I don't disagree that it's possible, but I'd like to find documentation. The income inequality analysis you linked to seems to be applying the equivalizing to p

I don't disagree that it's possible, but I'd like to find documentation. The income inequality analysis you linked to seems to be applying the equivalizing to produce a proxy for individual income. EDIT: Here's more on the datasets the Guardian used: http://www.theguardian.com/world/2016/mar/07/millennials-how-we-revealed-predicament-generation-y-data http://www.theguardian.com/world/2016/mar/07/millennials-how... It does say they used equivalized household income.