EDIT: This post presents open-access data from a large survey that anyone is allowed to download and analyze. It’s gone viral because a Twitter user named Philippe Lemoine downloaded the data and used it to investigate politics and mental illness. That analysis is not described here, it’s not my analysis, and I have various caveats about it, which are described here. Sorry for any confusion.
Thanks to the 8,043 people who took the 2020 Slate Star Codex survey.
See the questions for the SSC survey
See the results from the SSC Survey (click “see previous responses” on that page)
Some people expressed concern about privacy on the survey. Originally, respondents could see aggregate responses, including the responses of people who marked their answers private. I figured this was okay because nobody’s responses could be connected – ie you could see that one person put their age as 83, and another person put their country as Canada, but because the table order wasn’t the same you couldn’t link these together to form a coherent picture of an 83 year old Canadian. Some people still expressed concern about a few of the long answers, since some people might have put personal information in there. There’s no way for me to eliminate only the private people’s responses from Google Forms and still display the information to you like this, so instead I’ve removed all long answer questions. If you’re interested in those, you can find them in the downloadable data files. Sorry for not doing this earlier, and I hope this compromise is okay to everyone. I’ll try to get a clearer picture of what people want before the next survey.
I’ll be publishing more complicated analyses over the course of the next year, hopefully starting later this week. If you want to scoop me, or investigate the data yourself, you can download the answers of the 7000 people who agreed to have their responses shared publicly. The public datasets will not exactly match the full version, some overly identifiable questions (eg age) will be binned, and a few sensitive subjects will not be included.
Download the public data (.xlsx, .csv)
Finally, the game results. I randomly selected Game 3, “Prisoner’s Dilemma Against Your Clone”, chose a random respondent as the prisoner, and found someone similar to be his clone. Of the two clones, one cooperated and one defected, so the defector gets the full prize. That defector’s public key is “gwern is my waifu and paperklipot maximizer”. Please email me within one week at scott[at]slatestarcodex.com with your private key and a Paypal account where I can send you money (or a charity you want me to donate to).
That last paragraph is probably a metaphor for something.
I don’t know how I feel about this.
Getting to put that phrase into the blog is probably worth more than the prize money.
Who knows whether the winner exists at all, or whether or gracious host just wanted to write that with the cover of plausible deniability? 😉
Feel good about it? I always try to interpret ambiguity favorably – so see it as a sign of admiration or respect for the things you do. (Some people tend to take any ambiguity negatively – what did he mean by that? – but I think they generally lead unhappier lives because of it.)
It’s actually a morality play about how it doesn’t pay to be like a person who would write “gwern is my waifu and paperklipot maximizer”.
Although it does pay to be that specific person.
If you think you might be like that person, you better become that person.
Although that person did betray someone, knowing only that they were similar.
Yeah I think it’s more a morality play about how it totally does pay to be such a person.
At least when playing single-round games.
At first I thought it was a weirdly spelled “paperclip”
Is it not?
It’s a portmanteau of “paperclip” and “klipot”. Dunno if you’d define that as “a weirdly spelled ‘paperclip'” or not.
Did the person mean “gwern is both my waifu and my paperklipot maximizer” or “gwern is my maximizer of both waifus and paperklipots”?
Gwern does maximize waifus so the second one sounds plausible.
Is anyone able to load the results (after 15 minutes I gave up)?
Worked fine for me, just took a few seconds (both .xlsx and .csv)
Sadly it just doesn’t work for me after several attempts.
Doesn’t work for me either 😔
ETA ah! – for me, patience was the answer – took 5 mins to load..
Normally I’d be sad that I didn’t win, but let’s face it, no one could beat that 🙂
I read this and wondered whether gwern is meant to be the waifu, or the maximizer of waifus.
Why not both ?
Well he did make that random waifu generator…
Came here to say this.
…perhaps the question about blood donation should be retitled in the results? Right now it reads as though only 50% of readers have blood.
The rest of us have ichor.
I am in thralldom, for I am a Golem.
Clay is the way. Or mud, rather than blood.
Plainly the other 50% are vampires. They don’t have blood yet.
24M seeking blood.
Growth mindset!
Beep Boop
Or want blood, or have consumed blood.
Or are blood
Well it should be obvious from the colors that the other half are Crips.
If it’s easy to generate, could you post a list of the public keys for each clone pairing when you get around to that analysis? I’m curious to see what my clone did.
I selected clones manually by eyeballing.
Will you be posting the public key of the winner’s clone?
No. I felt bad about this because their survey answers suggested they were poor, from a country where $1000 would have gone a really long way, and I feel like it would be mean to let them know how close they came.
Or give the defector the opportunity to give the money to his poor cousin Vinny.
🙁
Hopefully, since this was the clone game, this also means the winner will benefit just as much.
Might create a bias incentive for next year.
Like in Pan’s Labyrinth?
For your background: I was looking through the answers, and noticed that I answered a question in the opposite way I meant to. I wouldn’t mention it, but this answer was the most extreme in its category, and I’d estimate this answer will only received ~250 like it (though I didn’t count). I assume you do not want people harassing you to edit their answers, but just wanted to let you know that the Lizard Man Effect is in full force!
My bad for trying to fill this out on my phone.
Was it the suicide one? In the earlier discussions several people pointed out that the answers were in the wrong order for that one, so people were likely to pick the opposite end from what they intended.
It was!
I thought it might be interesting to compare the SSC survey responses about the Democratic nominee to the current national polls. I used the averages from Real Clear Politics as a baseline: https://www.realclearpolitics.com/epolls/2020/president/us/2020_democratic_presidential_nomination-6730.html
RCP Candidate SSC Diff% Ratio
28% Biden 6% -22% .2x
20% Sanders 19% -1% .95x
15% Warren 18% +3% 1.2x
7% Buttigieg 10% +3% 1.4x
7% Bloomberg 4% -4% .5x
4% Yang 26% +22% 6.5x
3% Klobuchar 3% =0% 1x
2% Steyer 0% -2% 0x
2% Booker 2% =0% 1x
2% Gabbard 9% +7% 4.5x
Apologies for the formatting, but my first impression is that Yang supporters are tremendously overrrepresented here! My next impressions are that Gabbard gets a lot more love than in the outside world, and Biden gets a lot less.
