If the “happiness/satisfaction” metric is rubbish though, then all the correlations are fairly meaningless, and are of interest only to the extent that they do or do not back up one’s own political opinions.

Applying this scale cross-culturally, you just wind up measuring some weird convolution between actual subjective well-being and some kind of “what range of values is it culturally acceptable to answer this personal question about my wellbeing from a stranger with?” effect.