Please Take The 2020 SSC Survey!

Please take the 2020 Slate Star Codex Survey.

The survey helps me learn more about SSC readers and plan community events. But it also provides me with useful informal research data for questions I’m interested it, which I then turn into interesting posts. My favorite was 2018’s Fight Me, Psychologists: Birth Order Effects Exist And Are Very Strong, which I think made a real contribution to individual differences psychology and which could not have happened without your cooperation. But last year I also got to debunk a myth about how mathematicians eat corn, fail to replicate supposed dangers of beef jerky, and test a theory of how fetishes form. I expect this year’s research to be even more interesting.

The survey is open to anyone who has ever read a post on this blog before December 30 2019. Please don’t avoid taking the survey just because you feel like you’re not enough of a “regular”. It will ask you how much of a “regular” you are, so there’s no risk you’ll “dilute” the results. The survey will stay open until mid-January, and I will probably be begging and harassing you to take it about once a week or so until then.

This year’s survey is in two parts. Part I asks the same basic questions as previous years and should take about ten minutes. Part II asks more questions on research topics I’m interested in and should take about fifteen minutes. It would be great if you could take both parts, but if 25 minutes sounds like too much surveying to you, you can also just take Part I.

As always, the survey is plagued by fundamental limitations, poor technology, and my own carelessness, but a couple of things to watch for:

– Once you click a box on a Google form, you cannot un-click it – i.e. you can change your answer but you can’t unanswer the question. If you click a box you didn’t mean to, please switch your answer to “Other” if available; if not, then choose the most boring inoffensive answer that is least likely to produce surprising results. I realize how bad this is but there is apparently no way around it.

– By default, all responses will be included in a public dataset for anyone who wants to analyze them. Your responses will obviously not be attached to your name or any similarly blatant identifying information, and this year I’m going to further bin a couple of especially identifiable categories like age, but if you’re the only supercentenarian Mongolian reader or something, you might still be identifiable. There is an option to make all your responses private; if identifiability bothers you, feel free to check it and you will not be included in the public dataset.

– Due to poring over a 5000 entry spreadsheet not actually being that much fun, I am not up for changing your answers after you submit them. Please do not email me asking me to do this. This includes your answer to the privacy question. Please figure out whether you want privacy before taking the survey.

That having been said, you are all great, and I super-appreciate any survey-filling-out you are willing to do. If you can donate about a half-hour, I hope I can pay you back in interesting findings and useful crowd-sourced life advice. I also plan to pay up to two randomly selected respondents back with a large monetary prize (with some caveats I hope you’ll find fun, see the last section for details). So:

Take the 2020 Slate Star Codex Survey here!

This entry was posted in Uncategorized and tagged , . Bookmark the permalink.

408 Responses to Please Take The 2020 SSC Survey!

  1. Nnotm says:

    > It will ask you how much of a “regular” you are

    One question that might be useful is a rough estimate of how many posts one has read, e.g. I’ve read my first post here more than two years ago but I haven’t read as many posts in total as one might suspect from that.

  2. kupe says:

    The comment section on surveys are always rough :/

    Side note: How about a patreon goal for getting some better survey software?

  3. Edward Scizorhands says:

    When is the deadline to submit?

  4. DouglasGomes says:

    I think you forgot mentioning negative spiritual experiences, and I certainly did ignore mine when answering the relevant questions for this.

  5. Reasoner says:

    I wasn’t sure how to answer e.g. the Trump question. I think he’s plausibly the worst president in history, but I also think “Trump derangement syndrome” is an actual thing which I don’t have. So shall I give him a 1 or a 2?

    • anonymousskimmer says:

      If you refer to the concept of “Trump derangement syndrome” for your answer you aren’t answering the question on “your opinion of Donald Trump”, but instead answering a hybrid question that was not asked. This would be like changing your answer because a family member is favorable of Trump. The question isn’t asking whether your family members are favorable to Trump, or whether random other people reflexively oppose Trump, but about your opinion of him.

      Saying that you think he’s plausibly the worst (of 45) presidents places him in the bottom quintile of presidents. If you choose to answer the question as to your opinion of him in terms of your opinion of him as president, then you’ve pretty much indicated he merits a “1” (bottom quintile, or -2.5 SD, or least value log-scale).


      I personally don’t understand why people believe in “Trump derangement syndrome”. I can kind of understand the concept in terms of reflexive tribalism. But then why call this Trump DS instead of Red DS? And do these people believe there is LGBTQ derangement syndrome, Evangelical DS, Lizard people DS, (Anti-)Abortion DS, Qanon DS, etc… or if you only consider it DS because Trump is an individual, then AOC DS, Pelosi DS, McConnell DS, etc…?

      I have not seen those terms used, so assume “derangement syndrome” is a particular kind of opposition that (in the minds of the people coining and using this term) only happens in opposition to Trump. Why privilege Trump by claiming that vociferous opposition to him is the only thing worthy of the “derangement syndrome” suffix? What is in their psychological relationship to Trump that merits this privilege? I can posit guesses related to evangelical eschatology on persecution, and Trump’s own “everyone is against me, yet still I win because I’m just that great” marketing pitch. But I don’t understand why people who think Trump is a bad president buy this marketing pitch or persecution/martyr eschatology.

      I don’t want to turn this off-topic, so am not really looking for answers. And frankly don’t want a debate that will go even further and further off-topic. Feel free to add your opinion, but I won’t respond.

      • Acedia says:

        Feel free to add your opinion

        Thanks, I will.

        TDS does not refer to anyone who is strongly or outspokenly anti-Trump. It refers to people who spend very large portions of their day, every single day, obsessively thinking and talking about how much they hate Donald Trump, and who mindlessly gainsay anything that come out of Donald Trump’s mouth even when it’s something they would agree with if anyone else had said it. People who’ve allowed their hatred of the president to essentially replace large parts of their personality.

        These people exist and if you haven’t run into one yet you’ve been lucky. I’m a commited leftist so I know a few, and they are insufferable bores.

      • Edward Scizorhands says:

        I watched Bush Derangement Syndrome, and Obama Derangement Syndrome, so obviously there is Trump Derangement Syndrome. Every time the target of their ire so much as sneezes off-schedule, it is a call for impeachment or worse.

        It is a useful term. But, like all useful terms, it gets abused (because it’s useful).

  6. deciusbrutus says:

    Parenting Knowledge Number
    As of now, how many children total has your primary caregiver been the primary caregiver for?

    The answer to this question cannot be calculated from the answers above, if one’s primary caregiver has also been the primary caregiver for non-siblings, or if some siblings did not share the same caregiver.

    e.g. a blended family situation where both parents remarried and there are multiple sets of step-siblings under the default caregivers (The custodial biological parent and their current spouse).

  7. askwho says:

    I’m having trouble figuring out if Anglicanism is Protestant or not.

    I am going to have to skip the sleep part. I work a shift which pushes me around the clock. The first week 07:00 – 15:00. The second week 13:00 – 22:00. The third week 22:00 – 07:00. Then two weeks off. This makes it impossible to answer regular sleep questions.

  8. breviwit says:

    Game I: Growing The Commons
    I’m curious where Scott is going with this relative to the other two games. I feel like there’s a blind spot in many psych experiments like this. There’s a third party that the analysis pretends doesn’t exist, namely the person giving away the money.

    My order of preference for who gets the money is myself, Scott, then some random SSC reader. Given these preferences, and the fact that it’s far less likely that I would get the money, I should answer in a way which gives Scott more money. So I defected.

    How many other people thought of it that way? I suspect I’m a weird outlier here.

    For the two Prisoner’s Dilemma games, my answer only counts if I have already won, so my preference between Scott and other readers had a much much smaller effect on my answer.

    I have seen a few psych studies where one subject is choosing whether the other gets money and the results are explained as “punishing” or “rewarding” when the subject might really be saying “I’d rather the researchers kept the money”.

    I wish it were more common in psych studies to ask the subjects why they did what they did, and to publish that information (or at least summarize it).

  9. demost says:

    Hm, when I made the test on Open Sex-Role Inventory (male/female), I had the strong feeling that the survey was tuned for US americans. To go with the most obvious example, according to their database, the question “I like guns” has an extremely high positive loading for males. I don’t really know what “loading” means, but I find it hard to believe that the portion of European males who agree to that statement is outstandingly high.

    It is just a gut feeling and might be my prejudices. The authors of the study say that they have demographic data, but they don’t share it. I would find it quite interesting to see an analysis of that once the data from the SSC survey is available. Perhaps I’ll try it myself, though I am not trained in statistics.

    • Matt says:

      . I don’t really know what “loading” means, but I find it hard to believe that the portion of European males who agree to that statement is outstandingly high.

      Hopefully the test controls for that. The OSRI folks can tell if you’re accessing their website from the US or not, (unless you’re using a VPN, I suppose) and can score the question accordingly. I presume that, even in Europe, “I like guns” has a higher positive loading from males than females, even if both groups there are less likely to give a positive response. So they might disregard entirely answers from Germany OR they might give positive answers from Germany even more ‘male’ points.

      The question is also probably highly geographically correlated here in the USA, too.

      • demost says:

        Hm, I don’t think the term “controls for that” is applicable to what I am concerned about. Of course, there are a lot of questions where we can control for US vs. non-US. For example, we can ask whether there is a significant difference between males and females for this question after we control for the country.

        But when I make the test, I get a certain score, and I wonder how much this score depends on the country. I don’t think that the score I received is adapted to — what even? Me being non-US? Being European? Being from Germany? So probably my score is relative to the average person who did their test, where the average is set to 100. Of course, they could rescale the score values so that they represent the typical human being, but that would be odd, too. Would that mean that they rescale Chinese IPs to 20%, because 20% of the people are Chinese, although many of them don’t even speak English? That doesn’t sound like a good strategy either.

        In any case, I would be interested to know whether my low score (way below 100, both for male and female) indicates that I have untypical personal traits for a male, or whether they are typical values for my country. If all countries have roughly the same average, then my score indicates that I am a special snowflake. That is why I would like to know how much of the variance is due to geography.

        • Matt says:

          Well, here is what they say:

          From the items on this list that had the highest correlation with gender, a selection of 44 was taken as a beta version of the OSRI and made available on the internet. Using this beta version, responses were collected from 316,256 internet participants along with demographic information

          So they could do what I say, if they had the correct demographic information. If the US-based participants have males scoring 2.5 ‘higher’ than females on some question and German-based participants have males scoring 0.7 ‘higher’ on the same question, then they could give the same modifier to a US-based person who scores 2.5 higher than the average US-based person as they give to a German-based person who scores 0.7 higher than the average German-based person. They could theoretically even correct for cases where a country might have sign change. Maybe for some countries, the maleness/femaleness trait switches.

          It looks like they do not do that, though, so the discussion is moot.

          • demost says:

            Yes, we agree that they could do “that” (correct their individual answers for “demographic information”). I was just saying that the exact definition of “that” has no natural Schelling point (should they correct a score for being European? or rather German? or for even smaller regions like US states?), which is why they probably don’t do “that”.

  10. Signore Galilei says:

    PLEASE in future surveys let people pick multiple races, or at least add a “two or more races” option. I ended up skipping the race question altogether rather than lie by omission about one or the other of my racial categories. “Other” doesn’t work because then I am lying about both.

  11. Two McMillion says:

    For the charity question, it might be useful on future surveys to distinguish between charitable giving which is to a church vs. which is not to a church, though I don’t know if your readership is religious enough for it to make a difference, or if you care about the difference between giving to churches vs non-churches.

    For the police question, I was in fact guilty when I was last pulled over, but I still feel it was unreasonable of them to pull me over.

    • Lurker says:

      why would the church matter in this context?
      I’m not particularly religious, but if a specific set of people were doing some kind of charity I approved of and I was in a position to support it, I would do so, regardless of whether it was connected to a religion (or even the religion I was raised with) or not, since the one organizing the charity is only relevant insofar that they’ll actually do what they say they’ll do with the money/help/donations they got.

      • anonymousskimmer says:

        I believe that tithing is seen as obligatory by most churches. You’re effectively paying to be a member of the church. Why would this be considered charitable giving given that you receive value in return?

        Though yes, this neglects donations to churches that are not tithes.

        • Lurker says:

          in germany those tithes are literally called “church taxes” and the state actually does the taxing for the churches, so I thought charitable donations were just money that went towards charity, which is why I wondered how “this charitable donation goes to a church” is relevant.
          If that charity includes church “fees” in other countries, then it’s actually a good question to ask.

  12. Midge says:

    First time taking the survey ever — and the second comment, ever, I think, that I’ve left here (I had to re-register, it’s been so long). I stayed up wayyyy past my bedtime to complete it, didn’t have some of the information on hand and was too exhausted to ballpark it, so sorry about leaving those parts blank. Some stuff I genuinely couldn’t know: I only ever learned the threshold my IQ exceeded to qualify for a gifted program, never knew what the score itself was. When I could give a descriptive answer, though, I tried.

