31 ms·

Going with your experiences over soundly collected data is a very risky move prone to bias and discrimination (only noticing what affects people like you). Dat

Going with your experiences over soundly collected data is a very risky move prone to bias and discrimination (only noticing what affects people like you). Data, when reported by a reputable organization, is going to be substantially more predictive.

aww_dang 5y ago
Reputability is equally subjective. It is not immune from bias and discrimination. Incentives are another interesting factor to consider.
- TameAntelope 5y ago
  Reputability is not equally subjective (though it is somewhat subjective), it is based on past performance and current utilization through a trust network. If an institution has a history of providing a certain quality of service, and is successfully used to make highly consequential decisions, it is objectively more reputable than an institution with neither of those things.
  - dragonwriter 5y ago
    > Reputability is not equally subjective Reputability is the aggregate of individual opinion, which is the definition of subjectivity.
    - ChrisLomont 5y ago
      By that argument, all knowledge and belief and fact is the aggregate of individual opinion, so nothing would be objective. More likely there is a continuum, and more reputable places have more objective (and hence less subjective) output than less reputable places.
      - aww_dang 5y ago
        Consider the contrast between these statements: "The Bureau of Labor Statistics is a reputable institution" "A triangle has three sides" "Governments, politicians and their bureaucrats are trustworthy" "Two parallel lines will never meet" http://steve-patterson.com/deduction-induction-and-axioms/ http://steve-patterson.com/deduction-induction-and-axioms/
    - TameAntelope 5y ago
      Aggregating individual opinion (sometimes) has a washing-out effect on bias, which makes it valuable as a tool to push towards objectivity. [0] This is why I much prefer to follow guidance from institutions over individuals. There's also a "stewardship" in institutions that seems effective in countering other forms of bias (but not all). Of course there's a whole lot of bias left over that doesn't make this a fire-and-forget strategy, but comparatively it seems closer to objectivity than to just trust one's eyes alone, which has effectively zero systemic corrections for bias. [0] - You may know about this already, but I think of it like crowds singing, and if you'll afford me a bit of metaphorical license, here's a better explanation of how that works than what I could write: https://physics.stackexchange.com/questions/382429/in-concerts-why-is-the-audience-never-heard-as-off-tune https://physics.stackexchange.com/questions/382429/in-concer...

aww_dang 5y ago

Reputability is equally subjective. It is not immune from bias and discrimination. Incentives are another interesting factor to consider.

TameAntelope 5y ago
Reputability is not equally subjective (though it is somewhat subjective), it is based on past performance and current utilization through a trust network. If an institution has a history of providing a certain quality of service, and is successfully used to make highly consequential decisions, it is objectively more reputable than an institution with neither of those things.
- dragonwriter 5y ago
  > Reputability is not equally subjective Reputability is the aggregate of individual opinion, which is the definition of subjectivity.
  - ChrisLomont 5y ago
    By that argument, all knowledge and belief and fact is the aggregate of individual opinion, so nothing would be objective. More likely there is a continuum, and more reputable places have more objective (and hence less subjective) output than less reputable places.
    - aww_dang 5y ago
      Consider the contrast between these statements: "The Bureau of Labor Statistics is a reputable institution" "A triangle has three sides" "Governments, politicians and their bureaucrats are trustworthy" "Two parallel lines will never meet" http://steve-patterson.com/deduction-induction-and-axioms/ http://steve-patterson.com/deduction-induction-and-axioms/
  - TameAntelope 5y ago
    Aggregating individual opinion (sometimes) has a washing-out effect on bias, which makes it valuable as a tool to push towards objectivity. [0] This is why I much prefer to follow guidance from institutions over individuals. There's also a "stewardship" in institutions that seems effective in countering other forms of bias (but not all). Of course there's a whole lot of bias left over that doesn't make this a fire-and-forget strategy, but comparatively it seems closer to objectivity than to just trust one's eyes alone, which has effectively zero systemic corrections for bias. [0] - You may know about this already, but I think of it like crowds singing, and if you'll afford me a bit of metaphorical license, here's a better explanation of how that works than what I could write: https://physics.stackexchange.com/questions/382429/in-concerts-why-is-the-audience-never-heard-as-off-tune https://physics.stackexchange.com/questions/382429/in-concer...