| ▲ | raw_anon_1111 11 hours ago |
| You’re way off the median household income is $80K https://dqydj.com/household-income-percentile-calculator/ |
|
| ▲ | twoodfin 11 hours ago | parent | next [-] |
| Median isn’t the average and Matt was computing the average household Amazon spend. |
| |
| ▲ | raw_anon_1111 11 hours ago | parent | next [-] | | The mean is almost always a meaningless statistics. It only takes a few people to buy stuff like this to skew it. http://www.sellersprite.com/en/blog/most-expensive-thing-on-... | | |
| ▲ | twoodfin 11 hours ago | parent [-] | | Be that as it may, the point at issue was the Amazon spending of the average US household. I’m not sure what point relevant to the discussion you’re trying to make, other than reflexively arguing with any use of means in economic analysis. OK, sure, tell Matt Stoller. | | |
| ▲ | ryandrake 7 hours ago | parent | next [-] | | I guess it's just always important/helpful to keep in mind that the average is almost certainly going to be misleading when the distribution is extremely skewed, as is the case for household income. It's usually a mistake to talk about averages in these cases, when the median is almost always going to be more meaningful. | | |
| ▲ | danwills 7 hours ago | parent [-] | | Agree, but can't we just include both average _and_ mean? And maybe min/max while we're at it? Seems like that could give a much clearer picture (without even needing a graph!?) | | |
| ▲ | tsimionescu 3 hours ago | parent [-] | | Min & max are also meaningless for most distributions, so probably you should instead look at P1 and P99 or something, and all of a sudden you're now talking about 5 numbers when all you wanted was a quick point. |
|
| |
| ▲ | vintermann 7 hours ago | parent | prev [-] | | The average Amazon spending of a US household, not the Amazon spending of the average US household. That second one gets weird. | | |
| ▲ | twoodfin 4 hours ago | parent [-] | | I’d be comfortable assuming household income and household Amazon spend are highly correlated. |
|
|
| |
| ▲ | cyberax 11 hours ago | parent | prev [-] | | It actually is for the normal distribution. | | |
| ▲ | twoodfin 11 hours ago | parent | next [-] | | Sure, but I’m certain US household income is not normally distributed, and I’d bet all the money in my pockets that US household Amazon spend isn’t normally distributed, either. | |
| ▲ | bandrami 10 hours ago | parent | prev [-] | | Household income is not normally distributed. In fact nothing with a hard zero can be normally distributed. |
|
|
|
| ▲ | h2zizzle 8 hours ago | parent | prev [-] |
| You're conflating two different things, but what you point out is still useful because it suggests that there are a few people on the higher end who make a LOT more and are dragging the mean up when compared to the median. The mean is probably not as indicative of the fortunes of most Americans as GP suggests. $3000 is a lot of money for most families, but there are a few for which it's increasingly not only inconsequential, but more like a rounding error. |