    For most multiple-choice, including politics (which was hard), I tried to give the nearest “stupid” answer that kindasorta fit, figuring that would be more useful than styling myself an exception. But some were impossible, like sleep latency, where the only answer could have been “my latency variance is too high to fit any of these categories”.

    For several choices, in fact, my honest answer would seem to be “my physical state is too variable to pick one”. (People have reliable nap habits and can reliably tell whether naps refresh them or not?…) My body is less reliable than normal, but I’m not sure that makes me so abnormal on SSC, even if it leaves me with too much variance to answer questions directed at SSC-ers, including questions nominally about habits and psychology rather than physical function.

    By the time I got to the end of the survey (which seemed to take hours, perhaps because I was so tired), I had long lost any motivation to be clever. I could do the math on cooperate-defect problems, but by then I lacked all motivation to give calculated rather than whimsical answers. I’ve always sucked at strategy, since I get little joy out of applying what I understand for mere “winning”. Card problems that are just combinatorics? OK! Applying combinatorics to actual card games? Meh. Can’t bring myself to pay enough attention, and truthfully I’m usually happy to be the game’s unfocused, congenial loser.

    Despite my virulent anticompetitive streak, completing your survey showed me I’m a gal who types as more masculine than feminine on the OSRI — an inventory I’d never heard of before. I wasn’t surprised by that result, just amused.

    I hope I helped, and am glad to think I might have. I’ve lurked at SSC forever, though I seem to be the kind of hot mess who makes for a highly ineffective altruist.

  13. valuevar says:

    Also, I imagine you are asking for the year in which people took their SATs so that you can renormalize them yourself? The scale has been changed several times (and the conversion tables are public).

  14. Ezra says:

    I filled out part of the survey, then switched out to check my SAT score. When I switched back, the page had been refreshed and all my answers were gone.
    So I filled it out again with my SAT score, and on the assumption the previous was a fluke, switched out to check my income. Nope. When I came back, all my answers were gone again.

    Could there please be some sort of “save answers” each section? I can’t muster the will to fill it out a third time tonight, and if tomorrow it drops my answers a third time, I might not be able to submit at all.

    • Matt says:

      I’m not sure what you mean by ‘switched out’, but I feel like it would help you to know you can open a new tab in your browser and leave the current tab open to the unfinished survey, which will save your answers well enough.

      But maybe you’re having a problem that will not be helped by this?

      • Ezra says:

        I left the tab open in my browser. In one case I switched to a new tab to gather the information. In the other I went to the filesystem to get a file.

        It might matter that I’m on a mobile device (Android).

    • Ezra says:

      Made it to the end of Part 1 and start of Part 2! But then I left for dinner and by the time I got back it’d reser to Part 1 and cleared itself again.

      I don’t think it’s feasible to answer this long a survey on mobile. Our blogger will have to do without my data this year.

  15. Denis says:

    I encourage you to put the decision to share answers publicly at the end of the survey, since it’s not clear what will be on the survey when you answer that question.

  16. valuevar says:

    Honestly, there shouldn’t have been a free form field for “Please list some ways in which you believe specific questions in this survey to have used ill-defined terms, or involved questionable implicit assumptions.” Looking at some of the comments above, I don’t seem to be the only one to think this way.

  17. switchnode says:

    I was once the victim of a hit-and-run; I had a literal moment and answered ‘yes’ to the question about having been a victim of a violent crime since it a) involved violence and b) was a crime. (My jurisdiction doesn’t have a legal definition of “violent crime”.) It has since occurred to me that this is probably not what you meant, and that the sample size might be low enough that I should mention it. Sorry about that.

  18. Max Chaplin says:

    The Race question doesn’t have Jewish as an option. In the past I’ve marked Middle Eastern but now I have doubts this is what Scott meant. Is this the option I should select?

  19. hnau says:

    For the lying question, did anyone else select “falsehood due to self-deception” but *not* “deceptive though technically true”?

    My reason for doing so is that both seem like culpable attempts to mislead, but the first kind is much more insidious than the second. I can detect and compensate for “technically-true” by paying strict attention to what the person says, and asking for clarification if necessary– there are still rules of the game that both of us can follow. But if they’re saying falsehoods that they ought not to believe given their knowledge, then I have no way of detecting that; it’s just as if they were actually, consciously lying.

    This stems in part from real-world experience. I work as an engineer in Silicon Valley and have to base very impactful career decisions on information given to me by founders and executives– a class of people that is notorious for being unrealistically optimistic to the point of outright self-deception. If this isn’t considered “lying” then is it something we should accept? And if not, how are we supposed to hold them accountable?

    • switchnode says:

      I thought about it, as I find it similarly offensive, but didn’t, as it seemed too far a departure from the prototypical definition to be good communication. (One act can be as bad as another without being the same kind of thing; rape can be worse than manslaughter, but that doesn’t make it homicide.) There are many things other than “lying” which we need not accept, and from a practical perspective, methods which work to detect and (especially) deter conscious liars often fail on self-deceivers.

      • hnau says:

        I assumed that deliberate lying and self-deception were indistinguishable to most outside observers, but you make a good point about there being easier countermeasures for deliberate lies. If I could go back and change my vote, I would.

  20. Byrel Mitchell says:

    I’m curious why the ACT wasn’t a question in the IQ section. Is it even less rigorous than the SAT as a proxy for IQ or something?

    • hnau says:

      What I heard in high school is that the SAT tries to measure “aptitude” independent of background / preparation, while the ACT is just trying to be a standard measure of common academic achievement. So in some sense we should expect the SAT to be a better proxy for IQ (though everything I’ve read about it says that background / preparation does still matter a lot).

  21. kalimac says:

    Additional question:

    Did you A) know your SAT scores offhand, or did you B) have to dig out some dusty old piece of paper from your random document files in order to answer this question?

    My answer: B.

  22. VivaLaPanda says:

    For the Subreddit and Discord questions, I felt like I need an option for (occasionally read, would read more if I had the time/bandwidth). I like both communities, but have other communities that tend to take up my time instead.

    For the masturbation question, be aware that for many people like myself, the fact this survey is happening right around the holidays might skew the answers.

    On the blank wall question. I had a hard time determining how to quantify static vs intense static. It’s the “normal” amount from my experiences. I’d say that it’s very clearly visible assuming I’m looking for it, but it’s in no way distracting.

  23. kalimac says:

    “Other people are basically trustworthy/basically untrustworthy.” Depends on the context. People whom I meet in ordinary circumstances: probably trustworthy. People who call me up on the phone and try to sell me something: entirely untrustworthy.

    • hnau says:

      Hmm, good point. In practice the incentives created by social context determine most of trustworthiness / untrustworthiness. But the whole point of asking the question is usually to figure out how we go about building that social context– should we be more permissive or more restrictive? So I usually take it as a thought experiment like Plato’s Ring of Gyges: if we weren’t bound by any of the usual constraints would we still generally act in a trustworthy manner?

  24. MikeInMass says:

    I was completely unable to picture a red star except for one on a Macy’s gift box, white background and all. Any Macy’s marketing execs reading this, go ahead and high-five yourself now.

    The variance in the amount of time it takes me to get to sleep is so high, I didn’t feel like I could give a good “average” or “usually” answer. If I spent the whole day doing physical work, or if the missus and I just finished gettin’ jiggy, I’ll be out within a few minutes. If I spent the day sedentary, it could be a few hours. Apologies if I missed the intent of the question, hope this might help you ask it differently next time if necessary.

  25. David Speyer says:

    Regarding the questions about relatives with mental conditions — I assume you mean blood relatives? Or do you want us to count spouses and parents-in-law as well?

  26. Byrel Mitchell says:

    The hardest question on here was the political philosophy one. I strongly oppose the US Democratic party (which was the example for liberalism), but the short summary of liberalism was the closest to my views of any of the short summaries. Either I’m somehow failing an ideological Turing test on people with the same beliefs as me, or that example and description are mismatched in a significant way.

    I ended up selecting liberalism (filling it out based on the summary), but I’m not at all sure that it was the best choice.

    • VivaLaPanda says:

      I think part of that might be that the US Democratic party in this moment is more on social democrat side, but the Obama (and before) era party was more like the description of “liberalism” given.

  27. kotrfa says:

    Hey. Just a note – in Europe, we usually mention our salary per month, not per year – I am sure people are going to mistake that (e.g. Stack Overflow Survey often has this mistake). Also, I wasn’t sure if you mean gross salary or net.

    Thanks for this Scott.

  28. andrewducker says:

    “Psychedelic Change” – seems odd that “I never took psychedelics” and “I took psychedelics and they didn’t give me visual artefacts” are the same answer.

  29. andrewducker says:

    The “Night Time awakenings” question needs a checkbox for “I have a small child and a cat.”

  30. andrewducker says:

    “Ride-sharing like Uber”
    I’ve used a Taxi, like Uber are, in the UK. Does that count?

    • VivaLaPanda says:

      I would assume the question is specific to services where the drivers own their own cars and work in a contractual fashion.

  31. DNM says:

    I’m very surprised the “life satisfaction” section doesn’t have a “how satisfied are you with your health” prompt.

    Also, for me at least, you did not at all ask the right questions about SAD – I am curious where you are going with this.

  32. Ron says:

    Coffee was ambiguous for me, in the morning making me more awake, in the afternoon more tired. I think I once read a paper mentioning that this is at least somewhat universal, so maybe worth keeping this in mind when analyzing the data.

  33. brownbat says:

    Some scales had “other,” but some questions were more strict. I skipped a few questions where I felt like the scale or question had an embedded assumption I couldn’t come to terms with, like the left right political spectrum.

    One minimal effort change might address this — in future surveys, you could add a free form box at the end labeled “Comments, qualifications, explanations, or other nitpicks with any of the above questions.”

    Then add instructions at the top saying there will be this box to add quibbles at the end, and that respondents should just answer the closest answer to the question in the spirit in which it was written.

    Whether you completely ignore that free form box or deeply contemplate every comment there, you’ll probably still induce higher response rates on items provoking squirrely analytic objections. If nothing else, you’d go a long way to reduce the mild cognitive suffering experienced by anyone asked to describe themself in terms they don’t accept, given we all have a few nutty and eclectic worldviews or ontologies.

    I’d hazard a guess this problem afflicts this crowd more than most, but it might still be negligible and not worth addressing, only you’d know for sure.

    • switchnode says:

      On the one hand, +1.

      On the other, perhaps seeing which respondents leave comments to air their grievances is a useful way to filter the ‘analytical issue’ quibbles from the ‘discharge my mild cognitive suffering’ quibbles.

  34. eggsyntax says:

    For the record: I was halfway through filling out the second page when my computer froze up in a particularly odd way (I mildly doubt that it had anything to do with the survey, but can’t be sure). I couldn’t start again from the second page, so I did the whole thing over. So you have (I assume) duplicate first-page responses from me. If you’d like to weed the half-completed one out, I’m happy to provide identifying details (or it may be sufficiently identifying to say that I’m in North Carolina and submitted both within the last few hours, and first-page responses should be essentially identical).

    If it’s not worth weeding out, by all means ignore this comment.

  35. clipmaker says:

    For privacy reasons I wish it wasn’t a google form. I might fill it in from a library or something.

    • Edward Scizorhands says:

      I forget if I suggested this last year, but the privacy question should be at the end, after we have realized what kind of information we are giving.

  36. physticuffs says:

    I did not know how to answer my current socioeconomic class, given that I am a student with no dependents making ~25k/yr. A 25k income puts me pretty far below the American average, but is plenty for me at this stage of life. As a research student in a well-funded lab, my likelihood of being fired is miniscule. Plus, although I pay for all my own expenses, I know that if I were in serious trouble, my parents would be a safety net. In other words, I have a low income but almost no economic insecurity–it feels wrong to refer to myself as working- or middle-class but I’m definitely not higher than that.

    • ilikekittycat says:

      “Is most of the money you’re spending (including parental supplementation) from generational inheritance/investments/rents/financial instruments or from selling labour time?” is a good heuristic in a pinch. Just my opinion: there are self made tech guys, football players, rappers etc. that have 7 figures but never escape working/middle class; there are also used car moguls in Idaho who can’t crack $500k liquid but will never fall into working/middle class

  37. birdmaster9000 says:

    I had a professor once that would add a question onto coursework-review surveys asking about the “type of survey-taker” one regarded themselves as. As in, some just work through surveys to get them over with, a “sure whatever” survey attitude, while others answer as accurately as they can, a “can’t hinder usefulness of survey results” attitude. Of the latter, there was a categorization for those who answer any ambiguity in a question based on their inference of what they think the survey creator wants and a categorization for those who answer as literally as they can despite any inferred intention of the survey creator. There was another option, was always 4-answer multiple choice, but I do not remember it.

