4 ms·

The problem is that if you are trying to measure the effectiveness of government programs you can't exclude their value in your metric. In a very simple stripp

The problem is that if you are trying to measure the effectiveness of government programs you can't exclude their value in your metric. In a very simple stripped down version: * assume 20% of people need public assistance * assume that that assistance has been provided via government programs It is disingenuous to then assert that since 20% of people still need assistance (because you haven't counted the assistance in the metric) and therefore the government programs have not "reduced poverty" and need to be increased. I do agree that you might observe that 20% of people still need assistance and want to think about ways to make them more self-sufficient, but you can't use that number to obscure the fact that the assistance has indeed been provided.

coolgeek 5y ago
Part of the problem is that "poverty line" is a stand-in for "can (or cannot) afford food and shelter". Part of the problem is that there are two different statistics that need to be measured - the percentage of the population that could not afford food and shelter before assistance - the percentage of the population that still could not afford food and shelter even after assistance Part of the problem is that "poverty line" is an arbitrary demarcation point. (This doesn't even get into the fact that the "poverty line" is a national statistic, but the demarcation point in NYC is very different from that in rural Kansas.) You seem to be complaining that some percentage of recipients have moved past the arbitrary line, while ignoring that they still can't afford food and shelter. How is this not a No True Scotsman argument? The label you apply to a recipient (above/below an arbitrary line) is irrelevant. What matters is whether or not they can survive.
- gwright 5y ago
  > You seem to be complaining that some percentage of recipients have moved past the arbitrary line, while ignoring that they still can't afford food and shelter. No that isn't at all what I was trying to point out (or "complain" about). If I can refer to the two stats you listed as "before-assistance" and "after-assistance", then all I'm saying is that is terribly misleading to use the "before-assistance" statistic to assert that we aren't doing enough to help people. At least use the "after-assistance" statistic in order to measure the effectiveness of the programs. If you don't use the "after-assistance" metric then it doesn't matter how much assistance you give, the "before-assistance" number won't change and you'll continue to have people "living in poverty" even though that really isn't the case (because they have received assistance).

coolgeek 5y ago

Part of the problem is that "poverty line" is a stand-in for "can (or cannot) afford food and shelter". Part of the problem is that there are two different statistics that need to be measured - the percentage of the population that could not afford food and shelter before assistance - the percentage of the population that still could not afford food and shelter even after assistance Part of the problem is that "poverty line" is an arbitrary demarcation point. (This doesn't even get into the fact that the "poverty line" is a national statistic, but the demarcation point in NYC is very different from that in rural Kansas.) You seem to be complaining that some percentage of recipients have moved past the arbitrary line, while ignoring that they still can't afford food and shelter. How is this not a No True Scotsman argument? The label you apply to a recipient (above/below an arbitrary line) is irrelevant. What matters is whether or not they can survive.

gwright 5y ago
> You seem to be complaining that some percentage of recipients have moved past the arbitrary line, while ignoring that they still can't afford food and shelter. No that isn't at all what I was trying to point out (or "complain" about). If I can refer to the two stats you listed as "before-assistance" and "after-assistance", then all I'm saying is that is terribly misleading to use the "before-assistance" statistic to assert that we aren't doing enough to help people. At least use the "after-assistance" statistic in order to measure the effectiveness of the programs. If you don't use the "after-assistance" metric then it doesn't matter how much assistance you give, the "before-assistance" number won't change and you'll continue to have people "living in poverty" even though that really isn't the case (because they have received assistance).