I think there are a few additional factors that complicate the picture:
-front-page news is consumed way more than the rest, so the bad/good ratio in the top stories matters a lot. When I did an informal survey of two weeks of NYT front pages, bad was more prevalent than good by like 5:1.
-the badness of bad stories is much greater than the goodness of good stories. What's the opposite of "child brutally murdered"?
That could add up to a situation where, even if there's an equal amount of bad stories and good stories, the average news reader will come away feeling like things are very bad.
Oh yeah absolutely. As I was writing my post, I realized I had already said most of what I wanted to in that old French blog post, but the framing there was "it's not true that bad news are more frequent". So now I've reversed the framing, because whether it's true or not (which is a complicated question as my points and yours make clear) the feeling is what matters.
A good rule of thumb on good news vs bad news are from Gottman and Losadas: In general the amount of good news to bad news should be between 2:1 and 10:1, and optimally 5:1. If the positivity exceeds 91% then it is should raise eyebrows on being a Potemkin affair. Since the Lizardman ratio (minimal percentage of contrarians) is 4%, if the positivity rating is higher than 96% then the news can be considered homogeneous.
Anything like group coherence (Marcial Losada), couples wellness (John Gottman), and conformity (Scott Alexander) in public discourse. A more humble example is YouTuber EmpLemon's "Ratio" chart, and that extremely high positively reflects a kind of excessive fanaticism, which is good for entertainment but risky for truth-seeking. My hypothesis: the golden ratio is 4.5:1, and that the ranges of tolerance (~1.8:1 to ~11:1) and optimality (~2.8:1 to ~7.2:1) forms a log-scaled quadratic curve, or that doubling excessive positivity and halving needed positivity is the same. https://www.findingthefulcrum.org/antidote-positive-thinking-pn-ratio/https://knowyourmeme.com/photos/1489688-youtube
I think there are a few additional factors that complicate the picture:
-front-page news is consumed way more than the rest, so the bad/good ratio in the top stories matters a lot. When I did an informal survey of two weeks of NYT front pages, bad was more prevalent than good by like 5:1.
-even if people encounter a mix of good and bad, bad draws more attention (https://assets.csom.umn.edu/assets/71516.pdf). This is true in news consumption as well (https://www.pnas.org/doi/full/10.1073/pnas.1908369116)
-the badness of bad stories is much greater than the goodness of good stories. What's the opposite of "child brutally murdered"?
That could add up to a situation where, even if there's an equal amount of bad stories and good stories, the average news reader will come away feeling like things are very bad.
Oh yeah absolutely. As I was writing my post, I realized I had already said most of what I wanted to in that old French blog post, but the framing there was "it's not true that bad news are more frequent". So now I've reversed the framing, because whether it's true or not (which is a complicated question as my points and yours make clear) the feeling is what matters.
A good rule of thumb on good news vs bad news are from Gottman and Losadas: In general the amount of good news to bad news should be between 2:1 and 10:1, and optimally 5:1. If the positivity exceeds 91% then it is should raise eyebrows on being a Potemkin affair. Since the Lizardman ratio (minimal percentage of contrarians) is 4%, if the positivity rating is higher than 96% then the news can be considered homogeneous.
Huh, what is this supposed to apply to? Surely not news orgs?
Anything like group coherence (Marcial Losada), couples wellness (John Gottman), and conformity (Scott Alexander) in public discourse. A more humble example is YouTuber EmpLemon's "Ratio" chart, and that extremely high positively reflects a kind of excessive fanaticism, which is good for entertainment but risky for truth-seeking. My hypothesis: the golden ratio is 4.5:1, and that the ranges of tolerance (~1.8:1 to ~11:1) and optimality (~2.8:1 to ~7.2:1) forms a log-scaled quadratic curve, or that doubling excessive positivity and halving needed positivity is the same. https://www.findingthefulcrum.org/antidote-positive-thinking-pn-ratio/ https://knowyourmeme.com/photos/1489688-youtube