    Probably hard to make useful and might not be relevant to the analysis of data from any specific questions, but might serve as a nice debug question in the event any question is recognized as ambiguous after survey creation. Sort of like a more rigorous “Check answer C to prove you actually read and answered these questions.”

  38. jonabar says:

    I find melatonin does help me fall asleep faster (taken 15 minutes before bed) but makes me sleep longer (if I don’t use an alarm) and feel less refreshed when I wake up (either way). The last reason is why I don’t take it regularly, only if I’ve been drinking alcohol or working late at a computer.

  39. noahyetter says:

    I found some of the “Life Effects” options deeply puzzling, enough so that I bothered to log in and post about it.

    Firstly and quickly,

    I reduced the amount of meat I eat, or otherwise changed meat consumption patterns

    I feel like this is a topic that Scott never talks about, beyond occasional commentary on his personal habits, at least not while I’ve been reading the site (since mid-2015). So I struggle to see how reading SSC would cause such a change.


    I changed (or am seriously considering changing) careers to work in AI safety

    AI safety is a topic Scott semi-often references, but never actually explores. Even those references I find confusing. He will often make an oblique mention of MIRI, in a way that obviously assumes we all know who they are and what we should think of them, without linking to them or footnoting or parenthesizing or anything. It gets me every time because I do not, in fact, know who they are, and a google search does not make it particularly obvious. I looked at their site just now and I still don’t know. Setting that little bit aside, I can recall zero posts in the time I’ve been reading SSC that have actually been about AI safety. Perhaps there have been some, but if so they clearly did not leave an impression, and the topic is not something that springs to mind when I free-associate “SSC”. So again I cannot understand the hypothetical connection between reading SSC and the described “life effect”.

    But even beyond the SSC connection I find the very idea baffling. Changing “careers” to “work” in “AI safety”? How many jobs are there in AI safety, globally? 100? 50? Zero? I’m sure there are academic researchers engaged in this, but that’s not a career change you can just make, that’s putting your life on hold and going back to school for 4-8 years and then hoping it lands you somewhere. This just isn’t a thing that you can do.

    What am I missing here? How is this relevant, and even if it’s not how is it even possible?

    • VivaLaPanda says:

      It’s just a result of this blog coming from LessWrong which was (especially back in those days), like 80% about AI safety. Yudkowsky is high up in MIRI, and was the founder of LW and much of the rationalist community et large (of which this blog is usually considered a member). Scott probably just expects the readership to be familiar with the LW-sphere, especially given how many live in the Bay Area where MIRI is located and the rationalist community is centered.

      As far as references go, the Moloch post, maybe the most famous post here, essentially ends by concluding that AI research is the most important thing and the only way to beat Moloch.

      As far as jobs go, I think that probably goes back to the percentage of readers in the Bay area where between MIRI, Future of Life Institute, OpenAI, and others, there’s a decent number of AI Safety jobs.

  40. zz says:

    Re: “SSRI effectiveness” on page 2:

    I have taken SSRIs, but not as an antidepressant. Should I answer this question?

  41. JohnNV says:

    I was confused by the wording on the biodiversity question – the two extremes are favor and disfavor, but are you asking whether the respondent believes the proposition to be true? That’s different from favoring it which implies some sort of value judgement. I can not believe in the afterlife but still hope that it exists.

  42. Nancy Lebovitz says:

    This year, I’m taking notes as I go.

    I’m agnostic but spiritual, I suppose. I believe in a certain amount of weird shit, especially qi, but unlike New Agers, I don’t believe the universe is overall benevolent. nor that it’s possible to make things happen by wishing emphatically.

    I wouldn’t mind a question about how intense one’s religious upbringing was. Mine was definitely Jewish, but no one seemed to have a strong belief.

    I haven’t formalized it, but I’m inclined to think that a consequentialist/deontologist mix probably makes the most sense. I’m not sure what natural law morality is. A basis for choosing your moral premises?

    Why did you drop the internet IQ tests?

    I don’t consider commenting many times a week to be frequent, not compared to the really frequent commenters. Frequent is more like many times a day. I wouldn’t mind a question about how much of ssc people read.

    I put myself in the middle of the political spectrum because I’m not far left or for right. I’m a harm reduction libertarian, and I don’t have a political home. Maybe a question about whether people feel they have a lot of people who agree with their politics.

    Would it be worth distinguishing between mild-to-moderate depression vs. severe depression?

    I probably have ADD, but not ADHD. I answered positively for ADHD, but it might make sense to include both next year.

    I always take the SSC survey. I generally don’t take surveys.

    I have as many tabs open as Chrome leaves visible, sometimes with one or two more. I’ve had a dream about having two rows of tabs.

    How about a climate question about whether where you live gets too much/too little water?

    In self-help: What is circling?

    • Deiseach says:

      Why did you drop the internet IQ tests?

      I’m going to presume it’s because there are so many, and no real way of knowing which ones are any use or if the results are accurate. I’m at the stage of thinking that, before the IQ/SAT/other alphabet soup tests questions, there should be an “Are you American? If not, skip ahead to Q. 25” choice 🙂

      In self-help: What is circling?

      Same here. For that whole section, I more or less went “The hell is that?” and answered not at all.

      • Lambert says:

        Att he risk of revealing my alpha mater, circling is where you yet a couple of pints of dark snakebite and play various drinking games with your society,

    • thisheavenlyconjugation says:

      In self-help: What is circling?

      This thing.

    • Elementaldex says:

      I’m with you on the survey taking question… I take the SSC survey every year and so it is unsurprising. But it might be literally my only survey taken this year.

  43. Alkatyn says:

    I realise you can’t change it, but might want to be wary of cultural specificity in the OSRI scores, as I strongly suspect things like guns will score very differently in the US and non-US regardless of gender. And things like how much you communicate emotion are known to be very culturally variable

    • Jerden says:

      It’s definitely a US centred quiz, I noticed that as a Brit.
      I struggled with the gun question, mostly because I think guns (and weapons generally) are cool but have no desire to own them. I went with “slightly favour”, probably helped me assert my masculinity out of the undifferentiated zone.

      • Nancy Lebovitz says:

        I was wondering whether there should be a few questions about hot issues in countries other than America.

    • anonymousskimmer says:

      Whether its culturally over specific should come out in the results, assuming there’s enough people from other countries for statistical reliability.

  44. chaosmage says:

    How long will the survey be open? Will there be a warning before it closes?

  45. Templar15 says:

    I’d just like to say that the raffle/contest questions are really funny, and I hope we get some good insights out of them.

  46. some fairy says:

    If it turns out that AGP is correlated with [controversial thing for AGP to be correlated with], please contact me before you post an analysis about it. We have some things to say that you might want to have heard first. Our disposition is that it shouldn’t be discussed yet.

    We’ve been throwing around an evolutionary theory of kink in the lw slack channel for a while. At least two of us arrived at it independently afaict, others think they did (but I question this cause we’ve been talking about it for a while in their presence). I’m reluctant to explain the theory just now, for two reasons, one is that it’s a bit inflamatory and I’d prefer not to talk about it until we can evidence it, it’s especially inflamatory in this context, and also if the results don’t come out the way I’m worried they will, it probably doesn’t matter if people remain ignorant of it and I’ve got other stuff to do. The theory.. roughly speaking.. inverts the meaning a correlation here would have. Discussing those results without the theory would fairly inevitably result in the wrong thing happening rather than the right thing.

    [if email notifications fail to work for whatever reason, it’s easy to find us on the lw slack channel. That’s probably where we should talk about it anyway.]

    • BlindKungFuMaster says:

      If it turns out that AGP is correlated with [controversial thing for AGP to be correlated with] … Our disposition is that it shouldn’t be discussed yet.

      – no shit.

      So you came up with a theory that allows you to draw the opposite of the obvious conclusion. That’s a neat trick. Probably another AGP correlation.

      Not so wild guess at the theory: balancing selection. (Not sure this is the technically accurate term, but I guess you know what I mean)

      “People” will remain ignorant of this anyway, because it will never ever feature in the media. And people in the LW, SSC circle have the brains to notice it immediately if they just care to look.

      • some fairy says:

        balancing selection

        Hmm… ah yeah, I recognise that one. That’s not the theory. I have been thinking about weird variants of that one (pathological internalisations of “what is virtuous? Do as you would have others do” sort of thinking), but the one I had in mind is pretty much orthogonal to that. It’s worth remembering at this point that both things could be happening and combining effects and maybe neither were strong enough to produce the symptoms on their own.

        I anticipate a shift towards looking for feelings’ EEAs in clinical psych more often. That’ll probably be common sense among people who believe in evolution, one day shockingly soon. It’s sort of an entailment of an entailment of that, so I don’t think propagating it would be all that hard once we had our ducks in a row.

      • Pink_Creosote says:

        So you came up with a theory that allows you to draw the opposite of the obvious conclusion. That’s a neat trick. Probably another AGP correlation.

        What’s the obvious conclusion, if I may ask?

    • tailcalled says:

      If you are talking about the correlation between AGP and gender dysphoria or transsexuality, it is already well-documented (without Scott’s survey) that it correlates, see e.g. this, which finds a d~1.9 to 2.9 effect:

    • Deiseach says:

      if the results don’t come out the way I’m worried they will, it probably doesn’t matter if people remain ignorant of it and I’ve got other stuff to do

      You do realise that sounds awful if you’re talking about anything over the level of a SurveyMonkey ‘create your own quiz – which gemstone goddess are you? find out based on your ranked favourite vegetables coloured orange!’ timewaster? “If the results don’t turn out the way I want them, I’m gonna pretend I never said that thing in the first place and besides you are all wrong and deliberately choosing the wrong answer to mess with me so I’m sticking my fingers in my ears and ignoring it!” That’s not an ‘evolutionary theory’ of anything, that’s only bullshitting on the Internet (which we’ve all done).

      Yeah, I think I’ll stick with Scott’s version of “I have this hypothesis in mind, so I’ll ask you all these questions, and then we’ll see if the spaghetti runs up the flagpole after it moves the cheese and wears the empty raincoat”. And if it moves to Chicago instead, then he’ll tell us so instead of pretending it never happened.

      • some fairy says:

        :/ I’m aware it would bear negatively on the theory. I may actually post negative results if it turns out the theory isn’t true, or is making a claim that turns out to be vacuous/useless when you really take it apart, but I’m not totally sure that would be worth anyone’s time.

  47. babarganesh says:

    i suppose this is as good a place as any to thank Scott for the melatonin post. my wife (who is not the sort of person who would enjoy reading SSC – and yes we have an autistic offspring) now takes it in the morning and is no longer dead tired by 9 pm. we sometimes even can interact at night now.

    i (no big sleep issues) now use melatonin regularly though at first i was in the “wake up at 4 am feeling way too refreshed” camp. that was with an approx 0.3 mg dose however. i tried scaling back to 0.1 mg or so and it does help me get to sleep. it’s a small effect though.

    our autistic son refuses to try it, because, you know, autism.

  48. kazimircz says:

    Hi Scott. Love your site, took the survey. But i have a question – you ask about masturbation frequency, yet do not ask about sex/lovemaking frequency (i dont think??). I do not understand why. In my experience they are somewhat correlational (causal?)
    Thanks a lot.

  49. Hackworth says:

    Typo thread:

    “What is the age gap between you your closest later sibling?” should contain an “and” between “you” and “your”.

  50. the verbiage ecstatic says:

    Mild question-wording gripe: it’s not obvious (to me, anyway) for the mental illness questions which answer you should pick for “I was diagnosed with X but believe I am now completely cured”). I went with yes, I have it, as that seemed closer to the spirit, but a “have or had” would go along way to make it clearer

  51. acymetric says:

    In “referrals” what does “message board” count as? Other social media? Referred by a “friend”?

  52. Hackworth says:

    I disliked the question about Feminism, it should have been more clearly defined what exactly you mean by that. If you mean the basic definition of equal rights etc., then I’m all in favour; but the strand of feminism that currently dominates the public discourse, especially on the web, is to feminism what PETA is to animal rights: it can go die in a fire. So which is it?

    • thisheavenlyconjugation says:

      Neither, it rather refers to the devil’s brew, the poison scourge, the bloody monster, that defiles innocence, dethrones reason, destroys the home, creates misery and poverty, yea, literally takes the bread from the mouths of little children.

    • eh says:

      Yeah, I don’t like those kinds of questions, because there’s no way to tell what neutral actually means. Are you strongly in favour of gender equality but strongly against exceptionalism? Are you trans and not particularly happy with TERFs? Have you never heard of feminism before? Are you a feminist activist who’s disgruntled with her own movement? Are you religious and against abortion but agree with women’s lib? Are you scared of internet feminism but a lukewarm supporter of IRL feminism?

