5 ms·

Some say it was a decline in environmental lead, as the linked article does. Some say it was the advent of mandatory sentencing and three-strikes laws. Some say

by lutusp 13y ago

Some say it was a decline in environmental lead, as the linked article does. Some say it was the advent of mandatory sentencing and three-strikes laws. Some say it was the easy availability of birth control and abortion, meaning fewer unwanted children were born. But, social psychologists to the contrary, we will never know, because this kind of study can never ascend to the level of science. All we can be sure of is a rise in ... not crime, but contentless articles whose titles end in a question mark.

lclarkmichalek 13y ago
Antipositivism is a thing, like it or not. I've aways loved the quote from Heisenberg: > The positivists have a simple solution: the world must be divided into that > which we can say clearly and the rest, which we had better pass over in > silence. But can any one conceive of a more pointless philosophy, seeing > that what we can say clearly amounts to next to nothing? If we > omitted all that is unclear we would probably be left with completely > uninteresting and trivial tautologies. Rejecting the entirety of a science because it deals in confidence values rather than "facts" is foolish; it might be better to judge works based on their quality, rather than some ranking that deems X to be a science and Y to be something other. I particularly love that you used the word "ascend"; I shan't be the first to compare positivism to scientism
- beagle3 13y ago
  "science", to some, implies a discipline subject to the scientific method - none of this discussion is. Replace "science" with "discipline" in your post and it will make sense.
  - lclarkmichalek 13y ago
    Nope. Still not seeing how "this kind of study can never ascend to the level of discipline (née science)" is something that can be stated without any supporting argument.
    - beagle3 13y ago
      I was referring to "science" in your description of "Rejecting the entirety of a science because it deals in confidence values". It does not give confidence values. It is not science in the "scientific method" sense. It is a descriptive discipline, yes, and it can be useful - but it could be a random correlation, or correlated through a hidden variable. An observational study is, scientifically, not a valid way to establish cause and effect.
    - lutusp 13y ago
      > Still not seeing how "this kind of study can never ascend to the level of discipline (née science)" is something that can be stated without any supporting argument. If you understood the basics of science, you would know what's missing from an opinion article that picks a quantity at random and discovers a correlation with something else, but without a testable theory about how they might have something to do with each other. Science requires something, that something can be expressed in s single word, and once that word is uttered, all who hear it will realize why this kind of article cannot represent science. And I am astonished by how few people know this requirement of scientific practice. Science isn't the answer to the world's problems, but for certain problems, it has no meaningful substitute.
      - qubitcoder 13y ago
        As a thought experiment, just imagine Max Planck, Werner Heisenberg, and Richard Feynman gathering their teams together in a room for a major announcement. All three scientists have reached a unanimous conclusion. Each time they try to measure some aspect of an atom, their results are inconsistent! They declare to to their teams that this cannot be real science. After all, we need reliable measurements in order to do science. So let’s not waste our time any longer. Go home, get some rest, and come back tomorrow. And please take a moment to review the equations governing the path of flying baseballs. We think this merits a second look. Fortunately history tells a different story. By the way, I’d highly recommend reading the outstanding book “Quantum Mechanics: The Theoretical Minimum” by Leonard Susskind of Stanford. If you’ve seen a matrix before, then you’re good to go (with some effort). He walks you through the rest. As a fair warning, try to mentally prepare yourself. Your seemingly rational & logical viewpoints, while admirable, will be shattered one by one. And you’ll see that even our precious boolean logic is merely an illusion. EDIT: Oops. Wrong level. This reply was geared to downandout in above thread. Apologies.