But maybe this is misleading. From what I can tell, the RCP polls seem to be of “likely Democratic primary voters”, whereas the SSC poll includes everyone. Maybe what the numbers really mean are that Yang and Gabbard are abnormally popular among those who are not likely to vote in the Democratic primary? Maybe someone could do a breakdown of what things look like when cross-referenced by the answers to Party Affiliation?
FWLIW, I plan to vote in the primary, I’m leaning towards voting for Biden, and I said so in the survey.
Gabbard is widely considered in right wing circles to be the scariest potential opponent for Trump, because she has oodles of centre and right-of-centre appeal. Gabbard’s greater appeal here comes from asking not-Democrats.
If an establishment Democrat or leftish Democrat runs against Trump, he’ll almost certainly win again off rust belt votes. The Democrats haven’t yet realised this and are still pretending that the Great Inversion hasn’t happened.
Yang is technocratic Yang Gangery and UBI. This sort of place is his home turf. That one doesn’t surprise me.
Remember that Gabbard is an Assad apologist. If people are getting anything from this blog I don’t understand how they can think someone deciding to go against the grain in supporting a murderous dictator who used chemical weapons is a good idea.
Just a quick comment, hopefully not starting something. Before I even parsed the meaning of the sentence, the word “apologist” made me wary. It’s usually used in contexts where the bias against one party is so strong as to be unquestionable, or used intentionally to suggest that. I’d try to taboo it.
Indeed. The only reasoning I’ve seen for the use of ‘apologist’ applied to her is the idea that “advocating an end to wars designed to overthrow or destabilize a government” can only be grounded in support for the government involved, which is unconvincing. She doesn’t speak favorably of him.
I’m slightly surprised for the high (i.e. similar to population) level of support for Warren. MR’s list seems pretty devastating to me, and I would expect a significant fraction of SSC readership to lean similarly. Is Tyler just biased, did you not see this before, are there good counterarguments, or what? Also surprised by the low support for Biden, who seems like the safe, establishment candidate. I guess he might be a good number two or three on the list for many, though.
Has Tyler critiqued platforms of other candidates?
Much of it seems fairly uncompelling, and I’m no Warren supporter. What’s his point on healthcare anyway? That taxes would have to be raised and Warren is denying this? Sure, maybe. That’s not the important question, the important questions are 1. Will it cost more or less 2. How will these costs be distributed 3. What will the quality be. Whether it’s funded through taxes is of little intrinsic interest.
That healthcare will cost more and/or the quality will be worse? Well given the global experience that seems unlikely.
On colleges- why think that free college would destroy quality? It hasn’t elsewhere. This is a general problem with a lot of this stuff- Americans speculate about this stuff a priori too much, they should consider the lessons of other countries.
Why is he only counting the benefits and not the costs of fracking?
Also, this is a pet peeve of mine:
“Another part of the plan is to pay lower prices — 70% lower — for branded prescription drugs. That is supposed to save about $1.7 trillion, but again focus on which opportunities are lost. Lower drug prices will mean fewer new drugs are developed. There is good evidence that pharmaceuticals are among the most cost-effective ways of saving human lives, so the resulting higher mortality and illness might be especially severe.”
If a sector is doing valuable research, and collecting very high monopoly profits, the solution is not to keep letting them collect indiscriminate monopoly profits, it’s cut the monopoly profits, and give them money specifically tied to research through grants or bounties.
“Another unstated cost of the Warren plan concerns current health-insurance customers: Many of them prefer their current private coverage to Medicare for All. Switching them into Medicare for All is an opportunity cost not covered by Warren’s A$75.78 trillion
estimate. Even if you believe that Medicare for All will be cheaper in monetary terms, tens of millions of Americans seem to prefer their current arrangements.”
These people, at least in the main, fall into two categories: 1. People who have much better healthcare at present then it would be viable for the general population as a whole to enjoy 2. Rubes. Category 1. can’t elicit much sympathy from me- healthcare outcomes (all other things being equal) would be better if healthcare money were spread evenly. Category 2. shows the folly of valorising revealed preferences.
If marginal tax rates are exceeding 100%, that’s a worry sure, although I’d have to look at the precise details.
Overall it looks like “Libertarian disagrees with centre left politician”.
Free college tends to reduce responsiveness to students and other funders since the government provides the funds. Look at the British universities c. 1990 compared to today if you want a good example of how fees help improve a sector.
I don’t see what the intended conclusion is. British universities went through three phases — a small, private sector,
a larger, grant funded sector, and a debt funded sector which was larger still. Where did the rot set in?
It’s not clear to me that reduced responsiveness to students will necessarily result in reduced quality of education.
Depends on the formula used. If it’s subsidized on a per-student level, it shouldn’t reduce incentives for quality. If anything it should increase quality if students don’t care about price as much and colleges switch from competing on price to competing on quality.(At those cost of raising prices.)
I think Tyler’s unwilling to say mass education subsidies aren’t worth it in and of themselves since higher education skepticism would lower his status. So he has to seem like he’s defending the quality of higher education.
What costs? Do you deny that fracking has reduced America’s carbon emissions?
She’s not proposing bounties though. Do you deny that reducing the prices paid and doing nothing else will lead to less drug innovation? A better way of handling the supposedly high monopoly profits is to simply cut the patent terms.
If she’s center-left where’s the left left?
Fracking is a huge source of externalities; I’m not sure how it would be at all possible to deny this. Water pollution is severe, even compared to other methods of oil extraction, and the water usage is also severe compounded by the reality that most fracking is happening in areas with poor water tables to begin with. Air pollution and ground pollution don’t fare much better. It’s got bad effects on other industries – Wyoming’s hunting and tourism suffered because of sharp declines in deer population in the heaviest fracking areas, for example – and studies are finding that fracking-dependent regions tended worse economically than those that were more varied in economic output (of course, this is less fracking specific and more “single-income economy” specific but fracking is often the cause of a single-income economy for a county).
