> in the past three years the global developer population grew about 50%, from just over 31 million to just over 47 million
That's not really plausible. Unfortunately this is all you get on the methodology:
> Our methodology is based on two main pillars. First, we make use of reliable sources of developer numbers or direct indicators of their activity. This includes the number of GitHub accounts, Stack Overflow accounts, and employment statistics from the USA and the European Union. Second, we rely on our Global Developer Survey data, where we directly measure developer activity. So far, we've run 29 waves of this large survey, and in each, we reach more than 10,000 developers globally. We combine these two main sources to derive our estimates.
That's not really plausible. Unfortunately this is all you get on the methodology:
> Our methodology is based on two main pillars. First, we make use of reliable sources of developer numbers or direct indicators of their activity. This includes the number of GitHub accounts, Stack Overflow accounts, and employment statistics from the USA and the European Union. Second, we rely on our Global Developer Survey data, where we directly measure developer activity. So far, we've run 29 waves of this large survey, and in each, we reach more than 10,000 developers globally. We combine these two main sources to derive our estimates.
Plenty of space for them to screw up I think.