- lutusp 13y ago
  > Rejecting the entirety of a science because it deals in confidence values rather than "facts" is foolish ... I'm not rejecting pop psychology. I'm saying it's not science. Science requires something fundamental, it can be expressed in one word, and I invite the readers of this post to say what it is. As soon as the word leaves their mouths, they will realize why social psychology cannot be science. When the physicists at the LHC observed what they thought might be the Higgs Boson, they refused to announce a discovery until the had a five sigma result. Five sigma is approximately equal to 2.8 * 10^-7 (explanation: http://blogs.scientificamerican.com/observations/2012/07/17/five-sigmawhats-that/ http://blogs.scientificamerican.com/observations/2012/07/17/...), and it expresses the probability that, if there is no Higgs particle, a random observation might create the observation that has been made. Five sigma. That's science. Opinion articles whose titles end in question marks are not science. And I should not have to explain this to educated readers. > I particularly love that you used the word "ascend"; I shan't be the first to compare positivism to scientism Scientism is an irrational confidence in science as an answer to all the world's problems. Social psychology is an irrational confidence in the power of articles that observe correlations without either proposing or testing theories about those correlations to accurately represent some aspect of reality.
lambda 13y ago
What makes you say that this kind of study can never ascend to the level of science? This kind of study is science. It's imperfect, but all of science is; science is just a collection of tools to improve our knowledge of the world, coupled with the body of knowledge so gained. Some of those tools are stronger than others, or can provide more certainty. Controlled studies in which you vary only one input to a system at a time, and observe the effect, are obviously a stronger tool than mere observational studies. But that doesn't make observational studies not science. It's perfectly possible to determine which of these effects are more or less likely to be responsible; the different hypotheses that you make would make easily testable predictions, which can be tested independently because different jurisdictions have made these changes at different times. Lead was banned in different countries at vastly different times (https://en.wikipedia.org/wiki/Tetraethyllead https://en.wikipedia.org/wiki/Tetraethyllead). Likewise, three-strikes laws have been introduced at different times in different places, and many places have never introduced such laws (https://en.wikipedia.org/wiki/Three-strikes_law https://en.wikipedia.org/wiki/Three-strikes_law). Easy availability of birth control and abortion may be a little harder to quantify precisely, but that can be done too. Since some countries have only banned leaded fuel in the past 10 years, within another 10 or 15 we should be able to say with considerably more certainty whether this has a causal relationship. If the crime data for countries which banned it later follow the same pattern, that's a very strong indication that the theory is correct. I would find it hard to believe that the three strikes laws would have the same effect, as the reduction in crime has been happening fairly consistently nation-wide, while three strikes laws have been implemented at vastly different times in different states, or not at all. Lead levels are much more consistent nationwide, as those have mostly been national laws (with the exception of California banning leaded gasoline a few years earlier). Anyhow, I strongly disagree that you can say that this isn't science. It's imperfect science; anything as complex as the social phenomenon of global crime levels is going to have a lot of complex, interacting factors and no one of them is going to be completely explanatory, and you are never going to be able to do significant studies with controlled inputs. But there are scientific tools for dealing with these somewhat more ambiguous situations, and we can and should apply them as the policy implications are quite substantial.
- lutusp 13y ago