      There’s so much nuance that gets squeezed out, and I think there are interesting patterns and edge cases that would emerge from a free-text entry for pretty much any question on the survey.

      • Montfort says:

        I think taking the nuance away is intended, and it’s more of a bellyfeel question than an intellectual one. How I interpreted it was something like “if you heard of a new idea or public figure described as ‘feminist’, what would be your gut reaction before you read the details?” Whether your opinion on “modern”, “historical”, “radical”, “mainstream”, etc. feminism predominates also determines your reaction to the movement/brand, I imagine.

        If he wanted to really study attitudes about feminist policies and positions, we’d have an extra long section about it. Which might be interesting some year, but it would be very time-consuming.

    • Deiseach says:

      I have no idea what wave or post-wave version of Feminism we’re supposed to be up to now, or Officially Supposed To Be (as in the usual “women’n’minorities” galumphing by politicians and media types). I think I’m the kind of feminist that came after Emmeline Pankhurst but before Andrea Dworkin, if that helps any 🙂

      Countess Markievicz but without the gun feminism?

    • Bugmaster says:

      I took it to mean, “Feminism as it is most commonly practiced today” — i.e., referring to the modern movement and its adherents, not just to some abstract philosophy.

      • blacktrance says:

        I interpreted it as the philosophy, because past versions of the survey had a separate question for SJ, which I took to be the movement.

      • Deiseach says:

        referring to the modern movement and its adherents

        Lord God Almighty, I hope not – because any adherents of the modern movement I am aware of are all either screeking about the Patriarchy under the bed (if nobody wants to read their stunning insights into how 2+2=4 is a colonial invention to oppress the wisdom of indigenous peoples) or are orbiting Pluto, they are so far out.

        I suppose there must still be a few ‘here are real concrete problems of economic and cultural justice that need to be addressed but without collapsing the entirety of global civilisation or saying every man should be shot’ feminists out there, but I don’t get to see them represented much. I would five times rather see something from feminists working on the problems of dowry deaths than some physics professor moaning about how she is insufficiently respected as a non-European because people aren’t changing their conferences to what she thinks they should be about.

        Yep, I’ll definitely be over there in the corner, showing off my plumed hat to Chinggis Khan 🙂

    • Plumber says:

      @Hackworth says:

      “I disliked the question about Feminism, it should have been more clearly defined what exactly you mean by that. If you mean the basic definition of equal rights etc., then I’m all in favour; but the strand of feminism that currently dominates the public discourse, especially on the web, is to feminism what PETA is to animal rights: it can go die in a fire. So which is it?”

      I’ve no idea if “feminism” means my Mom wanting some laws changed back in the ’70’s, women to be treated better in Saudi Arabia, or some on-line screeds that our host objected to a while back, so I just put “Neutral”/middle bubble and didn’t think about it too much.

      • Garrett says:

        Broadly speaking:

        1st wave feminism: women are people and should have individual rights. Eg. should be allowed to vote and own property.
        2nd wave feminism: women should be accepted in whatever roles in society they want to pursue. Eg. it’s fine if a woman wants to be a welder, lawyer, doctor or politician (or plumber), and any woman might be capable of the work.
        3rd wave feminism: Intersectional feminism. That there is a lack of gender representation in most fields is evidence of systemic (perhaps unintentional) bias or oppression which must be overcome.

        There are enough significant differences between major waves, not to mention other near offshoots that it’s hard to truly be certain which issues are being discussed. As always, there’s a

        West Wing reference.

        • Plumber says:


          I’ve mentioned this in other threads that there seems to be a lot of ink and pixels on the “representation of women in STEM” but from my observations (the DA’s and public defenders at work, physicians at the Oakland Kaiser hospital, and just like 50 years ago: teachers) most “professionals” are now women, cops and deputies are still mostly men, while building trades are still overwhelmingly men (about one in fifty electricians I’ve seen are women, about one in a hundred plumbers are men) and it really looks to me that the more “female representation” of any above median pay profession/trade there is the less “non-white representation” there is (with physicians and public defenders being the exception as those professions have lots of Asians, but still few blacks and Hispanics), which I very seldom see remarked on.

          In comparing electricians to plumbers it looks like a trade may have more white-women or more blacks and Hispanics (including a very few black women) but not both.

        • Aapje says:


          1st wave feminism: women are people and should have individual rights.

          This is a very bad description of (any wave of) feminism, which seems based on a revisionist misunderstanding of prefeminist society. Those didn’t threat women as property and they gave women special individual rights that men didn’t get (and vice versa).

  53. Nicholas says:

    Oh, hey, I see ads on my desktop! Usually read the site from my phone, just noticed there’s a slide-out pane with ads on the phone, so I *could* see them there if I wanted to.

  54. DinoNerd says:

    With regard to visual trails etc. I had every effect you can think of, while I had cataracts and/or in the immediate aftermath of cataract surgery. I now have nothing of the sort. Your questions seem to presume such things never go away.

    So my null hypothesis for this isn’t psychedelics etc. – it’s real physical eye problems, sometimes correctable by surgery. (I’d expect astigmatism to have some of these effects too.)

    I think you may need a question about eye problems, to distinguish the baseline visual problems from whatever else you are looking for.

    • Bugmaster says:

      You are absolutely correct about astigmatism; high blood pressure also has an effect.

      • Midge says:

        Also, migraines.
        And migraines + astigmatism.
        And fatigue + astigmatism…
        Whatever defects in the eye, why wouldn’t stuff like migraines and fatigue alter visual processing, if drugs do?

    • disquietingmalodorous says:

      If the visual effects people describe as persistent hallucinations are the exact same visual effects that come from optical defects, maybe psychedelic use just tweaks the experience of users so that they are able to notice the defects in their vision that their visual field would normally messily filter out.

      The purpose of the question really could just be to ask how many users have a direct awareness of their eye defects, or for cross-comparison against the imaginary visualization question.

    • Robin says:

      I think it would be a good idea to swap left and right on the astigmatism question, so the scale matches the sides of the photo. It might confuse people that it is the other way round.

  55. Deiseach says:

    Cue the usual complaints about the survey questions 🙂

    I constantly have a problem with the orientation/sexuality/romance and love life ones. There’s really no option for “I am very happy with my romantic life; I don’t have one; I don’t want one which is why I’m very happy”. We get the “are you happy with your current love life?” and “do you have a current love life?” ones, but then it’s juggling between “do I pick ‘happy’ which makes it sound as if I am in a relationship; do I pick ‘none’ which makes it sound as if I’m looking for one?” Ditto the orientation ones: I emphatically am not in the camp of “Asexuality is a completely different orientation and should be treated as queerness” but picking plain “heterosexual” doesn’t give the full story, I feel; it sounds like “yeah I’m straight and I’m pining for a partner like a Norwegian Blue pines for the fjords”.

    EDIT: I also had real problems this time round with the politics one, so I had to leave it blank; I’m not Democrat-ish because I can’t be on board with the extreme social liberalisation/multi-culti stuff but at the same time I’ll die in a ditch before I consent to be lumped in with the modern-day UK Tories. Anyway, all the parties nowadays are going for social liberalisation/multi-culti stuff regardless of whether they’re left, right or centre, so at the rate I’m being left behind as a fossil of history I’ll probably end up with the renascent cult of Chinggis Khan!

    Anyway, apart from that – re: lying, I picked an additional answer which puts me in the same basket as St. Augustine rather than the Jesuits, even though technically equivocation isn’t lying. I also liked the good old BEM test, even though I think it’s definitely showing its 70s roots (and I always end up slightly more masculine than feminine on it, which I think is down more to ‘weird brain’ rather than ‘ah, so your real gender is…’) 😀

    It was fun though, and I liked participating, even in the lucky dip for the money (though I never win anything)!

    • DinoNerd says:

      so at the rate I’m being left behind as a fossil of history I’ll probably end up with the renascent cult of Chinggis Khan

      Fossils of the world unite!

      Different reasons, same effect – none of the available parties represent me.

    • tailcalled says:

      The OSRI is actually somewhat different from the BSRI, despite the OSRI having officially been constructed to be a replacement for the BSRI. For instance, the gender differences on the OSRI seem bigger than those on the BSRI. I think it’s because the OSRI more focuses on interests and activities, while the BSRI more focuses on personality.

    • Aapje says:


      I don’t understand your objection to the love life happiness question. If you answered that you are single and gave a 10 to your love life, then you would accurately claim to be a perfectly happy single, right?

      • Deiseach says:

        It’s more that it’s lumping a lot of things into the same basket, which isn’t really garnering much useful information for a survey, do you get what I’m saying? It’ll crudely sort out who is self-identifying as gay/straight/other but not much more; or if you match up “are you in a relationship?” with “how many kids do you have?” it’ll give you an idea of “are the smart people having more, fewer or the same number of kids?”

        But plain “are you in a relationship now?” sort of question lumps together and does not distinguish the:

        A) “no but I really want to be”
        B) “no and I chose not to be, but I’m open to being in one if I change my mind”
        C) “no and I don’t ever want one”

        There’s a lot of difference there as to “I really want an ordinary relationship but can’t get one and it’s making me so miserable I wish I was dead” from “I never ever want a romantic relationship and I’m happy”. A is very different from C there, and I think it does make a difference when you come to do something with the data you’ve gathered.

        You’re correct though that it’s minor grumping: with all the talk about X rights for sexual/gender/expression minorities, I do feel that people who don’t want sexual and/or romantic relationships get left out or pushed aside. If we’re going to ask for consideration that not everybody is a Christian at Christmas (so “Happy Holidays!”) then I think we should also consider not everyone is a romantic on Valentine’s Day (so, er, I suppose “Happy Cyril and Methodius Day!” instead?) 😀

    • Plumber says:

      @Deiseach says:

      “…I also had real problems this time round with the politics one, so I had to leave it blank; I’m not Democrat-ish because I can’t be on board with the extreme social liberalisation/multi-culti stuff but at the same time I’ll die in a ditch before I consent to be lumped in with the modern-day UK Tories….

      Oh join the club!

      I’ve long ago posted my political goal that decides my vote: U.S.A. 1973, specifically February, which I judge as a move to the Left compared to now. 

      @Nick has linked to a couple of speeches by a couple of Republican Senators that looked like good ambitions to me, but I’m even more doubtful that they’ll achieve them than some goals I’ve seen spoken by some Democrats that I’ve seen, and I’ll probably vote for Biden on March 3rd.

      “..I also liked the good old BEM test, even though I think it’s definitely showing its 70s roots (and I always end up slightly more masculine than feminine on it, which I think is down more to ‘weird brain’ rather than ‘ah, so your real gender is…’) 😀..”

      I got a “99” for “masculine”, and a “108” for “feminine” which I’m guessing is due to my putting “slightly” for most stuff (which is likely how I tested out as a “fascist fellow-traveller” that other thread quiz link) except for guns which I put as “strongly dislike” which I guess is girly (I’ll be sure to “feel in-touch with my feminine side” the next time I’m unclogging drains in the autopsy room)!

    • Acedia says:

      I’ve never seen a political survey acknowledge the existence of people who are economically leftist and socially conservative. Even though there are huge numbers of them, particularly in the lower classes. It’s such a weird blind spot.

      • Plumber says:

        In a survey of the 2016 electorate that I’ve often seen cited by opinion/analysis in The Atlantic/New York Times/Washington Post/et cetera, about 28.9% of voters both lean “Left” on economics and lean “Right” on ‘social issues’, and a slight majority of all 2016 American voters were ‘social conservatives’, and a majority was (slightly) ‘left-of-center’ regarding economics.

        Of course that implies that the supposed ‘center’ that is being measured from is bogus.

      • acymetric says:

        That demographic is exactly why I believed Bernie Sanders had a better shot than Hillary last election (I don’t quite feel the same way about this election). Definitely a group that seems to be regularly ignored.

      • hnau says:

        Current political commentary tends to gloss economically left + socially right as “populist / nationalist / Trumpist”. I think this misses a lot of subtlety; in particular I would guess that many who get lumped into that category aren’t Trump fans and aren’t really opposed to immigration as such, though they tend to choose Trump / Johnson over the alternative when push comes to shove.

  56. Arb says:

    In the hope you’re keeping a list of improvements for next time…

    Employment is missing option “Retired”.

  57. Plumber says:

    I was very tempted to put “90”,”95″, or “100” as a guess and to counterbalance the exalted ones but I left it blank as I have never heard/seen any results nor do I know where to get any, I haven’t taken the SAT either, though I did take the PSAT in 1983 with a higher verbal than arithmetic score but I don’t remember what.