There’s also some credible health concerns. Comparisons of population sites close to fracking sites (<3 km distance) to those further away (3 – 15 km) found that for counties with similar trends pre-fracking introduction, the closer population sites had worse health outcomes post-fracking. Natal health was a particular concern, with underweight and premature births becoming more common in post-fracking sites than non-fracking sites. And water contamination doesn't seem to be the issue here, as this trend held even considering sites that share the same water supply but are at different distances.
Links?
Not true.
That should be accounted for in the price, as they have to pay for said water. People who want to give everything away for free tend to think in terms of everything being given away for free, thus the demands to ban things so those free resources won’t be used up. There’s a better way: market pricing.
The existence of fracking caused all the other sources of income in those economies to magically disappear.
That’s not the important question
Everyone loves to set the narrative, but there are lots of things that are “the important question.” Let me get back to this in a second…
That healthcare will cost more and/or the quality will be worse? Well given the global experience that seems unlikely.
Americans (regardless of party) have a very ignorant view of the health care of the rest of the world. We pay more, but that’s about the only thing that is accurate.
There are many different systems, and right now the US and Germany have systems more similar to each other than either does to the UK.
Politicians have seized on this ignorance. The vague plans are typically “what the rest of the world has” but, like I said, we are already more similar to Germany. The vague plans have all the best things cherry-picked from every system with none of the drawbacks.
Now I am getting back to talking about taxes.
Let’s pretend, just for a minute, that the reason that other countries have better/cheaper health care is that people do not choose their own doctors [1]. The wonks and policy-makers all agree on this. They all agree it is a sane trade-off to make. But when it is time to sell it to the public, it only gets mentioned occasionally, and really just ends up debated in the political op-eds, where we are again told it is a fair trade-off.
The politicians running on the plan get elected, and start implementing it.
But wait! Halfway through Congress writing the plan, there starts a movement called Keep Your Doctor. They talk about how horrible an un-American it is to be Assigned A Government Doctor like this is Russia.
The politicians stick with it for a few days, but it keeps on building up momentum, and vulnerable Congressmen start defecting, and people start challenging “if we can put a man on the moon we can figure out how Americans can choose their own doctors!”
So the ability to select your own doctor is restored. This was, by assumption, the critical component necessary to make the plan work. And we end up with something even more of a kludge than we have right now.
So, yes, it does really matter if people are being told the truth up-front. If we were under single-party rule or a benevolent monarchy, this wouldn’t matter, and our rulers could just do whatever they thought was best. But we don’t have that. The plan needs political buy in, especially on the drawbacks. Deciding “we will debate the drawbacks later” is eating ice cream for dinner, saving the broccoli for dessert, and of course we never end up eating the broccoli and then wondering why we can’t have good health like other people do.
and give them money specifically tied to research through grants or bounties.
Our patent system, whatever other problems it has, actually works and generates new drugs.
Another system might work. But it might not. Who is going to decide what counts as an actual breakthrough drug?
[1] No, this is not the reason. But I am choosing it as an example because we can all understand it.
Let’s pretend, just for a minute, that the reason that other countries have better/cheaper health care is that people do not choose their own doctors [1]. The wonks and policy-makers all agree on this. They all agree it is a sane trade-off to make.
I don’t think that’s true. I thought it was because Americans use more healthcare and pay doctors more. You can see this clearly in the health care share of employment:
https://i0.wp.com/randomcriticalanalysis.com/wp-content/uploads/2018/01/oecd_2015_employment_share_gdp_share.png?resize=549%2C549&ssl=1
Of course it isn’t true. I said it was pretend. That [1] was a footnote where I said it wasn’t the reason.
I really really wanted to talk about “we must be willing to address the drawbacks of any new healthcare system, and acknowledge what we are giving up, and fuck yes there will be drawbacks for everyone not just drawbacks for other people.”
Our patent system, whatever other problems it has, actually works and generates new drugs.
Another system might work. But it might not. Who is going to decide what counts as an actual breakthrough drug?
A big part of the problem is that the costs of getting a potential drug through clinical trials is in the tens to hundreds of millions of dollars. Universities already generate lots of potential new drugs, even if two-thirds of their research doesn’t replicate. But giving hundreds of millions of dollars to the organizations (corporations) capable of taking a potential drug through clinical trials is going to look an awful lot like corporate welfare.
Of course, we could reduce the amount of testing required before releasing drugs to the public, but good luck getting buy-in on the drawbacks (thalidomide) of that change.
@Edward Scizorhands, Apologies for misrepresenting your point.
@Ketil
Biden’s mental state is deteriorating rapidly, much more so even than Trump. Policy wise he’s the most favorable candidate for centrist type Democrat voters, but mental capability matters, and it’s not unfair to say the guy is losing his mind.
This is a little brain melting, and a little cruel, but hilarious:
https://soundcloud.com/episode-one-868768631/113-joe-biden
I’ve yet to be convinced Biden is really senile, but I’m in the middle and it’s hilarious.
What the hell am I listening too?
Biden comes across to me as both condescending and ignorant, which make me despise him as a person more than anybody else running at the moment, even though his policies are … less opposed to my values than many.
If others get the same vibe, it’s possible that they react the same way, more viscerally than on policy at the moment.
Warren has a lot of appeal to relatively well-off, educated liberals: see NYT effectively endorsing her for President. Her picking up a few points in the SSC crowd isn’t all that surprising through that lens.
As for her policy, yeah, sure, but you could say the same about Yang, and we all know why Yang is so popular on SSC.
“Is Tyler [a conservative libertarian literally funded by the Koch brothers] just biased?”
I do not like Warren one bit, but the question answers itself.
Would you say the same for a liberal literally funded by Soros?
When the specific question is about bias? Absolutely. I’m not saying Cowen is wrong. (Though he is wrong.) I’m saying that, in regards to a question explicitly asking about his priors, he has priors and context that would be relevant to the person asking.
I think dismissing someone because of how they’re funded is basically just a fallacy. It’s some Bayesian evidence, but there’s no reason to rely on it. It’s not like we have to speculate about how good his arguments are; they’re right there.
It doesn’t seem surprising to me at all. Tyler Cowen is a libertarian, and people inclined to vote for Elizabeth Warren would be more likely to be paced on the far end of the “anti-libertarian” spectrum of US politics. As one of those people myself, I walked away with more positive feelings about Warren after reading his piece.
