June 01, 2012

Statistical Deception

Credit: Alan De Smet 
Ed Conard, a former director at Bain Capital, relies upon the presumed inability of people to scrutinize his assertions, when he indicates that the "distribution around the median income is very tight and hasn’t changed that much over time." This is presented as evidence that income inequality simply is not the problem it is purported to be. Well, if I remember my statistics correctly, the median is the middle data point in an ordered array of data points. So, even if the clustering of data around the median hasn't changed over time, that says nothing about:
  1. the distance between the median income and the top income
  2. the percentage of total income clustering at the top
You see, if the incomes above the median are multiplied by any factor whatsoever and the other incomes remain the same -- the median stays exactly where it is. Additionally, if you only do this to the incomes that are above those clustering around the median, i.e., the top incomes, then even the clustering around the median doesn't change. 

We must start challenging those who would mislead us into complacency or resignation.