    I didn’t include the friend who moved into my Mom’s house at 17 (his mother became a drug addict) as a “social sibling” as I didn’t want to share my room so I stayes at my Dad’s place then.

    My mother said her grandmother was an Ashkenazi Jew and her second husband (my step-father) was, but my Dad wasn’t and my mother was raised Lutheran.

    I’ve seen some “visual trails” but usually only when tired, dizzy, and/or dehydrated.

    I was amused to see that I’m more “feminine” than “masculine” (I’m sure my wife would say “If only!”) which I’m guessing is due to my saying that I really don’t like guns, which is due to (while I have owned guns and enjoyed target practice) I really didn’t like gunfire on city streets and having guns pointed at me.

    • VivaLaPanda says:

      The guns one was hard but I took it as “do you find the idea of guns cool in abstract”. Does the John Wick fantasy of being Mr. Guns Man appeal to you.

  58. Apogee says:

    A couple clarifications to my experience with the survey. Interested in seeing if this happens to anyone else.

    -I’m only able to switch the direction of the moving train by focusing on another part of the image (it seems to be heading toward the point I’m looking at).

    -Caffeine seems to have a delayed effect on me (on the order of 3-5 hours after consumption). I could be confusing it for some other effect but it seems pretty consistent.

    -When I first started taking melatonin, it helped me get to sleep faster but not to wake up earlier, leading to frequent 11-12 hour sleeps. Now it seems to do the opposite: I get to sleep maybe a bit earlier than I would naturally but I wake up much earlier. I’ve only been using it regularly for a few weeks. Like my experience with other sleep-inducing drugs, I find melatonin dreams to be weirder and more easily remembered, though I wouldn’t classify them as nightmares. I currently use it as a short-term hypnotic but I’m going to experiment with using it as directed for delayed-phase.

    • Nicholas says:

      I could switch the direction of the train by changing where in the image I looked, but it would move away from my focus point: if I looked at the tunnel it was coming at me, if I looked at the ‘close’ part of the train, it was going away (into the tunnel). I was not able to see different directions for the Einstein mask, it was always convex.

      I find caffeine changes my heart-rate, but not my subjective feeling of tiredness. Taking caffeine at night will make it harder for me to sleep, but not less desirous of sleep.

      I am not sure melatonin helps me at all, I tried it at 5mg, it seemed to help for several weeks, though my insomnia tends to come and go, so could be coincidence. After several weeks when I had trouble sleeping again I upped the dose to 10, then 15mg without a noticeable effect. Now I take 5mg in combination with ~10 – 25mg benadryl, and this tends to work pretty well, though I think the benadryl is doing the heavy lifting in this cocktail.

      • Lambert says:

        I’m starting to think the mask is actually convex and Scott is playing games with us.
        By which I mean I spent a very long time squinting at the exact reflections of the light in the image, to little avail.

        • Lurker says:

          I can change it from convex to concave at will, so if Scott’s just playing a prank, either my brain or my eyes or my laptop screen are broken.

      • sty_silver says:

        5mg is way too much! Try 0.5mg, and if that doesn’t work, try 0.3mg or 1mg.

        Just recently, I experimentally upped my dosage from 0.5mg to 1mg and it didn’t work out at all, I took quite long to fall asleep. I switched back to 0.5mg and it was fine. From what I’ve read, it seems very unlikely that > 2mg is a good idea.

    • NoRandomWalk says:

      Interesting, for me it’s based on the color. If I look at the red it’s coming at me, white away.

    • Laukhi says:

      I wasn’t able to switch the direction of the train merely by focusing on another part of the image alone. However, I was able to do it by rapidly moving my sight towards a certain direction; the train would then appear to move in that direction to me until I did it again the other way.

  59. DinoNerd says:

    (If American) What state do you live in?

    But not if you merely live in the US?

    And while I’m whining – I don’t “identify with” any race. If you want to know what race people automatically identify me as, you need to ask a slightly different question ;-(

    • iphthime says:

      Agree 100%. I’m not going to do one of these as long as you can pick only one race to “identify with”. Anything I picked would feel like a lie. Please fix this. Seriously, even the Census doesn’t do this anymore.

      p.s. DinoNerd has an excellent point that there are lots of people who live in the US without living in any state…

      • A1987dM says:

        DinoNerd has an excellent point that there are lots of people who live in the US without living in any state

        I think he meant foreigners living in the US, not people living in e.g. DC.

    • thedufer says:

      The question is worded ambiguously, but I chose to read it as an assertion about whether the state is American, not you. In that reading it solves both your problem and mine (I’m American but live abroad).

    • Pink_Creosote says:

      The question is valid either way. Wording it as the race you identify with helps with people who are of mixed race or ambiguous in appearance. Since you don’t consider yourself to be any race you can just leave it blank if you like.

  60. onodera says:

    The answers to the “Sleep Latency” question are too granular. It really needs at least a “how the hell am I supposed to measure *that*?” as an answer.

  61. Dacyn says:

    I put that I have 3 unread emails, since that is how many I have not read, but I have 30 marked in my inbox as “unread” (meaning I have not finished dealing with them yet)…

  62. Uncountably says:

    Did the same/opposite sex of siblings question mean birth sex (assigned gender at birth) for both of them? I answered assuming that was the case.

  63. therm says:

    I’m self-employed in 3 different fields. One of them I make the most money at, one of them contributes to society the most, and one of them I have the most fun doing.

    Which is the most important?

    Also, my family moved from poor to solidly upper middle class over the course of my childhood.

    • Ttar says:

      Also, my family moved from poor to solidly upper middle class over the course of my childhood.

      Then they were probably upper middle class the whole time, unless they went from smoking blunts while watching UFC to drinking California table wine while reading the Economist over that same time period, which is possible but unusual.

      • therm says:


        They went from a net worth of basically zero and income of
        less than $200/week to net worth in the millions and income of hundreds of thousands / year.

        All in 2018 dollars according to some random internet inflation calculator.

        • Ttar says:

          I don’t think we were supposed to interpret “poor” and “rich” quite so directly in terms of income. Maybe Scott could clarify more on next year’s survey.

        • Lurker says:

          net worth in the millions counts as upper middle class?!
          I would have put that firmly in the “rich”- bin

    • Aapje says:

      Given how American-centric this survey is, probably the one you make most money at.

  64. googolplexbyte says:

    Survey done. I choose cooperate thrice, but I was very tempt to defect against my clone.

    I checked after and I was nearly two orders of magnitudes off with the police killings section.

    • NoRandomWalk says:

      I did a quick fermi estimate, got something that was 1 order of magnitude too low, felt it had to be higher so multiplied by 10x. Was annoyed that my intuition was so much better.

  65. blashimov says:

    Are police killings 2 and 3 also supposed to be guesses?

    • blashimov says:

      I answered all 3 as guesses, was off by factor of 5. Sad really. Still working on my fermi estimates and accounting for anchoring bias.

      • DinoNerd says:

        You did better than I did. And it was hard not to look them up before entering my answers ;-(

      • Rana Dexsin says:

        I answered all three as guesses. Without revealing which is which, and subject to conscious-recall error, I recall that I was surprisingly accurate about one (< 0.25 decades off), somewhat overestimated one (≅ 0.5 decades off), and drastically overestimated one (≅ 2 decades off).

        Edited to add: I did something along the lines of Fermi estimates for all of them, not purely intuitive guesses.

    • Dacyn says:

      Presumably, otherwise Scott is just testing how well we can look up numbers.

    • Protagoras says:

      I wasn’t off by more than a factor of 2 in any of my guesses. That’s a Fermi win, right?

  66. JEA says:

    Please be gentle if my answers are wrong.

  67. thevoiceofthevoid says:

    Since I had to close all my tabs to update my browser last night, my “current tabs open” was about an order of magnitude lower than average. Graphing that stat over time for my laptop would probably give you a sawtooth-ish graph with an amplitude of about 100 and a period of about a week. I even have a Chrome extension that counts how many tabs I have and shames me by turning from green to yellow to red as the number gets above about 20!

  68. CyberByte says:

    Thanks for doing this, and for another year of interesting posts! Because I’m a nitpicking asshole, I will now provide a list of things I thought might be improved in the next survey:

    * Move the public/private question to the end (if you need one at the end of the first part, also make one at the end of the second part). That way people can answer it with full knowledge of what information they disclosed. Keep the announcement at the beginning that you will ask this.
    * In the employment question, I’m a bit confused about what to answer as an academic on the research (not teaching side).
    * For people who haven’t taken a professional IQ test or the SATs (e.g. because they’re not American), you could consider adding more options like the GREs or something. I might personally add a question that asks people to estimate their IQ (perhaps while providing links to (conversion) sites that help with this, even in not-ideally-scientific ways). You can always throw away the information if you want.
    * There’s a question about whether we identify as a “Less Wronger or aspiring rationalist” but those seem like entirely different things to me.
    * In the options about psychological disorders in family members, could you clarify if you’re interested in just blood relatives (genetics) or also e.g. inlaws, adopted and step family (environment)?
    * In the question “On average, how many hours do you sleep on a weekend?” I assume you mean on a weekend night, to be consistent with the previous question about a “weekday”, but that’s not technically what the question says.
    * For the “Napping 3” question, I wish you included options for feeling the same (not better or worse). Or perhaps just divide it into two questions, asking about duration and effect separately.

    I look forward to seeing your analysis! Happy New Year!

    • b_jonas says:

      > Move the public/private question to the end

      I actually prefer it in the beginning, and not only because fewer people will skip it this way. Since that question is in the beginning, then some people will precommit to publish their answers and then omit answers to questions that make them uncomfortable or to questions that would identify them. If it was at the end, then people would fill all questions and then decide that they don’t want their answer published. I prefer the first outcome, that of having public answers to some questions, over only Scott having access to the answers.

  69. blacktrance says:

    Bedtime 3
    In a perfect world, where you had no responsibilities and could listen to your body as much as you want, what time would you naturally go to bed?

    I tried this during spring break in college once. I started with a normal sleep schedule of going to bed around midnight; by the end of the week I was going to bed at 7 AM. During this experiment, I’d feel out of it for most of the day and only start feeling alert several hours before I’d go to bed. So ultimately it’s better for me to ignore my body and force myself to go to bed at the same early time.

  70. Jeltz says:

    Aggregate responses are visible after you complete the survey (“See previous responses”). You might want to disable that.

  71. googolplexbyte says:

    I put down 66 tabs open in your browser right now, but it’s actually 5 browsers with 15, 14, 15, 13, 8 tabs open…

    And that’s just counting my desktop browsers not my mobile ones.

    • Rachael says:

      I wasn’t sure whether to put the total number across all devices, or all tabs in all browser instances on the current device, or what. I was doing the survey on my mobile so I entered the number of open tabs on my mobile browser, which is less than the total on my desktop.

    • meh says:

      yeah, this question is just measuring time since last forced restart.

    • Jack says:

      Question only allowed numerical answers so I couldn’t put the actual amount of tabs: “I feel very attacked right now”.

    • nkurz says:

      Similarly, I answered “1”, but that’s because I prefer windows to tabs. I have dozens of browser windows open, but only 1 tab in each. Maybe this is rare enough not to affect the results, but depending on the intention of the question, maybe it should be reworded.

      • chridd says:

        I counted the total of all the tabs in my many, many windows. (I have lots of windows open, and some windows have multiple tabs.)

  72. googolplexbyte says:

    My train got stuck wiggling back and forth, while I was trying to get to switch directions.

  73. fortybot says:

    I pictured a spherical red star (and also what it might look like from earth). So none of the examples were like what I had pictured.

  74. Lambert says:

    ‘Poor’ and ‘Rich’ are not social classes.

    • Ttar says:

      Lumpenproletariat and élite, if you prefer?

      • Lambert says:

        It’s not even that.
        If someone from the lower end of working class gets a load of money, they don’t suddenly become UMC.

        • Ttar says:

          Sure, but people from lower class backgrounds can become higher class by changing social circles and manner as an adult, right? And I’d wager that’s also often associated with a higher income? Plus Scott asks about income separately. I guess I assumed poor and rich in the class section were euphemisms for lowest and élite classes, which certainly do tend to correlate highly with income..

    • Anteros says:

      I was also surprised by this. I couldn’t believe that ‘Middle class’ and ‘Upper middle class’ were followed by ‘Rich!!’.
      Of course, my surprise is explained by my upper middle class Englishness – class divisions when I was growing up had precious little to do with money.

      • Lambert says:

        When I was growing up, my family had comparable disposable income to enterprising blue-collar types.
        But we spend far more time listening to Radio 4.

        Can you go into more detail on socio- vs economic class next year?

        Maybe if we can get Americans to recognise that class exists, they’ll stop sublimating class issues onto racial ones.

        • DinoNerd says:

          My family (Canada, not UK) was blue collar. But my father read so many books that his co-workers called him “the professor”. We would have called ourselves “working class intelligentsia”. Where would we have fit into your scheme?

