One point that the pandemic has brought home to me is just how narrow people's expertise is. I'm regularly surprised by how a celebrated professor of X can exhibit a sub-college-level understanding of Y, even when X and Y are related. /1
Ask a professor of (say) virology about the properties of viruses and I assume you'll get a dependable answer; but ask about human behavior, public policy, causal inference, the law – or God forbid, ethics – the answer can go whichever which way. /2
Call me naive, but in the microenvironment where I work – in the intersection of science and the humanities – people take a certain amount of pride in being well read, even outside of their official domain of expertise. /3
As Warren Buffett has emphasized, having a narrow circle of competence is not necessarily a problem. The problem appears when people don't know where the boundary is, and operate with the same confidence inside and outside of it. /4
By all means, let's continue to rely on the judgments of experts operating within their circle of competence. But let's not assume that they have any particular expertise – or even knowledge – about stuff outside of it. /5
It might help to stop calling people "experts" tout court, and always append the area they're an expert in. That would make it a little easier to assess their location relative to the circle of competence. /6
We should probably be attentive to halo effects, in which our positive view of somebody's competence in one area bleeds over into other kinds of positive judgments. /7
Finally: whenever we're dealing with complex, multi-faceted problems that don't belong to any one discipline, we might want to pay particular attention to people who've attained journeyman status in more than one relevant discipline. /8
• • •
Missing some Tweet in this thread? You can try to
force a refresh
Periodic reminder that in terms of outcomes, Swedish corona policy is thoroughly average in EU comparison – not exactly a model to be emulated by the rest of the world, nor a crime against humanity that should be prosecuted in the Hague.
Only about two weeks separate Sweden and the EU in this graph, which shows cumulative confirmed COVID-19 deaths per 1M people. Source: ourworldindata.org/coronavirus-da…
There's a lot of disinformation. Here's @ForeignPolicy, which I've always considered a reputable publication, writing that Sweden's death rates are among the highest in Europe: foreignpolicy.com/2020/12/22/swe…
Let me preface this by saying I think "war" is a terrible analogy for scholarly discourse: (a) it trivializes the real thing, (b) it suggests science is a zero-sum game, (c) it triggers norms suggesting that "all is fair," and (d) it encourages people to choose teams. /2
Anyway, let's ignore the clickbaity headline and the hard-charging introduction and go straight to the conclusion. Here's what Sauer thinks the evidence shows. /3
The FT and @gideonrachman go down a well-worn path when criticizing the Swedish corona strategy – and end up with a predictably misguided conclusion. /1 ft.com/content/4f6ad3…
In Rachman's narrative, "Sweden's failure" can be traced back to a single "decision" – viz., the "refusal to go for a hard lockdown" – which was "a policy error" driven by "self-confidence … shaded into arrogance about the country's supposedly superior rationality." /2
As is well-known by now, the Swedish constitution does not permit a hard lockdown. There were things on the margin that the government could have done differently, but home confinement, travel bans, etc., were simply not in the cards. /3 bppblog.com/2020/04/23/the…
Terrific assessment of projections of demand for Swedish ICU beds. The first two panels are model-based projections by academics; the third is a simple extrapolation by the public-health authority; the fourth is the actual outcome /1 dn.se/nyheter/vetens…
tl;dr Model-based projections drastically exaggerated the actual demand – sometimes by more than an order of magnitude. Today the number of patients in intensive care is about 450; it never exceeded 600 /2 icuregswe.org/data--resultat…
Around the same time, if I read their data file correctly, the IHME projected a demand of 4400, with a 95% uncertainty interval of 1400–11000. The real number is therefore way outside the interval /3 healthdata.org/covid/data-dow…
Dear colleagues, fellow academics, and experts of all kinds: Now would be a good time to practice the intellectual virtue of epistemic humility. /1 plato.stanford.edu/entries/modest…
Being an expert involves not only knowing stuff about the world, but also knowing the limits of your knowledge and expertise. It requires both cognitive and meta-cognitive skills. /2
It is, of course, fine and good to have opinions and to express them in public. The point is that true experts express themselves in a way that reflects the degree of confidence they are justified in having in their beliefs. /3
Beware of people claiming to be really good at logic and rationality <A THREAD> /1
One thing we learned from Kruger and Dunning’s pioneering work is that people who are in fact very good at logic tend to *underestimate* their ability doi.org/10.1037/0022-3… /2
People to the right in this graph did very well on K&D’s logic test, and on the mean they underestimated both their ability and their performance on the test /3