Right. I get that. I didn't think that's what you meant.
Now that's an actual criticism I would be happy to discuss.
I don't think I've ever heard the term anamolous cluster in relation to statistics. So I'll try to answer without being 100% certain of your intent. I'd suggest that the result I got was expected - some groups of whites will have greater poverty than some groups of blacks. It's not unusual, or an anomoly. It's a 100% expected result for a non-uniform distribution. I'd suggest the unusual anamoly would occur if the data didn't result in this.
In short calling it anamolous cluster seems to me more like a way to uncritically dismiss the data point. Ultimately though, no matter what you call it or whether you agree with the above answer - what matters is 'what does this fact mean, especially in regards to privilege?'
Um....ok. I assumed you meant there was something significant or meaningful about that conclusion. But if that's it, if you're really stopping there, then I take it back: you were right, you have demonstrated that some groups of whites will have greater poverty that some groups of blacks.
But OF COURSE THAT'S TRUE. I mean, wtf? Do you think it's
surprising there's an overlap in the two distributions? Like, in order for systemic racism to be real, the richest blacks would have to be worse off than the poorest whites?
That sounds like a correlation vs causation issue and one that seems on the surface to have a fairly plausible explanation. People live around power lines. There's more powerlines around higher population areas. So all it's ultimately measuring would what areas have higher population. That is this is coorelation instead of causation. I'm not particularly sure how what i did falls into the coorelation vs causation issue. Maybe you can explain?
First of all, it has nothing to do with higher population areas, so I think you're missing the point. Cancer may not be actually random, but we can treat it as quasi-random. So pretend you are throwing darts at a map on the wall, and each hole is a cancer case. Although the dots are random, they aren't distributed evenly: some areas have more dots, some have fewer.
Now, what you want to "prove" (because you are a sleazy pseudo-journalist) is that high-tension powerlines cause cancer. So you mark all your high tension powerline towers on your map, and then you look for the ones that happen to end up inside your random clusters of dart holes. And some of them will be. Boom: there's your "evidence" that EM causes cancer. So you go to the town where that cluster is, and demonstrate that the cancer rate within 1/4 mile of that tower is 5.7 TIMES THE NATIONAL AVERAGE and the newspapers go nuts. (This is basically what happened in Long Island in the 80's.)
But, of course, there's no correlation OR causation. It's just that random distributions are not smoothly distributed. They are clumpy.
The same is true for any other demographic data. Such as poverty rates.