After seeing the figure [1] I'm pretty confident this relationship described is artifact. Rather than comparing raw medians a statistically robust comparison wo

After seeing the figure [1] I'm pretty confident this relationship described is artifact. Rather than comparing raw medians a statistically robust comparison would have been log(sentence counts) or geometric means, and there would not have been a significant change. The mean number of sentences per decade is presented in the original study [2] and it only increases. Similarly I find the position in the textbook of the climate context analysis unconvincing. OTOH there appears to be a move towards including more information about the effects of climate change on biology and ecology rather than just the physical premise/greenhouse effect, so that seems appropriate. [1] https://media.nature.com/lw767/magazine-assets/d41586-022-04487-6/d41586-022-04487-6_23849190.jpg https://media.nature.com/lw767/magazine-assets/d41586-022-04... [2] https://journals.plos.org/plosone/article/figures?id=10.1371/journal.pone.0278532 https://journals.plos.org/plosone/article/figures?id=10.1371...