It’s the second time you’ve had a snarky reply so I can’t tell if you’re having a good faith conversation.
The average wealth between me and Elon is several hundred billion dollars. That gives you very little information about me. Which is why people can hang too much inference on a simple average. Like Nate Silver said in The Signal and The Noise, the real discussion for the data literate is about uncertainty in models, not just drawing conclusions from “averages”
I'm being snarky because your criticism of the tool, and its data sources, is a weird form of nut-picking.
You're able to purchase groceries for your family, for your diet, in your locality, from your available stores, for less than the stated average.
You think the diet should be different, and blame "society!" for the nutrition goals not resulting in a lower budget.
This is not a serious criticism. It is an unverifiable anecdote coupled with generic contrarianism.
The fact some people spend more or less on groceries is already factored into the data, as it's an average of prices. Averages are imperfect. The fact it's an average of prices (instead of spending) makes it slightly better, but anecdotal data doesn't meaningfully contribute to a discussion about it.
The diet, too, is probably imperfect, but the tool needed to normalize costs, not assemble a Costco rice-and-beans nutritionally complete diet to minimize costs.
So I have absolutely no idea what point you're trying to make beyond "I like to sound smart."
Methodology is THE main criticism of research; it’s the most important piece. It’s more important than nit-picking results or anything else. Unless your the type to believe in starting with a conclusion and working backwards, but that’s bad science.
And that doesn’t mean the method has to be perfect. But if it doesn’t reflect the true problem, or if is too weak to drawn conclusions from, it’s just story-telling.
The point I’m making is we should acknowledge the model assumptions. If we’re saying, for example, a living wage is expected to provide the average car, we need to acknowledge it now becomes the floor and is no longer the “average”. That’s a fine point to debate, but it requires some data literacy that is absent in this discussion.
And, as an aside, if you think snark is somehow justified because someone is criticizing a tool or method in legitimate way that you don’t like, you need to revisit the HN guidelines.
It’s been discussed elsewhere , but we can’t get to details if we can’t get past the simplest of concepts. I think you are missing the critique because you’d prefer to argue than understand, but I’ll try to be more explicit.
Significant parts of the method are built on surveys. Surveys are often a poor measure because they tend to be more subjective and biased. That’s why nutrition surveys of dietitians have significant amounts of error. In addition, the surveyed data isn’t normalized for socio-economic class; that is, it sets the expected value at the “averages”. The implication is the living wage should provide the average level of subjective consumption. That, in turn, means the current average is now the lowest we are willing to accept as a society. That’s all well and good to discuss, but that’s more nuanced than anything you’ve brought up. And that’s doesn’t scratch the surface of the flawed reporting, where uncertainty isn’t part of the main discussion.
It’s clear the site is for laity but the problem (as we’ve seen) is that it just feeds people’s confirmation bias when they are more interested in being right than in understanding.
The average wealth between me and Elon is several hundred billion dollars. That gives you very little information about me. Which is why people can hang too much inference on a simple average. Like Nate Silver said in The Signal and The Noise, the real discussion for the data literate is about uncertainty in models, not just drawing conclusions from “averages”