As for the demographics of Warren/Sanders/Biden at SSC, I’m not terribly surprised at that either. Biden does atrocious with young people (and the average age at SSC seems to be roughly ~30), and less well known but equally as true is that support for Warren is strongly correlated with high incomes (which SSC readers also tend to skew).
If anything, I’m surprised Sanders has the high level of support that he does in the survey (I’m blaming a “silent majority” of left-oriented lurkers- I recall a previous survey showing a disconnect between average SSC reader ideology vs. SSC poster ideology)
Yes. When I looked at political affiliation of all survey takers, 62% were either Liberals or Social Democrats. Whereas for heavy commenters that was only 34%.
That sounds like it might have been mine. There’s a massive spike in the middle-left of the political spectrum that almost entirely disappears when looking at frequent commenters. I’m planning on digging into the data pretty deeply this weekend, and the primary support could be an interesting question.
Thanks for linking, very interesting results.
People who supported Sanders and didn’t identify as Marxist in the survey: anyone want to defend Bernie’s policy of national rent control?
https://twitter.com/BernieSanders/status/1219044653414146053
Of course one can support a candidate without supporting all their positions. But why else support him if you aren’t on board with all this far-Left stuff?
Some forms of socialism are explicitly non-Marxist.
“Marxist” is one of those nomenclatural asymmetries where far more people get called the term than who self-identify as it, e.g., Obama gets called a Marxist by Republicans, but even a socialist such a Sanders doesn’t necessarily self-identify as such. (I’m pretty sure, at least)
If Scott wanted to simply figure out how many of his readers are socialist, he could always ask this simple question. Instead he has us choose between “Marxist” and “Social democracy”, which I have to say is a bit of a tell as to what circles he’s been running in over the past few years.
> Some forms of socialism are explicitly non-Marxist.
I’d like to second this. Indeed, there are many forms of socialism and socialist thinking which aren’t Marx-related. It seems to me that Marx managed to get a lot of brain-space both because of his class-conflict approach which became popular in certain circles of academia, and his association with Lenin/Stalin/Mao.
>Instead he has us choose between “Marxist” and “Social democracy”, which I have to say is a bit of a tell as to what circles he’s been running in over the past few years.
Yeah that’s kinda like asking “Are you a white supremacist or a neo-nazi”, it’s not how the group would label themselves.
NSA regulation
It’s clear from Yang’s campaign material that he is familiar with rationalist ideas, although I wouldn’t call him a rationalist. I supported him because of that and because I think he is the candidate that is best prepared to respond to foreign interference in the general. He doesn’t seem to use rationalist ideas in his campaign though, which is a shame because the US presidential primary process is driven mainly by tradition and groupthink. He could have gotten a real edge had he used data driven methods to efficiently secure a sufficient number of delegates, rather than just throw all his resources into the early states like everyone else. I hoped that him not qualifying for the debate would have been a wake up call, but he hasn’t changed his strategy and I am now supporting Sanders.
I’m confused as to why you think not concentrating on early states would have been a winning strategy. The traditional reason for doing so seems sound: voters in later states are likely to perceive the primary as a contest between the candidates who did well in the early states, so they won’t vote for a candidate who didn’t even if they prefer his policies. Basically the early states are an opportunity to create Schelling points.
IA and NH award 1.63% of the total number of pledged delegates. That’s not a significant number, especially since they are not representative of the nation as a whole demographically. They main reason they are important is that the candidates agree that they are important. A candidate with fewer resources should defect in a public manner, allocating their resources to other areas. They can then plausibly blame their loss on their defection, and gain an advantage by allocating their resources more efficiently. This advantage should outweigh the loss of prestige that comes from losing the early states, especially since it is clear to everyone that it is a strategic retreat rather than a rout.
It’s pretty easy to point out why it would not. It’s a national election, so not being competitive in any territory cripples you. Moreover, the support is correlated across the nation and is perceived to be correlated. Media and voters will use early state results as guidance (the former for coverage priorities, the latter for strategic voting). Good results in IA/NH give you disproportionate returns for the relatively small expenses necessary to achieve them. And if you can’t achieve them, it’s a good hint you probably would’t have succeeded elsewhere anyway.
This primary season, we have Bloomberg trying the ignore-early-states strategy, with a lot more plausibility than someone of Yang’s status could pull off. We shall see the effects in about a month and a half, and while we won’t be able to compare his results to the counterfactual, we will find out if his bid succeeds. I strongly suspect it won’t.
The main reason they are important is that the voters, the media, and the donor base believe that they are important. If e.g. Biden and Sanders agree to sit out IA/NH, but Buttigeig and Warren don’t, then an awful lot of people are going to look at the returns and say “I had thought that Buttegeig didn’t have a chance and was going to hold my nose and support Biden just to make sure neither of the wacky socialists wins, but now that Mayor Pete looks like he’s got what it takes…”, or vice versa. And an awful lot of free advertising in the form of press coverage, that would have gone to Biden and Sanders, instead goes to Buttigeig and Warren.
However irrational you may insist it is for voters, donors, and reporters to behave this way, they will behave this way and candidates need to account for that in their strategy.
Has anyone speculated regarding Yang potentially running again, in 2024?
In my view, “maximize total delegates by running roughly equally in all states rather than concentrating on the early ones” is an absolutely horrible strategy to win the nomination (for the reasons you discuss) during that specific election cycle, but plausibly a very good way to boost one’s fame, profile, and status for the long term.
This is basically what Bernie did in 2016, isn’t it? Less by deliberate strategy but more by being the only person willing to continue opposing Hillary long enough to continue to earn delegates and get his name in the press.
If Yang’s goal is less “Be President in 2020” and more “Be President… eventually” then running in all states, refusing to drop out, and trying to acquire as many delegates and get as much attention as possible, seems like a plausibly good strategy to me.
I wonder if his medium-term goal might be to become a Governor or Senator in the 2022 election cycle. He’s a long-time resident of New York State, which has an incumbent Democratic governor (Andrew Cuomo) who is eligible to run for reelection (I don’t think NY has term limits for Governor), but is plausibly vulnerable to a primary challenge. Senate is less likely, since the incumbent Democrat there who would be up for reelection (Chuck Schumer) seems likely to run and very unlikely to lose a primary.
