3 ms·

> Poverty, which the study defines as less than half the median U.S. income This seems odd as poverty is a function of households, not individual income. Half

> Poverty, which the study defines as less than half the median U.S. income This seems odd as poverty is a function of households, not individual income. Half the median for a single person is very different from half the media. For a single person with 9 kids. Or two people living together. And why half the median? Why not first quartile or quintile? This is pretty complicated to measure, so it seems unusual to not use Census’ Official Poverty Measure [0] that has many factors and is based on family size. I would expect discussion around why they created their own measure and didn’t use Census. [0] https://www.census.gov/topics/income-poverty/poverty/guidance/poverty-measures.html https://www.census.gov/topics/income-poverty/poverty/guidanc...

giraffe_lady 3y ago
That's more precise but also requires much more information. Recently I ran through that with a few different scenarios and most of them did come out really close to half of median. I can see why they would go with that number for simplicity and to reduce dependence on specific data and avoid accounting for reporting error in the data.
- prepend 3y ago
  Definitely, it’s harder to use. But I’d rather not do analysis than do wrong analysis. If I can’t measure poverty correctly, then using a simple number isn’t useful to generalize and make conclusions. And I certainly wouldn’t want to target interventions. There’s a reason why it’s complicated.
  - giraffe_lady 3y ago
    > If I can’t measure poverty correctly, then using a simple number isn’t useful to generalize Why not? It can be good enough to identify a pattern to focus more precise study on. "I'd rather not understand this at all if we can't have a perfect understanding" is a harsh and anti-science stance on this imo.
    - prepend 3y ago
      More like “I’d rather have minimally acceptable understanding than incorrect and harmful understanding.” The reason is there are many confounding variables so measuring poverty incorrectly and reaching a wrong conclusion can focus resources toward people who don’t need it (eg, single people making half the median wage who are not in poverty). I think sometimes you don’t have a choice. But in this case, there is a more accurate measure to use, so not using it makes conclusions less helpful. I think claiming things as “anti-science” is a gatekeeping function that is counterproductive. Especially when people don’t understand statistical significance and are trying to make conclusions that aren’t accurate. I don’t think things are “anti-science” or “pro-science” as these are imprecise and meaningless statements. I’m not sure what you mean by saying anti-science. But I think it’s not logical to assume something that is minimally effective is striving for perfection and something that isn’t significant is good enough.

giraffe_lady 3y ago

That's more precise but also requires much more information. Recently I ran through that with a few different scenarios and most of them did come out really close to half of median. I can see why they would go with that number for simplicity and to reduce dependence on specific data and avoid accounting for reporting error in the data.

prepend 3y ago
Definitely, it’s harder to use. But I’d rather not do analysis than do wrong analysis. If I can’t measure poverty correctly, then using a simple number isn’t useful to generalize and make conclusions. And I certainly wouldn’t want to target interventions. There’s a reason why it’s complicated.
- giraffe_lady 3y ago
  > If I can’t measure poverty correctly, then using a simple number isn’t useful to generalize Why not? It can be good enough to identify a pattern to focus more precise study on. "I'd rather not understand this at all if we can't have a perfect understanding" is a harsh and anti-science stance on this imo.
  - prepend 3y ago
    More like “I’d rather have minimally acceptable understanding than incorrect and harmful understanding.” The reason is there are many confounding variables so measuring poverty incorrectly and reaching a wrong conclusion can focus resources toward people who don’t need it (eg, single people making half the median wage who are not in poverty). I think sometimes you don’t have a choice. But in this case, there is a more accurate measure to use, so not using it makes conclusions less helpful. I think claiming things as “anti-science” is a gatekeeping function that is counterproductive. Especially when people don’t understand statistical significance and are trying to make conclusions that aren’t accurate. I don’t think things are “anti-science” or “pro-science” as these are imprecise and meaningless statements. I’m not sure what you mean by saying anti-science. But I think it’s not logical to assume something that is minimally effective is striving for perfection and something that isn’t significant is good enough.