zlacker

>>Restle+(OP)
The methodology finally addresses a pet peeve of mine to adjust based on prices and hours worked: "we ranked 178 countries using three measures. The first is GDP per person at market exchange rates. It is simple and intuitive, and widely cited. But it ignores price differences between countries. The second measure adjusts incomes for these local costs (known as purchasing-power parity, or PPP). This offers a better guide to living standards but one that takes no account of leisure time: the share of people in work, and how long they work, varies by country. Our final yardstick accounts for both local prices and hours worked"

>>Restle+b
Why is average still such a persistent metric though? Isn't median more representative of an individidual's actual purchasing power?

If they paired average metrics with gini coefficient, I'd be happy

>>tetris+h4
NB: "average" can mean any of three measures of central tendency: the mean (sum/count), the median (middle value), or the mode (most frequently occurring value). All three may be valid choices in specific contexts.

Whilst it's common to interpret "average" as "mean", this isn't strictly accurate or reliable.

The Economist's article errs in failing to distinguish which measure of central tendency is actually meant by "average".