Re: Two Types of Distributions Found In Nature

In our last episode,
, the lovely and
talented Bret Cahill broadcast on alt.politics:
But if you so much as say "mean income" life on earth as we know it
> will come to a horrific end.
> And the most curious part of it all is no one can explain why.
Here's why: unlike many variables (strength of carbon fibers, for example),
income is *not* normally distributed. The mean is a useless (yes,
meaningless) statistic.
Reply to
Lars Eighner
Loading thread data ...
Bret is, I think, referring to analyzing income distribution in any way, shape or form. Mean, median, standard deviation, plots, curves etc. The rich hate this. It makes them look, well, kinda greedy. Like they have just a little too big a piece of the pie, for the good of the rest of us. Gee, I wonder why.
Reply to
Jerry Kraus
What would "normal" income distribution look like?
Reply to
ta
It's a statistical term -- the "normal" distribution is a standard curve. Income doesn't fit it very well. But the point is, most rich people don't want income distribution analyzed at all. It makes them look greedy.
Reply to
Jerry Kraus
Before jumping to any conclusions about monied interests' corp. media propaganda, we need only focus on one issue:
Why there is only _one_ distribution in Nature where it is OK to discuss the median value but not the average mean value?
This is getting curiouser and curiouser.
Bret Cahill
Reply to
Bret Cahill
Well, I suppose the average American might be irritated to discover that his income is nowhere near the mean value because the distribution is so skewed to the right by the super-rich!
Reply to
Jerry Kraus
formatting link
Explains his point.
-tg
Reply to
tg
We need a reference that suggests average mean only applies to normal distributions.
Right now we have _one and only one_ distribution out of the jillions found in Nature that is unique in that the median value is meaningful but the ave. mean value is meaningless.
Bret Cahill
Reply to
Bret Cahill
In our last episode, , the lovely and talented ta broadcast on alt.politics:
A normal distribution is generally described as "bell shaped." In a normal distribution, the mean, the median, and the mode are all the same.
The distribution of incomes is an exaggerated reverse J shape, which is to say, almost everyone makes almost nothing, and almost no one makes almost everything.
Reply to
Lars Eighner
I think that is exactly Bret's point. The rich don't really want the poor to know that.
Reply to
Jerry Kraus
Your wrong there are many issues that civilized human society considers to crass to discuss.
For instance. What is the mean number of sexual partners for every person in the world ... or in the US? I would venture to say the mean value is much larger then the median value. Also I would venture that nobody really wants to discuss this. And the answer for this is the same. The distribution is not normal. The distribution has a lower boundary condition but not upper boundary condition and the median value is too close to the one boundary condition to create a normal bell curve. This tends to flatten out the distribution causing a disparity between the mean value and the median value.
But then again. Since you so very much loving playing with these numbers, you already know that. You just like to try an stir up caste angst.
Reply to
MichaelNJ
In our last episode, , the lovely and talented sinister broadcast on alt.politics:
Modern economic doctrine is that wealth is used to produce income, so largely the distribution of wealth and the distribution of income are merely two ways of looking at the same thing. Indeed, the distributions are very similar and so skewed that is difficult to work with them at all except on logrithmic scales. See pareto distribution.
Reply to
Lars Eighner
Nonsense. We will discuss it until the sky falls if you like.
Reply to
pico
I know some millionaires and they don't mind such data. Not one bit.
Reply to
pico
No. The ratio of the mean to the median is one indicator of how skewed to the rich that distribution is.
(There are other statistics for testing for non-normal distribution. I happen to have the privilege of being among the first few to implement the Ozturk's statistic in computer code.)
Reply to
Robert Vienneau
Ah, very good. So you are concerned that those few who have most of the bucks places the majority is actually below the median. Right? What difference does it make? If it true that the majority would be below the median, and then the money of the rich were redistributed, would it make a difference? Not for long. Prices would rise until purchasing power was diminished to the acceptable norm.
Reply to
pico
You mean Median, not mean.
Reply to
pico
Such as in IQ tests?
Interesting. I wonder if there isn't at least one more such case. Let's see.... race and jail terms. There's one. No?
Reply to
pico
I believe Jerry means "mean", just as he says. But math is hard.
Reply to
Robert Vienneau
No. The distribution of wealth is more skewed.
Reply to
Robert Vienneau

PolyTech Forum website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.