pundits like 538. 3. the fact that HuffPo so aggressively mocked anyone who took a more measured view. http://www.huffingtonpost.com/entry/whats-wrong-with-538_us_581ffe18e4b0334571e09e74 http://www.huffingtonpost.com/entry/whats-wrong-with-538_us_...
    - function_seven 9y ago
      Drilling down to the state level is a good way to really trash those "predictions" from HuffPo. 538 really did get it right. They constantly noted the high correlation among related states, saying that if the polls are wrong in MI, they'd probably be wrong by the same amount in WI and PA. And that's exactly what happened.

haberman 9y ago

Nov 7, 2016 "fact": Hillary Clinton has a 98.2% chance of winning the presidency: http://www.huffingtonpost.com/entry/polls-hillary-clinton-win_us_5821074ce4b0e80b02cc2a94 http://www.huffingtonpost.com/entry/polls-hillary-clinton-wi... There are facts, but it does them no service to overstate what the evidence actually says.

Al-Khwarizmi 9y ago
You know that events with probability less than 1.8% are happening all the time, right? The probability of dying in a car crash is way less than 1% and still some people are unlucky enough to die in this way every day. It's a matter of sample size. In the same way, the vast majority of the time a candidate with a very slim chance won't make it to presidency, but it happened that Trump did.
ubernostrum 9y ago
It was factually true that their method of weighting and averaging polling data produced that result.
mark_edward 9y ago
This is an excellent non sequitur. What does that have to do with most of the poor being elderly or children? Are you saying they aren't?
- haberman 9y ago
  Grandparent's statement "there are such things as facts" seems to support the idea that the evidence provides unambiguous support for the author's thesis. I think people discussing complex phenomena (like poverty) tend to overstate the power of their evidence; they overstate its ability to support the conclusions they want to draw from it. This was illustrated in dramatic fashion when many/most of the most educated, "enlightened" people in the US had a reality check that their model was off last November. If there hadn't been an actual vote, people would probably still be insisting that Hillary Clinton would have won, had there been an actual vote on Nov 8. And that was the simple question of "how many people will show up to a polling station and mark a box for HC vs DT"! And the evidence consisted of pollsters who were asking hundreds or thousands of people that precise question every day! That should have been much easier to model and predict than poverty! It didn't even get into the causes of why people think or vote in certain ways. And yet still we have people looking at complex systems and problems like poverty and saying "it's not because of X, it's because of Y" with a feeling of certainty. And thinking that they are justified in their certainty because they claim this is what the "facts" say. I think, first of all, that grandparent hasn't even accurately summarized their own link, which says "Children and eldery make up slightly more than half of the market poor." Their statement that "essentially" 80% of the poor are children and elderly has already caused a significant amount of the underlying data to be "lost in translation". I think, second of all, that bucketing elderly, children, the disabled, etc into "vulnerable populations" and saying they comprise 80% of the poor doesn't really support the author's thesis at all. No one blames a child for being poor. It is not a child's responsibility to support themselves. So saying children are poor doesn't really speak to whether poverty is a personal choice. Children are the responsibility of their parents, so it is the parents choices that would be relevant here. Elderly people, on the other hand, generally spend their lives realizing they will be old someday, and therefore poverty of an elderly person does not necessarily say anything about whether poverty is a personal choice. It could be a lack of effort put into saving during their working years. I want to be clear that I'm not against safety nets, or in favor of doling out blame and judgement towards poor people. What I am against is the (to me) crazy amount of overconfidence many people (on the left especially) have in the ability of their data to "prove" their political positions and arguments. And I'm not even against people's right to hold convictions that the evidence doesn't fully justify. What I am especially against is when people use those convictions to pass blame and judgment on others. The clear subtext of this article is that we are all guilty, as a society, for making poor people poor.
rndmize 9y ago
What part of this is a fact? I'd argue this is literally the opposite of a fact - it's a model and a prediction based on it. Those are things that may be based on facts, but that in no way makes the theorized conclusion a fact. https://xkcd.com/605/ https://xkcd.com/605/
- haberman 9y ago
  Nearly every statistical fact anyone ever cites does exactly the same thing. When a fact says something like "13% of people are poor," they don't investigate every single person and determine whether they are poor. They create what is intended to be a representative sample, then they create a model that extrapolates the sample set to the entire population. They predict that their sample is representative and that a measure of the whole population would match the sample. The only difference here is that there's no "poverty election day" where the entire population goes to register their poverty level. So we never actually get to see if the prediction was correct!
  - metasean 9y ago
    There's a day every decade where the U.S. Census does their best to get an accurate count of every human in the U.S. [0] Combine that with the data from the correspondinc Apr 15th, and we should have something pretty close to a "poverty election day". [0]: cell phone and spotty network made it too hard to look up a good source - please feel free to add them if you can)
Dylan16807 9y ago
Sure, if you want to ignore that 538 put the odds as a toss-up. But the types of statistics here are fundamentally different. If you look at the huffington post numbers as how many people will vote for a candidate, then even their prediction is pretty close to correct. A few million votes make the difference between 50% and 99%, but a few million votes is a reasonably tight accuracy range for "how many". Similarly I would expect something like the 80% number to be pretty close to correct, at least based on the definitions chosen.
- haberman 9y ago
  I don't especially doubt the 80% statistic. I do think it's far overstating the evidence to say that it proves the author's thesis. I think it is far harder to prove the causes of poverty than it is to prove the incidence of poverty. I probably could have been more clear about this. The fault I find with the HuffPo prediction is overconfidence and under-skepticism about the ability of their data to say what they claimed it did.
function_seven 9y ago
Different pundits had different odds on that race. No matter which one you go with, what makes you think those odds weren’t true facts? Remember, n=1 in this scenario.
- haberman 9y ago
  You can look at: 1. the specific circumstances of the outcome. Hillary Clinton didn't lose just barely, she lost by a lot. And she lost states like Wisconsin, which wasn't even supposed to be in play. 2. the fact that predictions like HuffPo's were far off from more credible pundits like 538. 3. the fact that HuffPo so aggressively mocked anyone who took a more measured view. http://www.huffingtonpost.com/entry/whats-wrong-with-538_us_581ffe18e4b0334571e09e74 http://www.huffingtonpost.com/entry/whats-wrong-with-538_us_...
  - function_seven 9y ago
    Drilling down to the state level is a good way to really trash those "predictions" from HuffPo. 538 really did get it right. They constantly noted the high correlation among related states, saying that if the polls are wrong in MI, they'd probably be wrong by the same amount in WI and PA. And that's exactly what happened.