I am not at all surprised by support for the vaguely libertarian UBI guy amongst readers of a blog that writes positively about the UBI and is ensconced in the grey tribe. Among the candidates, there are multiple people splitting each coalitions, but he has this one all to himself.
This is a good point. I’m a registered Republican that is going to vote someone other than Trump in the Primary, but it will be the Republican primary, not democrat. I chose Gabbard, for the survey question, since she is the most similar to my views, but that doesn’t make her many friends among Democrats.
Biden seems universally much less popular online than offline (e.g. it’s rare to see a Biden supporter on Twitter, Reddit, Tumblr etc.) So I’m not at all surprised that his support is low here as well. If anything, I’m surprised it’s as high as 6%!
(Side note: I’d be interested to hear the take of anyone who put down Biden, since I don’t get to hear that viewpoint much.)
The two explanations I’ve seen proposed is 1) it’s an age thing and 2) it’s low-information voters because they’ve vaguely heard of him as working with Obama (many of whom will change their minds as information eventually trickles down to them). Neither of these really convinces me, though.
He gets his strongest support from old black voters, who don’t tend to spend time on Twitter, Reddit, Tumblr, etc.
Yep, that sure looks like the public key of a defector-against-clones to me.
What was the public key of the cooperator? The knowledge that you played has some value. I wasn’t sure whether a losing played would ever know that they lost, and that effected my decision to cooperate or defect.
Agreed. If nothing else, I’d like to know whether my psych profile indicates I should be waifuing gwern.
I cooperated but considered defecting because I figured my clone would be more likely than average to defect. I wondered for a moment if I’d lost, then realized no one who would call Gwern his waifu would be my clone.
I too would like to know the public key.
Scott has posted above he’s not going to release the other player’s key, because they could probably have used the money and he doesn’t want them to feel hurt.
Did anyone else have to double back on the “visualize a red star” question upon realizing that “in the skies above Pern” was not included in the option list?
No, but I’m so embarrassed I didn’t that disappearing between seems a good option.
I basically saw this: https://www.skyandtelescope.com/wp-content/uploads/Carbon-R-Leporis-Brimacombe_V2.jpg
Interestingly, SSC survey-answerers seem much more likely than the average American to have at least contemplated suicide.
To be fair, we’re more likely to have contemplated just about anything.
Probably not a surprise. The demographic skews young and male, which is a high-risk group for suicide. Less certainly, because I can’t substantiate at least one key link here: SSC readers tend to have higher levels of mental illness than the general population (unsurprising really: Scott writes lucid and interesting posts about it and so attracts readers researching such things for themselves) and that’s another high-risk group.
My understanding is that older rural males are actually the most suicide prone demographic. Partly due to access/familiarity with guns and fewer QALYs ahead of them.
Well, @Watchan did say “a high-risk group” not “the highest risk group”.
I mentioned this earlier on reddit, but I feared this result would be somewhat corrupted due to the options order not being coherent with the previous questions. Looking at the results now, I think this might be true.
Isn’t iq correlated with suicide?
Let’s find Mr. “gwern is my waifu and paperklipot maximizer” and punish him.
“Wishes suicide attempt succeeded” is more common than “glad suicide attempt failed” 🙁
The order of the answers was the opposite to the several previous questions and thus counter-intuitive, so the number of people who answered “Wishes suicide attempt succeeded” might be inflated.
I…guess, but that sounds like a question where someone would pay more attention to that sort of thing (as someone who paid quite a bit of attention to that question for those reasons).
As someone who just looked through the results and saw my answer on that question, I can definitely say that I put the wrong response.
For someone who hasn’t ever considered suicide in any manner, I can say that given the category of questions it was with, I didn’t pay any closer attention to that question than the rest of them, because it’s not any more notable to my lived experience than experiencing OCD or depression.
“I figured this was okay because nobody’s responses could be connected – ie you could see that one person put their age as 83, and another person put their country as Canada, but because the table order wasn’t the same you couldn’t link these together to form a coherent picture of an 83 year old Canadian.”
Are you sure this is correct? People were saying in the comments that the order was the same, so you could put together all the answers for each individual person, including enough demographic information to perhaps identify them uniquely, plus their answers to the sensitive questions.
Also, the public and private keys were both visible. This partly contributes to the first problem, in the case where the text in the key identifies the individual and/or reveals sensitive information. It also creates a problem where private keys aren’t private: before you removed long-answer questions from the aggregate results, all the public and private keys were displayed, and I believe they were in the same order. So you will probably get lots of people emailing you and quoting the private key of the paperklipot guy (unless he happened to take the survey very shortly before it closed, so the people downloading copies of the public/private key mappings didn’t have his included).
This is most definitely untrue. The order of respondents varied from page load to page load, but within a single page load each question listed the answers in the same order. (We can’t guarantee this by looking at most of the questions, but we can state it with total confidence for the public/private phrase pairs, because those frequently matched each other in a recognizable way.)
However, the summary view didn’t show the full list of answers; it showed a truncated list of a few hundred answers. So it’s possible that this guy was never exposed. Depends how heavily people polled the phrases, and also on whether this guy responded before or after Scott locked down the summary view.
Another note on correlating the answers to other questions: since answering questions wasn’t mandatory, it would be pretty difficult to guarantee that you correctly matched an answer on one question to an answer on another question. Missing answers are less of an issue for the public/private phrases, because it is pointless to submit one without the other one.
It shouldn’t be hard to find the guy who lives at the 83:rd latitude.
Not everyone answered every question – some questions got hundreds or even thousands more responses than others. As soon as there is even a one person difference in number of respondents, you can’t count rows any more.
SSC has steered many readers towards vegetarianism, and steered even more towards eating beef rather than chicken. The LifeEffects2 column contains many interesting comments about this. (Incidentally, I can’t see the corresponding question in the survey questions webpage that you linked to.) Row 4592 is the most interesting testimony for how deep this goes:
> Personally farmed, slaughtered and butchered rabbits to replace chicken, thus replacing the most ethically negative source of meat with a similar culinary meat, that I could ensure had a net positive life with minimal suffering during slaughter. Also, doing it myself ticks a major virtue ethics box.
