Focusing in just authors may make this fall into the Simpson's Paradox (https://en.wikipedia.org/wiki/Simpson%27s_paradox https://en.wikipedia.org/wiki/Simpson%

Focusing in just authors may make this fall into the Simpson's Paradox (https://en.wikipedia.org/wiki/Simpson%27s_paradox https://en.wikipedia.org/wiki/Simpson%27s_paradox), if you look at the bigger set the clear trend may vanish. Even if authors are part of the population and living (some of) the influences that the general population have, there are a whole ecosystem and pressures that should be taken into account, including changes in the editorial ecosystem with time. There are other sets of public data that may or not track the general population, like social networks activity, blog posts and comments on different sites. But that is affected by changes in culture, population and external influences (including disinformation campaings), and the selection of sites may select also the kind of users that may add a bias to the results. I wonder what deviations would be seen at i.e. slashdot that should have around 25 years of comments, with all the previous objections that I already said.