My criticism is about how the dramatic language differs from the banal content of the article.
Titling it "The World Happiness Report Is a Sham" and calling it "beset with methodological problems", I would expect some more serious scientific malpractices, like data fabrication, calculation errors, sampling problems, p-hacking, etc., not "I think there are some problems with this variable".
Disagree. Whether I'm entirely fabricating data that claims A by writing numbers into an Excel sheet, or whether I'm doing a survey that measures B and then claim it means A, isn't materially different in outcome. The outcomes are just as bad, and that's what people care about. Maybe you as a researcher care that the former is more immoral, but to everyone else it doesn't matter.
I think there's a difference in outcomes between fabricating data, and getting data that still remains validly gathered, but measures something subtly different.
And I think the general public can make meaning of that
difference and have a stake in both – in the same way that the general public knows that stock market values and economic security are different things, even though people still have a lot riding on retirement plans based on stock investments.
Titling it "The World Happiness Report Is a Sham" and calling it "beset with methodological problems", I would expect some more serious scientific malpractices, like data fabrication, calculation errors, sampling problems, p-hacking, etc., not "I think there are some problems with this variable".