However, several readers (row 578, 1374, 2269, 4246, 4410, 4641, 5893, 5998, 6222, possibly 5543) report an opposite effect. It’s good to know that the same blog contains arguments for both sides of the same issue.
Also, several readers mention your 2014 article “Nobody Is Perfect, Everything Is Commensurable” “https://slatestarcodex.com/2014/12/19/nobody-is-perfect-everything-is-commensurable/” in the LifeEffects2 column. I guess I know which article I should read next.
There are 33 distinct rows that were submitted more than once with exactly the same values. The highest frequency are the identical rows 5923, 5924, 5926, 5927, 5930, 5932, 5933, 5935. Some of these are probably the same reader submitting the form multiple times. Most of these rows, quite rationally, chose cooperate with their clone.
I’m really interested to see the groupings on the police killing/being killed questions. There were people who thought the police killed tens of thousands of Americans in a year? Somebody said “100,000” and some of the answers for police killing unarmed black people were in the multiple thousands or tens of thousands. Were these serious answers? Do people really think the police are killing tens of thousands of people, many unarmed, a year?
For reference, here’s the WaPo Police Shooting database. It was 933 people total in 2019, and 10 unarmed black people. Here’s the Officer Down Memorial Page. It’s 47 for shot and killed, but there were more if you count assault and vehicular assault.
I said ~800 people killed by police because that was what was stuck in my head from previous attempts to parse out the “gun deaths” statistics gun control advocates will cite*. I can’t remember if I said 30 or 40 for unarmed black people, which was a guess based on hearing about maybe 5-10 a year and figuring two or three times that many wouldn’t have made it into my filter bubble. But apparently not. For police officers I remembered hearing statistics that about 100 were killed every year, and I assumed ~80% would have been shot (the rest did not pay enough attention to the Surviving Edged Weapons video), so I think I said 80. But it’s more like ~100 total DIED in the line of duty, not necessarily KILLED. So like 16 of those are heart attacks and such. If you just look at things that look like intentional killings/murders, about 80% gunshots doesn’t seem that far off.
* GCA will say there’s ~30k gun deaths in the US, but about ~20k of those are suicides, which is a related but different problem, and (I thought from last time I looked) it was ~800 for “shot by police,” but we generally hope those people deserved it. But by including every person killed using a gun, even if they were using it on themselves in the statistics, it makes the US sound more dangerous than it actually is.
And there are still many large guesses for “how many people total were shot and killed by US police in 2019” even if you consider only the readers who claim to be residents in the U.S. (The section of that question starts with “Please answer only regarding police officers in the country you listed as your residence at the beginning of the survey”.)
These were the questions I was most excited to see responses for. Somewhere in Australia there is a 20-ish female law student who thinks the US police kill half a million people per year and that half of those (a full quarter million people) are unarmed blacks. The rest of the answers seem genuine (e.g. unlike 1E16 guy). There’s also a 20-ish male medical student, location unspecified, that also thinks US police kill half a million people per year, though he only thinks 150k are unarmed blacks. Similarly, the rest of the answers seem genuine.
This is a really off-the-wall matchmaking criterion, but the error is so large and so specific that there has to be some potential here. Unfortunately they’re both already in relationships (perhaps with each other?!).
A big chunk of on-duty deaths are car crashes. Cops spend a lot of time driving, sometimes at high speeds, and driving is dangerous.
Just realized I botched the link to the Officer Down page. Of the 134 Line of Duty deaths in 2019, 23 of them were auto crashes.
Damn I missed the unarmed qualifier in there. Now that I’ve seen this I don’t know what I’d answer but it would def be lower than what I put (I think I put around 500).
I’m more interested in the proportions than the raw numbers (since my own ability to estimate absolute numbers is pretty rubbish). There seem to be quite a few people who think that fully 80% or 90% of the people killed by police are unarmed black people (as opposed to about 1% from the data you quote). My own estimate was a too-high percentage, but nowhere near that high.
As one of the people who said in the 10s of thousands, I came to my conclusion from a flawed Fermi estimate. I assumed 1% of the population would be police officers (possibly too high) and that on average 10% of the force would have fired a gun that killed someone in the past year (also possibly too high).
Your first assumption is high by about a factor of two if you count just the US labor force, but it sounds like you’re including the children, retirees, etc, and that throws you off by another factor of two.
Second assumption, yeah, way to high. 10% of police officers firing a gun that kills someone in their entire career would be closer, but still high. Police officers are not gunfighters.
I suspect the question on auras around lights may partly (mostly?) be selecting for whether someone needs corrective lenses, since it specifically asked about what lights look like without wearing glasses or contacts. Maybe this is because my vision is particularly bad, but the fuzziness of my vision when I’m not wearing my glasses makes streetlights and other things look like auras/halos, even though with my glasses I don’t see anything like that. Was that the intent of the question?
As someone with an eye disease which gives me halos around and high contrast light, who has never used psychedelics, I was also unsure about these questions.
I recently discovered that astigmatism, even if corrected, can lead to poor experience with reflex sights for firearms.
I’m really interested in the results of the “red star” question. I myself rated it at a one, because that felt the most accurate to my experience.
I was under the impression that under true aphantasia, there wouldn’t be any visualization at all (A litmus test for this, under my understanding of the condition, would be to tell someone to imagine a ball, and then ask them what color the ball they’re imagining is. A true aphantasic would be confused by the question “There’s not really a ball”, unless my understanding is wrong)
That said, this seems to suggest that most people are far more visual in their mental processing than I am. (I knew that my psychedelic experiences seemed less visual than the average, but didn’t think much of it at the time) I am quite curious as to the further ramifications of this, because I am still convinced that I pictured a star, it was just closest to “1” than anything else.
This one is a weird one for me as well. If I’m really honest, I would answer a 1 as well, but I wonder if it’s just that the question is a bit ambiguous. Do people really “see” the red star when they close their eyes?
