Pigeon
coo coo
No bio...
User ID: 237
Pigeon
coo coo
11mo ago
This is not the "public consciousness" understanding of the DKE. That is the claim that "people who say they are real good and talk about how good they are are actually no better or even worse than the people who say they are bad".
In that case I stand corrected. That seems silly.
This is not how poor correlation is usually defined either in real life, that's normally given by r, and you can have very high r while the statement "poor performers overrate their performance and good performers underate theirs" (say r = 0.99, an out of the world level correlation for anything in the social sciences) is still true.
Well, of course "can have very high r and still have [that statement] be true" is true with r<1, especially since the data is bounded and poor performers are naturally going to be more room to overestimate and good performers underestimate. I thought the point of DKE was that r was low,
Pigeon
coo coo
11mo ago
Pending a detailed read of Nufher et al. and Gignac & Zajenkowski, this appears as one of three -- either the blogpost is simply wrong; or I have a misundertanding of the Dunning-Kruger effect; or the author has done a really shit job at explaining himself. Either way, I'm not convinced so far.
It means that we can throw random numbers into x and y — numbers which could not possibly contain the Dunning-Kruger effect — and yet out the other end, the effect will still emerge.
My understanding of the DKE is that self-assessment is poorly correlated with objective ability in such a way that poor performers overrate their performance and good performers underate theirs. In this case, the lack of correlation in Fig. 7 from y being a variable with a uniform distribution uncorrelated with x already shows the effect! I'm not sure how the author is so sure that plotting uncorrelated variables and "showing" the DKE disproves it, as the entire point is that they're poorly-to-uncorrelated!
If my understanding of the Dunning-Kruger effect is right, I suspect the author may be right to some degree (just based on personal experience, I think DKE is extremely oversold, and even if true is unlikely to be very important), but his working is definitely wrong.
Pigeon
coo coo
11mo ago
Granted.
What makes this different from rule consequentialism, then? Or are they the same in your definition?
Pigeon
coo coo
11mo ago
There is a passage in the Zuo Zhuan, under the 21st year of the reign of Duke Zhao of Lu, where a member of the lower aristocracy in Spring and Autumn China dies from allowing an enemy to take a shot at him after missing his own shot and, prior to a second shot he was readying, was chastised by his opponent (who shot him dead) that taking two shots in a row without allowing a return shot was dishonorable.
Even granting that "breaking decorum has social consequences" and thus you can offer consequentialist explanations for actions like these, I think it's important to acknowledge that there are many people throughout history who are much more on the deontological side than otherwise.
(In the end I am more of a consequentialist myself, but I see the value in deontological thinking and virtue ethics as proxies for these, and I can somewhat understand how deontological thinking turns in the heads of those that accept it..)
Pigeon
coo coo
1yr ago
I don’t know about the others, but unlimited world class healthcare is not unlimited healthcare to the maximal extent possible.
Pigeon
coo coo
1yr ago
I did once, back in college, have a conversation with someone whose position on death camps and genocide was "no bad tactics, only bad targets," and that whether such things are immoral — or not — depends entirely on who is using them against whom and whether or not the latter group "deserves it."
Out of morbid curiosity, who, in this someone’s mind, “deserves it”?
Pigeon
coo coo
1yr ago
If I see a guy using !!! in any context, I'm going to rip off his pants to see if he has any balls underneath them.
I have to admit I have written notes with !!! to indicate important developments or other critical issues…
Pigeon
coo coo
1yr ago
https://www.lbc.co.uk/radio/presenters/maajid-nawaz/maajid-the-left-is-no-longer-liberal/
It’s been around since at least 2016. I swear I’ve heard it earlier than that but not sure from where…
Pigeon
coo coo
1yr ago
Or you could just have no say when you get whipped in the House of Commons in the UK.
Pigeon
coo coo
1yr ago
Oh, I work as a doctor. History, especially the Great Divergence, is just a hobby thing I keep up with.
Pigeon
coo coo
1yr ago
Glad to be of service! If there are any questions I can try to answer to the best of my ability.
Pigeon
coo coo
1yr ago
If we are to reach for institutions that are lost to history (at least continuity-wise), surely we can reach farther back for the Sumerian city-states? We even have written history from this.
They had this weird, subordinate relationship with China, changing their allegiance as new dynasties and invaders took control.
Probably worthwhile to note that this is true mostly of the Joseon period, and less true as you go back from that point, fromTang intervention on behalf of Silla in the Korean Three Kingdoms period + the Silla-Tang war; all the way to Han conquests of parts of the Korean peninsula.
At home, Korea was run by actual civilian governors, to the point where they could actually have military coups in the 11th or 12th century when everyone else was feudal.
China certainly wasn't more feudal than Korea at this point in time (especially 11-12th century), I don't think. There's a good argument that feudalism proper ended in China with the Qin dynasty (221 BC).
And of course China flooded the peninsula with so many troops that the Japanese army was ground down.
IIRC ~150k soldiers were sent during each invasion wave from Japan. Ming China sent something like 50k soldiers each time. The Ming-Joseon side were quite outnumbered when it came down to soldiers (numbers may even out more if you count Korean militia).
There were actual Ming advantages, such as much superior cannon and field artillery. I don't think "flooding the peninsula with so many troops the Japanese were ground down" really is accurate.
Pigeon
coo coo
1yr ago
I'm not aware of any major religion that's more cool with atheism, barring perhaps Buddhism, but that's just a distant cousin.
Does Taoism or Confucianism work, if either count as religions?
Pigeon
coo coo
1yr ago
I haven't played piano in 15 years or more, and the piece I would go for would be Rach prelude op. 3 no. 2. Sometimes familiarity beats everything else?
Pigeon
coo coo
1yr ago
I think the left qualifies it in a sort of paradox-of-tolerance way
Not this again!
Pigeon
coo coo
1yr ago
Granted that Mao was not a good person, he didn't set out to kill 100 million people. He made some bad decisions that inexorably led to a famine which killed 100 million people.
Absolutely nitpicking, but I don‘t think Mao killed a hundred million people in the Great Leap Forward. The commonly quoted numbers are anywhere from 18 million (CCP official estimates) to 60 million (some of the more loony estimates), with most reputable estimates going from 30 to 45 million. (The commonly quoted one I grew up with was 36 million.)
100 million sounds like something right out of the Black Book of Communism, which has very artistic ways of arriving at casuality figures.
Pigeon
coo coo
1yr ago
Where’s that graphic from? I’m pretty sure Newport is the name of many suburbs around the world, and Propecia and Sinutab are actual brand names for real medications (for finasteride and pseudoephedrine + paracetamol respectively). This seems too stupid to be true.
In that case I certainly did not get that impression from the blogpost, and I don't think this has anything to do with "autocorrelation" as much as it has to do with the data being bounded, which I think is another argument entirely. Incidentally I think the bounded-data explanation, which seems the most obvious one to me (along with other ones like the better-than-average effect and simple regression to the mean), are much more convincing than wrangling about autocorrelation. It's also using "autocorrelation" in a weird way.
I remain of the opinion that DKE is probably artefactual or minor at best, but the blog is still either poorly written or wrong.
More options
Context Copy link