    • DinoNerd says:

      I found it interesting that “poor” was distinguished from “working class”, and could easily have answered “poor” as meaning “lower working class”, if I hadn’t been paying close attention.

      What would the intended answer be for (a parent who was):
      – a day labourer
      – a minimum-wage-when-I-can-get it worker, frequently unemployed
      – a low end member of the gig economy, working several unpredictable, non-full-time jobs, none paying (much) more than minimum wage, with neither benefits nor security
      – working illegally, and/or at work generally disapproved of by upright members of society, such as prostitution
      – on welfare
      – on disability

  75. Evan Þ says:

    That suicide question has shockingly different answer choices from the previous questions’ pattern! I was about to select the fourth option presuming it was (following the pattern) “I haven’t attempted it, and neither has anyone in my family,” but then I noticed what the fourth option actually was…

    • eucalculia says:

      It’s so jarring that I’m tempted to believe it’s deliberate. Like, there is going to be some non trivial number of people who have never suffered from depression or any other mental illness, but then say they have attempted suicide. This is some sort of measure of survey taking laziness (it still works as an experiment, even if it isn’t deliberate)

      • Deiseach says:

        Like, there is going to be some non trivial number of people who have never suffered from depression or any other mental illness, but then say they have attempted suicide.

        Which can happen in reality. Not everyone gets a neat diagnosis of “Ah, you’re depressed” and based on my own personal experience, I did get the “never self-harmed/attempted suicide? then you’re not depressed” from my own doctor in spite of the fact that the reason I was asking for anti-depressants was because I was seriously worried about suicidal ideation and increasing attraction to the idea of killing myself. This would lead some people to at least attempt the ‘cry for help’ suicide attempt to finally get the ‘okay, we’re taking you seriously now’ reaction.

        Or someone can attempt/carry out suicide even if they haven’t previously been suffering with mental illness/depression. A cousin of mine killed himself (though the verdict was brought in as ‘accidental death’ to save the family’s face, we all knew it was deliberate) shortly after his marriage had collapsed and his wife had left him. He was certainly depressed after that happened, but it wouldn’t have been a diagnosis beforehand. And if he survived, he could well answer “No, never suffered mental illness; yes, did attempt suicide” on this survey.

        There will be some amount of Lizardman’s Constant, but not every inconsistent answer is going to be ‘survey taking laziness’.

    • Elementaldex says:

      I also almost gave the wrong answer on the suicide question due to this.

  76. thevoiceofthevoid says:

    Concerning depression/anxiety/etc:
    “I think I might have this condition, although I have never been formally diagnosed”
    “I have family members (within two generations) with this condition”
    are not mutually exclusive, and though I assume the former is intended to override the latter it’s not entirely clear.

    ETA: Perhaps a good rephrasing would be “I do not have this condition myself, but I have family members (within two generations) with this condition” if that’s what’s intended

  77. Calion says:

    I have no idea what six options are being referred to in relation to this tweet:

    Is my iPhone not showing them, or something?

  78. Three Year Lurker says:

    All of the questions about whether the survey taker has a condition or someone in their family has it need an option for people who don’t know about conditions in their family.

    As written, there is no way to differentiate between someone whose family actually lacks the condition, and someone who does not know the conditions of everyone in their family.

    With the “within two generations” part, that set could be over 50 people, if it’s meant to include biological aunts, uncles, and cousins.

    • Garrett says:

      I’d also add that it’s unclear for “family member has official diagnosis” and “family member is strong suspected of having” cases.

      • Lurker says:

        that tripped me up, too. I went with choosing the diagnosis, since I’m pretty certain, even if there’s no way to confirm that now.

    • thepenforests says:

      Yeah, fwiw I left most of those blank for that reason. (I get it though, I imagine this survey is already kind of a nightmare to administer, you can only cover so much)

  79. DragonMilk says:

    So red star question, I imagined something like this, but I put 6 as it was the closest.

    • blashimov says:

      I imagined what you might see through a very good telescope (as opposed to but similar to Mars with the naked eye)

  80. caryatis says:

    How should I answer if I have been diagnosed with X in the past, but it no longer applies? (Long long list of mental health diagnoses, but the most recent is now 16 years old.)

    • Anteros says:

      I treated all diagnoses as current. Some of mine, too, were over a decade ago.

      • caryatis says:

        Yeah, that’s what I’ll do unless someone convinces me differently in the next few hours. But it does ignore the gap between, for example, “ever depressed” and “currently depressed.”

        • Anteros says:

          An expression that kept coming into my mind while I filled out the survey was ‘Don’t let the perfect become the enemy of the good’ which was helpful, but ironic as an inability to do that is generally my current mental health diagnosis….

  81. One problem with the IQ question. The only score I have is from my high school, one of the students having copied the list of student IQ’s and circulated it. The school is run by the University of Chicago, so I am sure the figures were “scientific,” but I do not know what definition they were using. Mental age/physical age for children — I was in the same school for most of K-12 — yields a different distribution than a definition based on mean and standard deviation for adults.

    I gave the figure anyway, but don’t know if it fits the question correctly.

    • gryffinp says:

      I left the IQ question blank, as the last IQ tests I’ve taken were so long ago that I don’t really trust my memories of the results.

      What is a good way to take a “professional” IQ test these days, as a random civilian anyway?

      • sty_silver says:

        Mensa’s tests might be the cheapest option.

        • Rachael says:

          Except Mensa don’t use one of the standard normed tests. They gave me a certificate with a number on saying “This is a true IQ rating”, but didn’t specify which test it used. I gave that number as my IQ for years, including on SSC surveys, before I realised this. The actual figure is lower.

          • Anteros says:

            I’m in the same boat – the only number I have is a Mensa one. By how much should I reduce it to get a standardized score? How much dumber am I going to get overnight? 😩

    • Loriot says:

      As far as I know, the only IQ test I’ve ever taken was as a young child, which is notoriously unreliable, but I figure that if Scott wants to ask silly questions, he can get silly answers.

  82. meh says:

    Possible limitation, SAT scores are subdivided by section, but preparation methods are not.

  83. A couple of minor problems:
    “Work status” did not include “retired.”

    “Religious status” “as of the last time your family practiced a religion.”
    It wasn’t clear whether “family” meant immediate family or if I was supposed to go back until some ancestors, in my case grandparents, practiced a religion.

    • Rachael says:

      I thought it did? Something like “Retired or post-significant income earning”.

      I have a similar issue with the religion question. I’m the Christian child of atheist parents who are presumably descended from Christians.

    • Rachael says:

      Oh, sorry, “work status” doesn’t but “income status” does. I misremembered.

    • DinoNerd says:

      Yeah, I noticed those too. I’m not retired yet, but I’m looking forward to retiring soon.

      And the religion thing was weird – I presume you meant “birth family” in the “as of the last time your family practiced a religion” question, and were looking for something more like “what were you raised as” than “what did your kids convert to” or “what does you wife practice” for those not currently practicing a religion.

      In my case, I’m somewhat agnostically not practicing the religion practiced by other(s) in my household, having been raised in an atheist household that was rebelling against a different religion, into which I was officially baptised for the sake of peace with religious grandparents. One of my sisters is doing the same as I, except the religion she’s not practicing is somewhat closer to the one our grandparents practiced.

  84. cvxxcvcxbxvcbx says:

    I took the SAT during the years when it was out of 2400.

    • acymetric says:

      It asks for each section separately so that shouldn’t be a problem 🙂

      He never asks about the writing section, just math and verbal (reading).

  85. A1987dM says:

    * On the “Work Status” question, I answered “Academics (on the teaching side)”, even though I’m not actually teaching any course, because my job is otherwise identical to that of people who are.
    * In questions such as “Subreddit”, “Discord” etc., in order for the answers to be mutually exhaustive, “don’t want to” has to be interpreted as “don’t want to now”, but the most natural interpretation would otherwise be “don’t want to ever”. Next time, consider adding “haven’t gotten around to”, either to “don’t want to” or as a separate answer.
    * In the “Immigration” question, do you mean de jure or de facto? It is possible to simultaneously believe that more people should be legally allowed to immigrate but more effort should be spent to keep everyone else out (or even vice versa, in principle).
    * “Income”: do you mean before or after taxes?

  86. vicoldi says:

    In the games, I don’t actually want to grow the commons. I feel that the money is in a pretty good place in Scott’s hands. This kind of reasoning might affect other people’s decisions too.

    If you do something like this next year, I would consider declaring that you will donate the remaining budget to the Scientologists. I would be very interested to see how much it motivates people to choose Cooperate.

    • Nicholas says:

      I said ‘defect’, but not because I want a larger chance of winning, I’m assuming a defector will win and therefore want to minimize their winnings. Punishing a defector feels like cooperating to me, but I had to pick ‘defect’ to do it.

      • Sandpaper26 says:

        Go one more level meta: if I anticipated that some would act as you do and wanted to reward that behavior, I would choose “cooperate.” Maximize the winnings of someone who has tried to minimize the winnings of someone trying to win. Out of spite, I guess.

  87. Rob K says:

    Is there a canonical answer to how to handle the income questions wrt family vs individual income?

    Or, rephrased: should I give my household income and charitable donations, my income and some fraction of our household’s charitable donations, or half of the household number for both? Or something totally different?

    • Elementaldex says:

      I was confused about this also and gave just my personal income which is considerably less than household income.

    • methylethyl says:

      Yeah, this. I’m a “homemaker”. I listed family income, because otherwise it sounds like I am starving in the streets. But it feels weird to say I’m making (x household income) as a homemaker– then it sounds like I’m a professional mommy blogger or tupperware saleslady or something. Also not true.

  88. Randy M says:

    Oh, now here’s a serious complaint. On the traffic cop question, you need a response for “Yes, and I was guilty, and it was unreasonable for them to pull me over.”
    Stupid 80 mph speed limit on empty roads. :/
    (Also, multiple of the choices might have been valid at different times)

    • Nicholas says:

      Or the cop made up some BS about me not signaling a turn, but when he pulled me over my tags were expired. I was guilty even though the pretense for the stop was false. My wife makes fun of me for over-signaling, like signaling turning into a parking spot in a parking lot, or signaling a turn when I’m a turn only lane: there’s no way I absent-mindedly forgot to signal a turn on the actual road.

      • methylethyl says:

        @Nicholas: There’s nothing wrong with over-signaling. I learned this the hard way: while trying to turn left from a side-road, where I was not very visible, onto a main road, at night, with virtually no traffic, I trusted that the one car going by on the main road, not signalling, was going to keep going straight, and I could just turn after them. Nope. They were turning ONto the road I was turning OUT of. But no blinker. I dinged the back of their car. My insurance rates went through the roof. Now, I signal EVERYTHING, just assuming there may be a car or pedestrian or bicyclist around that I didn’t notice. And I don’t trust people to *not turn* when they do not signal. Too many people don’t know what turn signals are for.

    • RC-cola-and-a-moon-pie says:

      “Gettier’s police stop.”

  89. Randy M says:

    Scott, here’s a minor little observation. In the mental health section, the responses go from most to least serious (“I have” to “I don’t have”), until the last question, where it switches to least to most serious (“I never have” at the top).
    I almost misclicked since the previous several questions had the same answers and I wasn’t reading them closely by that point.

    Also, when you said “close your eyes and think of a red star” I saw this”

  90. zeleza says:

    Uh, sorry to contradict you Scott, but I just managed to untick a box in a question on the form so it seems like Google has just fixed the issue…

    • thevoiceofthevoid says:

      You can untick a box but can’t unselect a radio button

      • googolplexbyte says:

        Couldn’t you always use boxes and not radio buttons then?

        • thevoiceofthevoid says:

          Not really; some things really are single-select (e.g. “rate X on a scale of 1 to 5”) and having checkboxes were there should be radio buttons would probably be more confusing and lead to more errors.

  91. Ragged Clown says:

    Whenever I read analysis of attitudes to religion surveys on religion-friendly sites, they always seems to interpret “spiritual” to mean “belief in supernatural forces” and they marvel over all the spiritual atheists. For that reason, I answered “not spiritual”.

    If I were to interpret “spiritual” to mean “I sometimes feel at peace with the universe” or “I sometimes feel like I am walking on sunshine”, or “I believe in Love”, I would answer differently.

    • sty_silver says:

      Eh. I interpreted “spiritual” as “interested in mindfulness” and answered yes. I categorically don’t believe in anything supernatural.

      It’s bad news if we have such different ideas of what the term means.

  92. Dan says:

    On the red star question, I was getting a blank page for the twitter link so I have no idea what it was asking and had to leave it blank. Not sure if this is a general problem (twitter down? Tweet was deleted?)