In my case, I can very vividly imagine it, but it’s not the same as seeing. Like, I’m sure I can imagine color, I can visualize in my head a blue or green star, but it is nowhere as vivid as actually seeing them with my eyes. Also, when buying clothes or furniture, I can easily think in my head if its color will match with my other clothes/furnitures. Also, I have always been a very visual learner, and use “spatial imagination” daily at my job as an engineer. So I find quite hard to believe I have some kind of aphantasia.
Like, the experience of thinking of the star with my eyes closed is very similar to thinking of the star with open eyes, the star isn’t really in my visual field, but in another space in my head. I can then kind of force it to superpose with my visual field, but then I don’t really get any visual or color qualia at all. Do people really see actual red qualia if they do this, as vivid as the red of the real star in your screen?
I can also imagine very complex and colorful scenes, spaces and objects in my head, but I would still answer 1 in the survey question, which seems weird.
I have no understanding what you’re referring to at all here. I’m staring at my desk, trying to visualize a pencil placed on the desk, but can’t seem to superpose it on the world at all. It’s either “pencil in the void in my head” or “pencil on the copy of the desk in the void in my head.” I get the idea that you can kinda map the mental image onto the real world in a way that I cannot?
When I imagine a sound, I would describe that as subjectively on the same stratum as my actual hearing, maybe that’s analogous?
How good’s your spatial awareness?
Just wondering.
Never had any real problems with it.
When I was back in college, we all had to take this spatial reasoning test, with those failing it being sent to some class to try to teach it to them over the course of a semester. The test was an easy 100 for me, but I’d say a good 10-15% of people ended up in that class.
I’m in a similar boat. I answered 6 initially, then I decided that it was probably asking what you could force onto your actual visual field (which is why you have to close your eyes). That’s still about a four for me if I really try, but more complex shapes are very difficult and I can hardly do it at all with my eyes open. I would consider my spatial awareness to be better than most (I have done some combat sports in the past and found, surprisingly, that they’re not much more difficult to do with my eyes closed), but it’s always a visualization that isn’t the same thing as what I sense as the signal coming from my eyes.
Actually seeing red qualia is a hallucination. No one would mistake their imagining for actually seeing something. Your experience is totally normal.
I agree with this, generally. It’s pretty straightforward, if difficult, to fool yourself into “seeing” a star shape in the random noise your retinas send your brain, if you look hard enough and want to see it. It’s almost impossible to convince myself that it’s red.
+1, this is pretty much what I’m confused about. I’m not convinced this isn’t (mostly) an argument about words. Like, for me to believe that people see #6 would suggest that you might be able to fool them by placing a led-lit red star in a dark room, and have them unable tell if it was their imagination or a real thing. But surely visualization isn’t so… vivid?
Does closing the eyes matter, other than removing visual clutter to distract your attention from the “minds eye”?
Correct. The mind’s-eye is like a memory. If you have never confused a memory for real life, you’re not going to confuse a visualized red star for a real one.
Closing the eyes doesn’t really matter except removing distraction.
I have had a vivid mind’s-eye experience indistinguishable from reality once before, though this isn’t usual for me.
Nikola Tesla’s autobiography reports him regularly experiencing such vivid imagery, which was apparently distressing in his youth before he understood it. (This may represent a different kind of experience, but it does seem like how such a thing would manifest.)
I pictured a red gas giant with a solar flare and was unsure what answer to pick.
I enjoyed filtering the Excel sheet to find my row and seeing the similar people who also had many other answers the same as mine, but some different.
2264 of 7339 (30%) were Computers (practical) for Profession. I didn’t realize it was so high.
Are you kidding? I’m amazed it’s not even higher!
I was astonished to see that roughly 40% of the people here have JD degrees. That can’t be right, can it?
5% lawyers, 35% former juvenile delinquents who weren’t lawyers?
on the allergies front, i’m one of those people who have near constant low level sinus issues. i don’t think these are “allergies” so much as sensitivity – my sinuses are a lot more sensitive to environmental irritants than most people’s.
it’s very rarely a huge problem for me but being constantly irritated means living with a cluster of symptoms that are always at a low level – including sinus drip, headaches, minor sleep disturbances, higher emotional reactivity, lower concentration / slight mental fog, etc.
these are things that i do well enough with if i can manage my exposure but, say, travel is a problem as hotel rooms tend to be pretty stinky and i don’t stay asleep very well in them. so, i don’t travel as much as i would like.
the thing is, i have never seen this studied. is it common? is it correlated to other chronic (or other) health issues? i have always been curious but have never seen anyone asking about it. i’d assume no, but the health effect of sinuses being irritated for decades is something i can’t just brush off.
Since sometime in the summer, I’ve been using air-tight mattress and pillow covers, to insulate myself from dust mites. Also, I had the sofa steam-cleaned. Occasionally I still need to blow my nose, but it seems my chronic sinus problems (which I had since childhood) have dissolved. I haven’t even had a cold since the change, though I do expect to catch one sometime this winter.
If any of the 20 people living in Spain who said they have never attended a Meetup because they “don’t know of any or can’t make it to any” is reading this, I would really like to know whether it’s because of the former or the latter. This is aimed especially at the 10 who identify as (at least “sorta”) EA. Our EA chapter in Madrid hosts activities like workshops and talks almost every week. We’re actively trying to expand our group of core members and we’d love for you to get in touch! If you’re interested, please reply to this thread or shoot me an email at javier[dot]prieto[dot]set[at]gmail[dot]com
I have created two Google Docs that have all of the long answer explanations of paranormal or mystical experiences. I was getting a headache trying to read them in an Excel file: this format should be easier for people to read. I was particularly curious to see what kind of experiences SSC readers would have had. Browsing through the list is interesting: if I have time I may read them all the way through and post my highlights to the next open thread.
If you are similarly interested in others supernatural experiences, here are links to the two docs below.
