Thread by @DrMatthewSweet

My thanks to @lionandunicornnews for altering me to this. This conversation with @ezraklein is a great example of how @johannhari101 gets his dodgy claims past his interlocutors. This was for the @nytimes. Not some gormless wellness podcast. nytimes.com/2022/02/11/pod…

First, he replies with a compliment to the interviewee. He does this all the time, no matter the quality of their points. (He even did it with Bill Maher.) Like saying your name over and over again, this is a technique of persuasion.

But it moves him on to making a point that supports the Stolen Focus thesis, but is deeply dubious. And @ezraklein knows it’s wrong, because he’s recently heard it debunked. So quite rightly, he challenges him on it.

Oh dear. Will Johann concede? Or reveal the source data from c.1922 that proves his point? (Spoiler: I don’t think it exists.) No. He just bulldozes @ezraklein with something that seems to me to be entirely false.

Johann says there is a consensus that our hours of sleep are falling, and that a minority disputes this. Let me give you some actual evidence. This is a study about UK sleep patterns since the 70s. onlinelibrary.wiley.com/doi/full/10.11…

Here’s another: A study of studies, collating material collected over decades. Pretty authoritative. ncbi.nlm.nih.gov/pmc/articles/P…

But @ezraklein doesn’t have all this at his fingertips, and probably trusted Johann and his editors at @bloosmburybooks to have done their job. Is the book any more trustworthy? Well, here, in Stolen Focus, is a melange of stats about sleep. Where on earth do they come from?

Well, that top one (and the 42 per cent figure) comes from a Gallup poll. Not a scientific study. A Gallup poll. And what about the National Sleep Foundation? They sound authoritative, don’t they? Uh-oh. Anyone who read Johann's anti-anti-depressant book won't like this.

So Stolen Focus ignores reliable studies that would dismiss its thesis, and seizes upon really ropey material that supports it. And the book is doing really well – which, I think, is the only strong evidence for its argument that nobody is paying attention.

Sorry got that handle wrong - @LionUnicornNews

and alerting not altering, obvs.