    On the masturbation question, I know there are survey-ological reasons for asking “in the last week” rather than “in a typical week”, but that assumes that the last week will have been typical for most respondents, whereas this past week is probably the most consistently atypical week of the year. But maybe you realized that and are trying to determine whether being home for Christmas and sleeping in your old bed makes you more or less horny. 😉

    • Lurker says:

      I got to see the twitter page but no matter what I did, I couldn’t see the picture that was supposed to tell us which of the answers to choose. maybe next time include the picture directly or host the picture yourself so the link is reliable could work better?

  93. A Lorenzen says:

    If I am currently studying for my bachelors, I have not completed my education and highest degree earned is high school, right?

  94. Machine Interface says:

    Since we’re complaining about stuff:

    The moral views question could really use a nihilist/anti-realist option!

  95. a real dog says:

    About the “Overall artifacts” question, the picture and scale are backwards! It’s really easy to get it wrong.

  96. Luna says:

    The aphantasia questions could have some more nuance. They don’t capture whether the visualisations are part of the same visual field or ‘off-screen’, or either way per choice. Don’t capture stability of the visualisation.

    They don’t capture variance over time, (did you become / fix your aphantasia?)

  97. Luna says:

    > Moving Train

    none of the above, the frame rate is so low it looks like a slideshow. There’s no illusion of movement.

  98. ShemTealeaf says:

    Suggestions about a couple of these questions:

    1) I checked the Washington Post police shootings database after taking the survey, and I see that they break out “toy weapon” and “vehicle” as separate categories from “unarmed”. I would consider people with toy weapons to be unarmed, and people with vehicles to be unarmed unless they are actively using the vehicle as a weapon. Lumping those categories in with “unarmed” seems to change the numbers very significantly.

    2) I take melatonin in the afternoon, based on the suggestions from you (Scott). As a result, I answered ‘no’ to the question about taking melatonin before bed, but I am probably getting a benefit from taking it in the afternoon.

    3) It might be helpful to clarify that the “think of a star” question is referring to a five-pointed star, rather than an actual star in the sky.

    • Ttar says:

      Ok, imagine it’s dark outside, and you’re holding at gunpoint someone you suspect of being a criminal who is not cooperating with an arrest and instead is behaving belligerently or erratically, and they pull a black-spray-painted airsoft pistol out of their waistband and point it at you. Do you hold your fire and shout “hey, is that thing real?” If so, you probably shouldn’t pursue a career in law enforcement, because more times than not you will get shot in that situation. Pulling a gun-shaped object on a cop is functionally equivalent to pulling a real gun on a cop.

      • ShemTealeaf says:

        Sure, but it depends on context. I would consider someone with a toy weapon to be unarmed unless they are actually brandishing it or otherwise behaving in a threatening manner.

        • Aapje says:

          Isn’t that same logic true for real guns? I don’t think the police is justified in shooting someone who carries a gun, but never comes close to using it or seeming so.

          • Lambert says:

            you can go from ‘not seeming to be about to use a gun’ to ‘shot someone’ in a couple of seconds.

          • Aapje says:

            Sure and that may be a reason to point a gun at a person, but not to shoot them.

            Also, toy guns and real guns aren’t always easily distinguishable, especially as some people intentionally carry real-looking fake guns.

          • Lambert says:

            I’m not defending the US police but guns, especially pistols, take longer to kill someone than people think they do.

            Shame how it turns out that a large minority of people wielding immediately lethal ranged force in a way that’s easily concealed tends to erode trust like that.

          • Aftagley says:

            I’m not defending the US police but guns, especially pistols, take longer to kill someone than people think they do

            I don’t understand what you’re saying. Are you implying that a gunshot wound doesn’t always kill you quickly or that it takes longer to draw and aim than people think?

            If it’s the latter, it’s still under 10 seconds, substantially lower if the person doing the drawing and aiming has put in a minimal amount of practice or if you’re at especially close range.

          • Lambert says:

            The former.
            You shoot the perp and they shoot you back.

      • Aftagley says:


        Heck, if the airsoft gun even gets out of the waistband before the suspect gets some added ventilation that cop is acting insanely negligent.

      • thisheavenlyconjugation says:

        What proportion of incidents where cops killed people with toy guns are similar to that which you describe, rather than (for instance) involving 12-year-olds who aren’t suspected criminals?

        • Ttar says:

          I’m saying that shooting 12 year olds with toy guns probably often looks like shooting a 5’7″ 195lb male pointing an object visually identical to a standard 9mm handgun at you, while you are responding to a 911 report of a guy walking around pointing a gun at people. Considering that the example you implied is kind of the toxoplasmic scissor case.

          • thisheavenlyconjugation says:

            You have distinctly failed to answer my question. For that matter, you haven’t even given evidence that the situation you describe (cop hesitates before shooting someone because they think the gun is a toy and gets shot because of it) has ever occurred.

          • Evan Þ says:

            Perhaps it never occurs because every cop is taught to preempt it by not hesitating.

      • Nicholas says:

        Police are given body armor and specialized training; if they can’t put their scaredy-cat reflexes on hold for 1 second to evaluate a situation before murdering people, they shouldn’t pursue a career in law enforcement, because actually, more time than not they end up killing the very people they’re supposed to be protecting (by a 20:1 margin). ‘Law enforcement’ as a category doesn’t even crack the top 10 most dangerous jobs in America, yet they are free to murder anyone who they claim made them feel afraid. And don’t give me the old ‘a few bad apples’ chestnut, the rest of ’em aren’t stopping those bad apples, so they’re all complicit.

        • Aftagley says:

          Police are given body armor

          Well, some police are given body armor. It varies by locality. Body armor is also good at taking one or two chest shots, it does very little for sustained fire or shots that aren’t aimed squarely at the chest. They way I’ve heard it summed up is that if you’re relying on body armor, things have already gone terribly wrong.

          and specialized training

          What specialized training makes getting shot less terrible? They’ve been given training, yes, but it’s almost entirely focused around how not to get shot and, sadly, the most efficient answer to this question is almost always “shoot first.”

          if they can’t put their scaredy-cat reflexes on hold for 1 second to evaluate a situation before murdering people, they shouldn’t pursue a career in law enforcement,

          Forgive me, but i don’t think your standard is actually “police should wait one second before engaging in lethal force.” Depending on the situation, one second is either a very long or very short amount of time – it’s long enough for someone to fire a pistol twice, but maybe not long enough to parse that the spray-painted air soft pistol isn’t actually a threat.

          ‘Law enforcement’ as a category doesn’t even crack the top 10 most dangerous jobs in America, yet they are free to murder anyone who they claim made them feel afraid.

          I don’t think you’re seeing the correlation here – “Law Enforcement is less dangerous than logging” and “law enforcement officers will definitely kill you if they perceive you as being a threat to their life and safety” are fully related factors. When there are so many freaking guns in circulation, there is either a world in which policing is safe profession that kills innocent people, or a dangerous one that doesn’t.

        • thetitaniumdragon says:

          1) They’re not “scaredy cats”. Guns are lethal weapons.

          2) Body armor isn’t magical. Being shot while wearing body armor is still very painful, and is often lethal, as body armor only covers the torso, and most shootings happen at extremely close range, so there’s zero guarantee they will shoot your body armor even if they do shoot you.

          3) There are often other people around.

          4) Most cops don’t wear body armor all the time.

          5) Anyone who pulls a gun on a cop deserves to be shot. In fact, anyone who pulls a gun on *anyone* deserves to be shot. That includes drawing a fake gun that will look like a real one.

          Anyone who thinks that you shouldn’t shoot people who pull guns on you probably shouldn’t have any opinions about *anything*, because they don’t respect the most fundamental of human rights – the right to be alive and free from harm from others. That shows an extremely high level of sociopathy on their part, as if you don’t respect that, then you basically respect nothing.

          When you pull a gun – or something that looks like a gun – on someone, you’re forfeiting that right. In fact, you forfeit all rights until you are no longer a lethal threat to other people.

          Moreover, almost all police shootings are legal and justified – less than 5% of police shootings are unreasonable.

          And they aren’t supposed to protect people who are trying to maim or kill other people; they’re supposed to kill them to protect the rest of us.

          The police should kill vastly more criminals than vice versa because the police are better trained than criminals are, and the police are a higher grade of personnel than criminals are.

          • Nancy Lebovitz says:

            If a police officer pulls a gun on someone, should the police officer be shot?

          • Evan Þ says:

            Anyone who thinks that you shouldn’t shoot people who pull guns on you probably shouldn’t have any opinions about *anything*, because they don’t respect the most fundamental of human rights – the right to be alive and free from harm from others.

            I believe that signing up to be a police officer means you accept some limits on that right while on duty, for the good of society. That’s one of the things we should be paying police officers for.

          • acymetric says:

            @Nancy Lebovitz

            No, because police are infallible.

    • acymetric says:

      I was way off on total killings. I was pretty much spot on with unarmed, going by your definition (which would also be my definition, I have some suspicion that “unknown” probably parses to “unarmed” in a significant number of cases). A slightly different point (because it doesn’t equate to unarmed) but I also have some suspicions that “other” and “knife” were probably not consistently warranted.

      I overestimated the number of police shot in the line of duty by a fair bit. I was closer to the total number of deaths.

      • acymetric says:

        Oh wait, I was looking at overall unarmed deaths, not just for black people. Even adding up every category except “gun” I was way over on that one. Oops.

        I do wish for mental illness it broke out yes, no, and unknown as 3 separate categories rather than lumping no and unknown together.

    • thisheavenlyconjugation says:

      Fun figures: in 2019, at least 25 people with toy weapons were killed by police. In comparison, 46 cops were shot dead in total.

    • thetitaniumdragon says:

      Vehicles are more dangerous than guns are when used as weapons; someone deliberately hitting you with a car, or simply not caring about you and hitting you with their car is very likely to kill you. In fact, I’d wager that being deliberately run over by a car is actually more likely to kill you than being shot is, as being shot has only about a 20% fatality rate, if that.

      And toy guns are counted separately because 1) some people threaten people with toy guns to rob them and 2) toy guns often look like guns and are a good way to get shot if you brandish one at the police/another person. It’s not the same as someone being unarmed.

      They break them out (correctly) because they’re different things.

      Someone who is in a vehicle and refuses to stop is a danger to the public. Someone who brandishes a toy gun at someone is committing a felony, and a felony which looks like imminent lethal danger at that.

  99. Björn says:

    Concerning the “thinking about your tongue” question: I saw a speech therapist in 2017 for tongue posture problems. When I read the “think about your tongue” question, I just put my tongue in the correct posture, and then it wasn’t uncomfortable at all. So maybe the prompt “your tongue is feeling uncomfortable” actually screens for incorrect tongue posture.

    Concerning imagining the red star: I imagined it even more detailed, like this.

    Concerning the sleep time questions. I sleep more than I would like to, and stay up too late, too, which can be seen in the context of atypical depression. I missed options expressing that I sleep too much.

    • Ttar says:

      Re: red star, your link is precisely what I imagined. Interesting. For some reason this has… Soviet associations in my brain? Is this just a super available image of a red star that I’ve seen over and over?

      • Bugmaster says:

        It’s super-Soviet, and it is almost exactly what I pictured in my head; except that my Soviet star also had a gold outline. I was raised in the USSR though, so YMMV.

    • googolplexbyte says:

      I have a hard time imagining the colour red. Yellow and blue stars are much easier. Green is trickier but doable.

      Question: Can you imagine colour gradients?

      • onodera says:

        I can imagine the standard MS Windows colour picker.

      • Björn says:

        Yes, I can. Btw, with the star, when taking the test, I tried how far I could push my visual imagination, so I imagined the red star attached to the head of Edward Norton rotation over a lake in the mountains without any trouble.

      • anonymousskimmer says:

        I have a hard time imposing color on the shape with my eyes closed, as the dark gray background that occurs with closed eyes swamps out any color.

  100. eterevsky says:

    I wonder if the experience of art appreciation counts as “spiritual”. It kind of fits the definition (“profound”, “outside normal life”), but at the same time it doesn’t imply any mysticism. I ended up answering “no”, even though I do sometimes experience profound emotions say from listening to live music.

    Also, I introduced some noise to the police questions, because I moved from one country to another, so some of the answers were about events in one country and some about another.

  101. Viliam says:

    Income in USD — pre-tax or post-tax?

    Autism in children — does it include asperger?

    • DinoNerd says:

      Income in USD — pre-tax or post-tax?

      almost always pre-tax, in any survey I’ve ever seen

      Autism in children — does it include asperger?

      in other places in the survey, he explains autism as including asperger’s, so presumably “yes”.

    • a real dog says:

      There’s a saying in my country that declaring income pre-tax is like declaring penis length including the spine.

      FWIW I put in post-tax.