Paranormal experiences:
https://docs.google.com/document/d/1i5QVW4OumnNk-dzgLVjKO99S5JiMCo8bc8BhoJRZGWU/edit?usp=sharing
Mystical/Religious experiences:
https://docs.google.com/document/d/17jrRezpP-lFht5xEuebD2RyP8GaTKt9qpMFUtfisRkE/edit?usp=sharing
I see some people being surprised at the ratio of “I have attempted it in the past, and wish the attempt had succeeded” to “I have attempted it in the past, but I am glad the attempt did not succeed” for the suicide question and speculating that a lot of it may be due to question ordering. As one of the survey participants who deliberately answered “wish the attempt had succeeded”, I feel some responsibility to make it known and maybe take questions if people have them. I confirmed that I didn’t make a mistake on the question by looking at the public data download. This is a temporary account because I don’t want to link my name to that answer. I will decline to answer questions that in my judgment expose too much information about which exact row was mine.
I guess I should add from the beginning that I am not currently intending suicide, and that my response was based mainly on my felt emotions and not on a conscious analysis of tradeoffs. I don’t think I consciously chose which interpretation to use at the time, but looking at the available answers, the second half of the other similar answer uses the term “glad” which I read as an emotional term, so I agree with my previous interpretation.
I am sorry to hear that you have ever felt that way. Would you say that your feelings on this matter are pretty stable, or do they fluctuate, such that you are sometimes glad that the attempt did not succeed? If the situation is more like the latter possibility, was your response based more on your feelings at the moment, or on some sort of average across time?
They fluctuate, but not to the extent you describe. Mostly between “I should’ve died” and “…” numb/neutral. There have been periods where I felt “maybe being alive now is actually a good thing” in the past, but those periods all ended with losing most or all of what I thought I had gained, not by unrelated losses coming in sequence but in ways (betrayal, foundational errors revealed, early mistakes boomeranging tenfold or hundredfold, etc) that made my positive interpretation of the preceding time seem absurd. I would say my response was based on a recency-weighted average of the last several months, but that all measurement curves that would change my answer would look clearly absurd and chosen for that purpose.
I’m sorry to hear you feel that way; may your username check out.
Are you saying that you still emotionally want to commit suicide? Or do you emotionally (as well as rationally) shrink back from the thought of attempting it again, even if you (emotionally) wish you had succeeded in the past?
Emotionally, I shrink from the idea of another attempt, primarily because of the image of gods holding me responsible for it and being unhappy with me afterwards. Also because of the risk of failure or being intervened on while planning such a thing, which again seems to mainly be about getting in trouble. A semi-recent discriminating example is that while discussing the fallout of a broken interpersonal situation a while back, the possibility came up that if things had gone differently, one of the other people involved might have raped and murdered me. That idea sounded very appealing in the moment, because if that other person had killed me instead, I wouldn’t have to deal with life anymore and it wouldn’t be my fault.
My semi-conscious immediate belief is that the last few decades of my life have been of such overwhelmingly negative total utility (from my own suffering and lost goals and from resources wasted by and harm done to others) that if I work at it for the rest of my life, I could plausibly raise the value of my life closer to zero, but probably not above. So, it might be worse for me to die now than to continue living, but it would be better for me to have succeeded at cutting things short twenty years ago to avoid the badness continuing for so long in the first place.
My conscious meta-belief is that I can’t trust my immediate belief, because it’s likely that it has been warped by severe developmental trauma and the ensuing depression and other problems. I finally saw a new psychiatrist recently, and I am trying a new medication among other potential fixes to see if this will help curtail the damage enough to live a life with some positive utility over time. This meta-belief influences the immediate belief as well, by being the main source of the “if I work at it for the rest of my life” part; if I believed I were completely out of treatment options, then I might also believe consciously that I should attempt suicide again.
That’s an interesting point that wishing that one had died at a prior time is entirely compatible with not wishing that one would die now, due to large negative utility that was incurred between then and now. Thank you for sharing such personal experiences. I hope that you can get to a point where your total valuation of your life is above zero.
I was interested in looking at how responses varied by US state of residence for the American respondents, but that’s one of the responses that was removed. Was this intentional? Too identifying?
Even though readers from the US have only marginally higher IQ scores than readers from outside the US, they earn vastly more (even after taking into account differences in PPP). Does this provide further support for Garett Jones’s Hive Mind hypothesis?
Most of those outside America are in Europe, no different in IQ. It provides support for the importance of the exorbitant privilege and not having Euro-style economic policy.
Not really — I don’t want to move to the US even though I know I could almost certainly get a several times higher salary.
Yeah, I have a congenital heart defect and I work as a cybersecurity consultant, and I would need at least triple the pay I get now to even consider working in a country with America’s healthcare and educational system.
In the US, most salaries are given before most taxes are paid (meaning out of that dollar amount, the employee would pay their national/state/local income taxes, payroll tax, etc). In Europe, as I understand it, the convention is to give an income post-tax, the amount left over after most of those taxes have been paid. That could account for some of the difference. I employ people of roughly the same position and skills/experience in both the US and the UK, and the market rate for those people in the US is indeed higher than the UK, but not vastly so. In my industry, maybe 20% higher. In my experience, the market rate is higher because of more competition for that person’s labor in the US versus the UK.
It just means the US is wealthier and has higher salaries for many professions. In as much as that provides support for the Hive Mind hypothesis, it should have already been taken into account, so it is not “further” support. But as @Alexander Turok points out, the US has higher salaries for many professions even compared to other countries with similar IQ, so it fails to support that hypothesis.
I occasionally skim this blog and marvel at the strangeness of views here for the most part. The survey helps a lot, not only the views are strange (to me). Definitely a self selected group, far from random even given the education level.
One definite like is that the posts are polite, a rare thing nowadays.
BTW I am very conventional but also skeptical of most claims. I am old a have time to consider and was fortunate to get an education which stresses analysis (STEM). I very rarely post something, usually skepticism in the abstract and get no responses at all. Belief seems to be a very prevalent mode for arriving at the truth.
Does the cash prize invalidate the results? I seriously considered giving untrue but more generically “ethical” answers in order to get a goody two shoes clone so that they would be easier to dupe out of money when I defected.
Does the google survey log IP addresses of participants?
… only to discover that the person you were paired with was so similar to you, they tried the exact same strategy, thus pairing you with a defector.
Will there be/is there some way to compile the data into a bar graph or something?
I feel quite stupid yawning reading statistics about yawning on reading about yawning I already yawned upon…