Citigroup: The rich will continue to get richer at the expense of everyone else

[Beckwin]

relatively close to the US based on the complete list, but many are appropriately far when you are only considering the top part (developed, first world nations) of the list. Japan, Norway, Sweden, France, Germany etc are toward the lower end of the first world on the Heritage list.

take for example this chart Kuya posted:
http://i45.tinypic.com/2crwqh1.png

NZ, Australia, UK are ranked more free/closer to the US than countries such as Japan, France, Norway, Sweden. There are certainly some exceptions, and it is a pretty loose correlation, but you have to keep in mind that these are Heritage's rankings and there are some extenuating circumstances to consider as well (for example in these graphs, though there isn't an absolute correlation between per capita GDP and well-being, on a case-by-case basis it would be worth examining the effects of it in relation to infrastructure, history, and whatnot).

[Cadsuane]

^ I would say those "feel" accurate based on OECD ratings I've seen over the years. And it's shit like that which tends to throw "Economics 101" out the window. Just for a grain of salt, here are Heritage/WSJ's rankings for economic freedom (I honestly have trouble believing they genuinely rank Canada above the US; I feel they do that for shock value/political reasons, but whatever)

http://www.heritage.org/index/ranking.aspx

even these rankings show some rough inverse correlation with the charts you've posted.

What, as in economic freedom is correlated to bad health care, social instability, and social mobility? Er, from my cursory glance at that link, no, not as they've defined economic freedom. (for example strict and transparent financial regulation is apparently something that contributes to economic freedom)

I made "economics 101" arguments that vindicate the position against income disparity. You need to realize that economics isn't a conservative mouthpiece, and that when I criticize your arguments with this sort of shit it's not a call to abandon progressivism etc., but to stop acting like a blustering ideologue and correlate your opinions with reality.

[Plow]

NZ, Australia, UK are ranked more free/closer to the US than countries such as Japan, France, Norway, Sweden.

The issue here is that the U.S. is an extremely obvious anomaly. 2 of the 3 "close" countries you listed are "free-er" than the U.S., but significantly below it on the charts.

Further, you've got Greece and Italy, 2 of the "farthest", being very high on all the "problem" rankings.


The biggest point I see this making is one I make a lot... we're not *that* far off. The countries that are a bit, but not a long ways "behind" us on freedom are the ones dominating the good end of the "problem" lists.

Really, these charts aren't particularly definitive, they're going to show what people want to see for the most part, because the raw correlations just aren't that justified.



Amusingly enough, the best correlation I see is this:

http://i45.tinypic.com/2crwqh1.png


The darker you are, the more likely you are to be below the line lol. Portugal is too poor to get there, but you've got Spain, Japan, Italy, NZ/AU below it, and almost all of the countries above it are like... the whitest in the world. lol

[Kuya]

The significance of the line is that there is a progressive increase in health and social problems as income inequality rises. Being below or above the line has no significance.

[Plow]

If the line has any meaning, being above or below it shows how that country is doing in respect to how you would expect it to be.


Now, if you're actively trying to make the graph into argument, sure, I can see trying to discount the importance of relation to the line, that's what scientific dishonesty is all about.


That's how you claim there's a clear correlation when there are clear indicators it's not the case throughout-- didn't look at exactly what the numbers are but there's at least 3 instances of countries being a couple inches to the left and being higher in problems compared to others.

[Cadsuane]

You didn't take statistics in high school or college I take it

[Plow]

There's a difference between deviation and half the fucking results being backwards.

[Cadsuane]

It's a linear regression, Plow. Significance is measured by r^2, which would've been nice to know (prolly around 0.4-0.6?), as well as the indices by which the axes are plotted.

[Kuya]

I haven't checked the source for accuracy.

edit; or definition

edit:
Why is plow talking about deviation?

edit: the income inequality is most likely measured by Gini

[Cadsuane]

Why is plow talking about deviation?

edit: the income inequality is most likely measured by Gini

Dunno, obviously not a distribution.

More than likely, but doesn't account for the vertical axis which is simultaneously indexing alot of crap. Sounds complicated, or contrived. Regardless, even the wikipedia article points out alot of similar research.

[Kuya]

The source is a book:

http://en.wikipedia.org/wiki/The_Spi...ways_Do_Better

Richard Reeves in The Guardian calls the book "a thorough-going attempt to demonstrate scientifically the benefits of a smaller gap between rich and poor", but also says there are problems with the book's approach. "Drawing a line through a series of data points signals nothing concrete about statistical significance [...] since they do not provide any statistical analyses, this can't be verified."[9] John Kay in The Financial Times says that "The evidence presented in the book is mostly a series of scatter diagrams, with a regression line drawn through them. No data is provided on the estimated equations, or on relevant statistical tests,"[10] although the evidence was available online.[11]

In response to criticisms of the book’s lack of statistical data, its 2010 paperback second edition added an Appendix which included such statistical data as Pearson r coefficients of the correlations between income inequality and the book's ten welfare indicators, for both its 23 rich countries and also for the 50 states of the USA. These revealed that important health indicators such as life expectancy and infant mortality both had r values of their correlation with income inequality of less than 0.5 for the 23 rich countries.

Swedish welfare economist Andreas Bergh thinks that Wilkinson and Pickett mistake correlation for causation, and points to Sweden as an example where increased economic inequality has gone hand in hand with better health and better gender equality.[12] Another Swedish Economist, Danne Nordling points out that while Wilkinson and Pickett rank Japan as the country with the highest income equality, all other sources rank Japan as being in the bottom of the OECD table.[13]

[Cadsuane]

http://www.equalitytrust.org.uk/why/evidence/methods

Some stats

[Kuya]

The US has no data on social mobility? Well that's kind of ironic.

edit: just checked, there is data for US social mobility, maybe they just didn't find it for the dates they were looking for

[Plow]

I understand that the idea is to show a trend, it still matters to what extent it varies.

Particularly when one is utterly disproportionate to the others, it shows other significant factors.

It also reflects more factors being unaccounted for when other singular factors show a distinct pattern in terms of being better/worse than the trend. Take out the U.S., Spain, Portugal, UK, Aus/NZ, Italy, Japan. Suddenly the line shifts in angle, showing a decrease in the strength of the correlation when you take out the highly diverse or "more brown" countries.

Do I have to explain why that's significant, too? If you can directly alter the correlation level with other statistics, that's strong evidence against the correlation actually being causation.


Further, imagine where the U.S. would end up if you considered all of those same factors except imprisonment and obesity.

[Cadsuane]

Correlation doesn't imply causation no matter how strong it may be. Seriously, just take a statistics course, it's not intuitive.

[Kuya]

It's a given that correlation does not equal causation.

Imprisonment and obsesity show moderate to strong correlation to income inequality.

Don't know what you mean by "more brown" country or what that has to do with this, but heterogeneity is probably related to income inequality, thus that a country is more heterogenuous should not disqualify it, as being heterogenuous is probably what exacerbates income inequality which in turn exacerbates social problems.

[Plow]

lol...

if you're not showing the graphs to imply that the correlation is a result of causation, what is your point?

[Kuya]

http://i49.tinypic.com/2jcyqm0.jpg

[Cadsuane]

http://www.threadbombing.com/data/me...g_facepalm.gif

[Plow]

So you're showing the graphs for no reason?


Your implication in attempting to display that "social problems" correlate with income disparity is *not* to imply they're related?


I perfectly understand that because they show correlation does not mean they are related... that doesn't mean that *you* personally are not attempting to display it as evidence that they are.


I've seen you do this shit before.