Found with AI but this article claims the opposite of the GP. https://www.americansurveycenter.org/research/the-state-of-american-friendship-change-challenges-

Found with AI but this article claims the opposite of the GP. https://www.americansurveycenter.org/research/the-state-of-american-friendship-change-challenges-and-loss/ https://www.americansurveycenter.org/research/the-state-of-a... And the various articles found seem to agree. The GP claims to have combined 30 surveys. I wonder if they wound-up with one of those statistical paradoxes where a combining of data sets points to something different from each individual data set. But something probably isn't true 'cause the individuals sets are well curated and the combination isn't.