5 ms·

What you're talking about is statistics. Statistics are not irrefutable facts. They're data points from a report, and they are often incredibly easy to manipula

What you're talking about is statistics. Statistics are not irrefutable facts. They're data points from a report, and they are often incredibly easy to manipulate depending on how the macro is assessed. Usually it's impossible to gather stats over large, complex, chaotic populations. Instead samples are taken and applied to the whole and interpolated in-between. And in that interpolation an incredible amount of manipulation and even pure laziness is possible. It's possible to misrepresent the error bars of your conclusion. It's possible to leave out important details. It's possible to be selective about your time frame. There are a myriad of ways to mess up or screw up statistics. The more chaotic the system, the more difficult it is.

ttoinou 2y ago
Fully agree. Statistics are not global irrefutable facts about society, it's literally just one or a group of person computing something random and claiming it is representing society as a whole, or a journalist saying he/she read that figure in a reputable source. Even from a mathematical point of view statistics are incredibly hard to manipulate, but even before that, reality cannot be really measured and put into numbers.
ball_of_lint 2y ago
If, in an argument, you want to go back to the data and do different or better statistics on it then by all means. I would _love_ to have a disagreement with someone that went in that direction and we could discuss the intricacies of how to interpret the information that we have. I have my own gripes about the statistics done by various groups, with changing the inflation calculation being a recent example of the bad side of this: https://www.nytimes.com/2022/05/24/technology/inflation-measure-cpi-accuracy.html https://www.nytimes.com/2022/05/24/technology/inflation-meas... However, I think the key point here still stands. Most disagreements (at least in my experience) are not reaching this level, and are instead diving towards anti-intellectualism and dismissing statistics and data interpretation wholesale.
- naasking 2y ago
  > Most disagreements (at least in my experience) are not reaching this level, and are instead diving towards anti-intellectualism and dismissing statistics and data interpretation wholesale. Not without reason though. Many statistics are pretty bad and very partisan. Unpacking this is not trivial, and who wants to invest that effort to settle a debate with one other person?
abtinf 2y ago
It’s much worse than that. Every single example mentioned by the GP isn’t just a statistical measure, they measure of a wildly political (as in, defined by humans in a deeply imprecise manner) issues: > - whether crime is up or down Which kinds of crimes? In which political boundaries? In which reporting period? Did definitions change? Is reporting down because of ineffective policing? Is reporting up because of effective policing? That statistical games played with crime stats are criminal. > - whether the earth is warming or not There is a reason the phrase “global warming” went out of fashion in preference of “climate change”. Warming up how much? Over what time period? With what error bounds? Assuming which runaway processes? In which areas? Due to which causes? What are the error bounds around the sign of the change? > - how many people live in poverty The government literally draws a line in the sand and declares anyone below a certain income level is living in poverty. Who set the level? Why did they set it there? What is the standard of living at that income level? In which areas? How long do people live in poverty? What, if anything, prevents them from moving upward? What is there effective standard of living after government programs and charitable giving is taken into account? > - what the rate of inflation is This is literally defined by bureaucrats at central banks. Inflation according to which index? How were the index components chosen? How are the index components weighted? Over what time period? In which areas? Even the concept of “inflation” is highly suspect and basically incoherent. - how much social security or healthcare costs Over what time period? How did the demographics change? How about inflation? Where did the cash flows go and how did they net out? Which purchasing regimes were in place? How did the programs change? What was the quality of the services?

ttoinou 2y ago

Fully agree. Statistics are not global irrefutable facts about society, it's literally just one or a group of person computing something random and claiming it is representing society as a whole, or a journalist saying he/she read that figure in a reputable source. Even from a mathematical point of view statistics are incredibly hard to manipulate, but even before that, reality cannot be really measured and put into numbers.

ball_of_lint 2y ago

If, in an argument, you want to go back to the data and do different or better statistics on it then by all means. I would _love_ to have a disagreement with someone that went in that direction and we could discuss the intricacies of how to interpret the information that we have. I have my own gripes about the statistics done by various groups, with changing the inflation calculation being a recent example of the bad side of this: https://www.nytimes.com/2022/05/24/technology/inflation-measure-cpi-accuracy.html https://www.nytimes.com/2022/05/24/technology/inflation-meas... However, I think the key point here still stands. Most disagreements (at least in my experience) are not reaching this level, and are instead diving towards anti-intellectualism and dismissing statistics and data interpretation wholesale.

naasking 2y ago
> Most disagreements (at least in my experience) are not reaching this level, and are instead diving towards anti-intellectualism and dismissing statistics and data interpretation wholesale. Not without reason though. Many statistics are pretty bad and very partisan. Unpacking this is not trivial, and who wants to invest that effort to settle a debate with one other person?

abtinf 2y ago

It’s much worse than that. Every single example mentioned by the GP isn’t just a statistical measure, they measure of a wildly political (as in, defined by humans in a deeply imprecise manner) issues: > - whether crime is up or down Which kinds of crimes? In which political boundaries? In which reporting period? Did definitions change? Is reporting down because of ineffective policing? Is reporting up because of effective policing? That statistical games played with crime stats are criminal. > - whether the earth is warming or not There is a reason the phrase “global warming” went out of fashion in preference of “climate change”. Warming up how much? Over what time period? With what error bounds? Assuming which runaway processes? In which areas? Due to which causes? What are the error bounds around the sign of the change? > - how many people live in poverty The government literally draws a line in the sand and declares anyone below a certain income level is living in poverty. Who set the level? Why did they set it there? What is the standard of living at that income level? In which areas? How long do people live in poverty? What, if anything, prevents them from moving upward? What is there effective standard of living after government programs and charitable giving is taken into account? > - what the rate of inflation is This is literally defined by bureaucrats at central banks. Inflation according to which index? How were the index components chosen? How are the index components weighted? Over what time period? In which areas? Even the concept of “inflation” is highly suspect and basically incoherent. - how much social security or healthcare costs Over what time period? How did the demographics change? How about inflation? Where did the cash flows go and how did they net out? Which purchasing regimes were in place? How did the programs change? What was the quality of the services?