I'm also not really sure about this article. I find the evidence in that graph not to be compelling enough for summarizing with "bottom line:" and some real clean explanation.

I find the trending cumulative percentage a little confusing to think about.

I also wondered, does this mean there is some income bracket losing people? That is, no more families making 25-75k but still as many making <25k, so the <25k have fewer slightly-above peers?

Then I wondered if maybe this means something about the <25k potentially having less income? That percentage stays flat, but those other earners are changing classification to rich... I don't really know.

After writing this up I realize that what I'd like to see is the whole income distribution. This cherry-picks certain points on the CDF and plots them over time (the income cutoffs of 25k and 75k).

What's interesting to me is whether or not we've gone from a unimodal income distribution to a bimodal one, in the sense that there's a trough between the lowest income families and the highest. If we have a 'middle class' we avoid that, and probably that is socially important.

I think this graph is better for understanding the main article, which plots the Gini coefficient for income distribution over time by country [a few, anyway]

http://en.wikipedia.org/wiki/File:Gini_since_WWII.svg http://en.wikipedia.org/wiki/Gini_coefficient

So, I'm not convinced that this is helping the situation:

- Is this data taking into number of working adults ($/house-hold)?

- What about the rising costs of living (like eating healthy)? [is this normalized based on any consumer metric of just inflation?]

See also:

http://en.wikipedia.org/wiki/Lies,_damned_lies,_and_statisti...