by lutusp 13y ago

lclarkmichalek 13y ago

Antipositivism is a thing, like it or not. I've aways loved the quote from Heisenberg: > The positivists have a simple solution: the world must be divided into that > which we can say clearly and the rest, which we had better pass over in > silence. But can any one conceive of a more pointless philosophy, seeing > that what we can say clearly amounts to next to nothing? If we > omitted all that is unclear we would probably be left with completely > uninteresting and trivial tautologies. Rejecting the entirety of a science because it deals in confidence values rather than "facts" is foolish; it might be better to judge works based on their quality, rather than some ranking that deems X to be a science and Y to be something other. I particularly love that you used the word "ascend"; I shan't be the first to compare positivism to scientism

beagle3 13y ago
"science", to some, implies a discipline subject to the scientific method - none of this discussion is. Replace "science" with "discipline" in your post and it will make sense.
- lclarkmichalek 13y ago
  Nope. Still not seeing how "this kind of study can never ascend to the level of discipline (née science)" is something that can be stated without any supporting argument.
  - beagle3 13y ago
    I was referring to "science" in your description of "Rejecting the entirety of a science because it deals in confidence values". It does not give confidence values. It is not science in the "scientific method" sense. It is a descriptive discipline, yes, and it can be useful - but it could be a random correlation, or correlated through a hidden variable. An observational study is, scientifically, not a valid way to establish cause and effect.
  - lutusp 13y ago
    > Still not seeing how "this kind of study can never ascend to the level of discipline (née science)" is something that can be stated without any supporting argument. If you understood the basics of science, you would know what's missing from an opinion article that picks a quantity at random and discovers a correlation with something else, but without a testable theory about how they might have something to do with each other. Science requires something, that something can be expressed in s single word, and once that word is uttered, all who hear it will realize why this kind of article cannot represent science. And I am astonished by how few people know this requirement of scientific practice. Science isn't the answer to the world's problems, but for certain problems, it has no meaningful substitute.
    - qubitcoder 13y ago
      As a thought experiment, just imagine Max Planck, Werner Heisenberg, and Richard Feynman gathering their teams together in a room for a major announcement. All three scientists have reached a unanimous conclusion. Each time they try to measure some aspect of an atom, their results are inconsistent! They declare to to their teams that this cannot be real science. After all, we need reliable measurements in order to do science. So let’s not waste our time any longer. Go home, get some rest, and come back tomorrow. And please take a moment to review the equations governing the path of flying baseballs. We think this merits a second look. Fortunately history tells a different story. By the way, I’d highly recommend reading the outstanding book “Quantum Mechanics: The Theoretical Minimum” by Leonard Susskind of Stanford. If you’ve seen a matrix before, then you’re good to go (with some effort). He walks you through the rest. As a fair warning, try to mentally prepare yourself. Your seemingly rational & logical viewpoints, while admirable, will be shattered one by one. And you’ll see that even our precious boolean logic is merely an illusion. EDIT: Oops. Wrong level. This reply was geared to downandout in above thread. Apologies.
lutusp 13y ago
> Rejecting the entirety of a science because it deals in confidence values rather than "facts" is foolish ... I'm not rejecting pop psychology. I'm saying it's not science. Science requires something fundamental, it can be expressed in one word, and I invite the readers of this post to say what it is. As soon as the word leaves their mouths, they will realize why social psychology cannot be science. When the physicists at the LHC observed what they thought might be the Higgs Boson, they refused to announce a discovery until the had a five sigma result. Five sigma is approximately equal to 2.8 * 10^-7 (explanation: http://blogs.scientificamerican.com/observations/2012/07/17/five-sigmawhats-that/ http://blogs.scientificamerican.com/observations/2012/07/17/...), and it expresses the probability that, if there is no Higgs particle, a random observation might create the observation that has been made. Five sigma. That's science. Opinion articles whose titles end in question marks are not science. And I should not have to explain this to educated readers. > I particularly love that you used the word "ascend"; I shan't be the first to compare positivism to scientism Scientism is an irrational confidence in science as an answer to all the world's problems. Social psychology is an irrational confidence in the power of articles that observe correlations without either proposing or testing theories about those correlations to accurately represent some aspect of reality.

lambda 13y ago

What makes you say that this kind of study can never ascend to the level of science? This kind of study is science. It's imperfect, but all of science is; science is just a collection of tools to improve our knowledge of the world, coupled with the body of knowledge so gained. Some of those tools are stronger than others, or can provide more certainty. Controlled studies in which you vary only one input to a system at a time, and observe the effect, are obviously a stronger tool than mere observational studies. But that doesn't make observational studies not science. It's perfectly possible to determine which of these effects are more or less likely to be responsible; the different hypotheses that you make would make easily testable predictions, which can be tested independently because different jurisdictions have made these changes at different times. Lead was banned in different countries at vastly different times (https://en.wikipedia.org/wiki/Tetraethyllead https://en.wikipedia.org/wiki/Tetraethyllead). Likewise, three-strikes laws have been introduced at different times in different places, and many places have never introduced such laws (https://en.wikipedia.org/wiki/Three-strikes_law https://en.wikipedia.org/wiki/Three-strikes_law). Easy availability of birth control and abortion may be a little harder to quantify precisely, but that can be done too. Since some countries have only banned leaded fuel in the past 10 years, within another 10 or 15 we should be able to say with considerably more certainty whether this has a causal relationship. If the crime data for countries which banned it later follow the same pattern, that's a very strong indication that the theory is correct. I would find it hard to believe that the three strikes laws would have the same effect, as the reduction in crime has been happening fairly consistently nation-wide, while three strikes laws have been implemented at vastly different times in different states, or not at all. Lead levels are much more consistent nationwide, as those have mostly been national laws (with the exception of California banning leaded gasoline a few years earlier). Anyhow, I strongly disagree that you can say that this isn't science. It's imperfect science; anything as complex as the social phenomenon of global crime levels is going to have a lot of complex, interacting factors and no one of them is going to be completely explanatory, and you are never going to be able to do significant studies with controlled inputs. But there are scientific tools for dealing with these somewhat more ambiguous situations, and we can and should apply them as the policy implications are quite substantial.

lutusp 13y ago