    • alexschernyshev says:

      Indeed. I post this every year. There needs to be a gross / net tax option. In certain countries employers are not allowed by law to write down the pre-tax salary anywhere in the contract or let the employee know them. It makes comparing numbers useless across countries and within them as well, since some people will interpret the question differently. If course, it’s very difficult to compare anyway with purchasing power and all, so the point is maybe moot…

  102. Ragged Clown says:

    Your police questions made me realize how much I appreciate British police. Being pulled over by police in California was always super stressful, even for minor traffic infractions. I was detained by British police for youthful delinquent behaviour after a lengthy chase and they were quite nice about it. “My wife washes my shirts every day and she hates it when little wankers like you make me get them dirty”. This makes my answers to police questions misleading because I answered about two different systems of policing.

    I would never, for example, ask an American policemen for directions but am happy to chat with a British policemen. I imagine that American police are trained to be hostile and to expect hostility (it’s better to be feared than respected?). I imagine British police encounter much less hostility and get more respect as a result.

    • DinoNerd says:

      Oddly, my personal experiences with US police officers have all been good. It may help that they have all involved (a) being stopped for vehicle issues (burnt out lights) and (b) them chasing a burglar across my property. And I don’t look like a potential perp – totally wrong social signals.

    • Nancy Lebovitz says:

      I’m not sure how much it’s formal training and how much it’s a culture they pick up from other police.

      • Lambert says:

        UK police culture came from the Metropolitan Police, which was founded quite explicitly to be an ethical community police force.
        The idea of an armed police force that might crush protests and prop up unpopular governments (e.g. the napoleonic gendarmerie) would never be accepted in London.

        Of course, one can debate the extent to which the police here live up to their principles.

    • ilikekittycat says:

      My experience, as a neither obviously white trash or high class person in an American red state (situations ranged the gamut from dealing with a burning house, to breaking up a loud party with drunk minors, to political protests):

      Default, by myself or in a homogenous white group: Helpful and chatty police. My only international police experiences are Alberta Canada and Berlin Germany and the police consistently acted the same in each place as they do in America when I’m by myself. The chat is common language in regular tone, like a principal talking to a student

      Hanging out with black friend, even dressed well and speaking eruditely: immediate suspicion, switching to hostile/”rules enforcer Karen” mode within 60-90 seconds, calling over any other available officers to circle us/stand behind our line of sight, shoulders back + hands on hips posture. The chat is always “formal register police jargon” where they’re like “We have ascertained a kinetic tactical situation with a fleeing perpetrator and as a consequence we must detain you for questioning” etc. etc.

      My personally positive survey responses would seem to indicate “police are better and friendlier than the newsmedia implies” but in my real experience, the shocking delta between experiences is more distressing than the stereotype of uniformly and equally hostile, authoritarian police would be. Its not even like a 90 negative/10 positive split, it’s clockwork 100% positive reaction or 100% negative reaction based on racial composition. The only social justice type issue I’ve (anecdotally) felt no movement on in 25 years of independent adult life

    • thetitaniumdragon says:

      I’ve never had any problems with American police, and they’ve always been polite and friendly towards me.

    • JRM says:

      I work in a business with a lot of contact with cops. Like a lot of things, where you live even within the US – and sometimes different jurisdictions very close to each other are very different – makes a difference.

      I was once detained by the police for at least an hour. At one point early on, the cop next to me said, “You’re not helping yourself.” Like many, I had the right, but not the ability, to remain silent.

      I was highly satisfied with the cops at the scene. I was not satisfied with the entire process.

    • imoimo says:

      Didn’t the survey say to only answer concerning your primary country?

  103. sharps030 says:

    The America-centrism of this survey is intense.

    An example that bothered me a lot: “A relatively southern climate” is *not* the phrase you’re looking for on the topic of how extreme seasons are: you mean “equatorial”. Cheers, the entire &*$F{*’n southern hemisphere.

    (Although I suppose asking for latitude as well will help disambiguate.)

  104. OrangeInflation says:

    For the question about the red star, I was surprised that there was no option for picturing a red circle, as in a red-colored star seen through a telescope.

    • eterevsky says:

      I actually imagined something like this.

    • Dacyn says:

      Such an option would merely provide information about the question “how does the survey-taker interpret the word ‘star’ ” rather than about their visualization ability. And once you see that it is supposed to be about a five-pointed star, you can just do the visualization experiment again. (Though it would probably have been better to just state “five-pointed” upfront.)

      • Michael Watts says:

        I didn’t see anything in the survey question or the linked Twitter thread explaining what the question was supposed to mean. There were many complaints in the Twitter thread over whether what was being asked about was

        1. A pattern in your visual field, something you actually see in the same way you see e.g. afterimages;

        or 2. An imaginary representation.

        So, prompted by the survey, I imagined a red star without bothering to close my eyes. Then I clicked through and saw that the question wasn’t necessarily asking about that at all, and there was nothing resolving the meta-question one way or the other. So I declined to answer the survey question.

        Why are you assuming that the question is asking about visualization ability, as opposed to whether and how much visualization effort affects the perceived visual field?

        • dentalperson says:

          Same here, I wasn’t clear on the task. The SSC survey question didn’t imply to me that I should pick the actual red star image only if I had a visual experience that was the same as looking at an image of the red star. I thought it was just about the ability to experience imagining a red star, which I guess everyone can do.

          Only later when I went back and read the twitter thread, I realized it was about the complete visual experience.

          • DinoNerd says:

            Ditto. I cannot imagine visuals. IIRC, there’d been a previous question asking about that in particular. So with that priming, I thought the question was fishing for detail level among those who can imagine visuals.

        • janrandom says:

          I also skipped the answer for comparable reasons. But even with the explanation to imagine a red star (given a reference red star) what I would see in my mental imagination would not fit either. It is basically a transparent red star (#ff000000) on transparent background (#00000000) if that makes sense. It is visual – at least in the sense of being able to do geometric operations like rotation/translation. Colors are hard but I can make them come up with some concentration.

    • Jack says:

      I imagined roughly this. Then I thought “that’s what the Sun looks like so a red star up close is probably… redder?”. Then I tried to imagine a redder star, then I imagined a small point or dot as might be visible in the night sky, but reddish pink, then I thought “oh wait that’s Mars or something”. But since it seemed the question was about the vibrancy of my visual imagination it had become clear by that point that the answer was six in any case.

    • thetitaniumdragon says:

      I took the survey, then searched this post for this response.

      I imaged a red star in space as well – not a circle or star symbol, but the sort of twinkly thing you see in some photographs through telescopes.

    • Elementaldex says:

      I envisioned the sun but red and with lots of solar flares. So I skipped that question…

      • Senjiu says:

        Hmm, I was kinda surprised by the picture because none of them actually showed the star in space that I envisioned (I thought of the sun and then thought “okay now take a much bigger one and put the sun next to it for comparison”.. and then the twitter picture was completely different) but since I envisioned mostly red I took option #6 since that was the reddest one of the bunch and therefore the closest one to my imagination.

        But it was like “picture one to five are 2% alike and picture 6 is 3% alike to what I imagined” or something like that.. none of them were close, the last one was just a small step closer.

    • Dictatortot says:

      I pictured an actual swirling ball of plasma–like Betelgeuse seen from something under an AU away. Quizzes can be rough on the literal-minded.

      • Lambert says:

        You underestimate how big a supergiant is.
        That thing’s an estimated 7AU in diameter. Comparable to the orbit of Jupiter.

    • Jeff R says:

      I actually pictured a red dwarf star, so none of the options was even close. Not sure if that’s my bad or the survey’s.

      Edit: oh, I see a bunch of other people did the same. Now I feel like less of an oddball.

    • imoimo says:

      I’m still a bit dubious about that question. I can firmly get the “concept” of a red star in my mind, and it even has a specific shade of red and sits at a specific place in front of me. But I don’t “see” it the way I see a picture. I assume this makes me a 1. But I can’t tell if some people who choose 6 are like me, just less nitpicky.

      • doubleunplussed says:

        Yeah it was really hard for me to tell how to answer, because I could imagine people having my experience could answer at either extreme depending on how literally they take things.

        Of course, measuring how literally people take this visualisation business is probably the point of the question, but it suffers from circularity given that answering the question is subject to the very issue it’s trying to measure…

        In any case, I can fairly clearly see a bit of a star—like, one or two star-points—but not all at once. I have move my virtual eyes around to see the whole thing. Most of it is not super ‘visible’ in the same way that things in my peripheral vision are not ‘visible’. They take up space and have the right shape, but are kind of provisional and subject to revision until I direct my attention at them.

        And colour seems really hard. “It’s red” seems more like a label than an actual experience, unless I picture something well-known like the New Zealand flag, then those stars are definitely red.

      • Shpoon says:

        I can see there are quite a few comments about the red star question – I also wanted to express this sentiment. A friend and I who took the survey both said our visualization corresponded to image #6. But after thinking about it a while, I couldn’t shake the feeling that the star *actually* resembled #2 or #3 more closely. My friend later agreed this is about his expetience – when we visualize the star, we can ascribe a fair bit of solidity and constancy to the shape. But the more I focus on the actual visual component the less substantial it seems to become. So in a weird way, I think I assumed my visual corresponded to #6 but thinking more deeply has made me question whether my visualizations are actually coherent. The strongest impression I am left with is that “my mental visualizations have the property of seeming coherent and substantial, while becoming much more incoherent and insubstantial once I focus more directly on them.” In some regards sounds kind of like dreaming, eh?

  105. Benjamin Ikuta says:

    Once again, I am faced with having to choose between identifying as white or asian…

    Why can’t I choose both?

    I guess I should choose white, since I was raised in a mostly white culture?

    Also, is there a way to save my answers, for my own reference?

    I work in retail. Is that considered white collar, or blue collar?

    Am I considered to have completed my formal education if I’m not currently in school, but I intend to return to university in the near future? This is making me feel bad about myself…

    • vaniver says:

      Also, is there a way to save my answers, for my own reference?

      Last year, I just marked mine as public, and then found them in the public dataset.

      I work in retail. Is that considered white collar, or blue collar?

      I think that’s called pink collar.

      • Benjamin Ikuta says:

        Thanks for the reply. I saw that article, but I don’t think it was an option in the survey.

  106. Florent says:

    I and a number of other responders thought that the “growing the commons” question was not actually growing the commons: it’s a winner-takes-all situation where you can increase the money that the winners takes. As a leftist, I would rather punish the rich guy while keeping the median income stable.

  107. OutsideContextProblem says:

    Some general comments and observations:
    The security/keyword system is broken, but does offer some interesting psychological insights.

    I did not answer large parts of the survey because I have no confidence it will be kept private and non-ascribable (see above), and truthful answers to many of the questions could have a significant impact on my life.

    The ‘professions’ section seemed to list academic disciplines, rather than professions (with the exception of IT monkeys for some reason). My own profession was not listed.

    Many, many of the questions were quite obviously american-centric to the degree that answers from non-usians will not just be wrong, but misleading. As examples: ‘professor’ means different things in different places, and understanding of class varies dramatically. I would have been able to tell the nationality of the author of the survey just from the list of possible ethnicities to choose from.

    I would have liked more places to comment, qualify, or answer multiple times because I felt the ‘most correct’ answer was often misleading.

    American Police shoot and kill far, far more people than I would have expected. This made me even more irritated with this article for not providing base rates then when I read it this morning. It also made me even less comfortable with plans to further arm our own police force.

    • elephantower says:

      totally irrelevant but are you the predictit commenter?

    • b_jonas says:

      OutsideContextProblem, sharps030, Aapje: Yes, the questions are america-centric. This is a feature, because more than half of the respondents are american, according to the 2019 survey.

    • doubleunplussed says:

      What’s wrong with the keyword system? Seems fine to me.

  108. Kuiperdolin says:

    For the medical conditions, I assumed you had to pick the highest one relevant. Because 3, as written, is compatible with either 1 or 2.

    • Rachael says:

      Yes, this bothers me every year and I’ve commented on it before. They should be checkboxes rather than radio buttons.
      I ticked “I might have” for autism, and this year “I have an immediate family member with the condition” can be inferred from later questions. But obviously that won’t work for other conditions.

  109. localdeity says:

    I took the SAT at age 12 and didn’t retake it later, so my scores are from then. Generally, I suspect asking for age on the SAT question would be useful.

    • I agree on age.

      But you should also ask the date it was taken, since that tells you if it was before or after the test was renormed.

    • Evan Þ says:

      Also, I remember taking a couple practice SAT tests back in high school, but I’m not sure anymore just how long I spent preparing for it.

    • DinoNerd says:

      Yeah. All my test scores are from long long ago. I’d either do worse now from age, or do better because they’ve been renormed (dumbed down). For what it’s worth, my GRE (taken 20+ years after the SAT) scores were slightly higher than my SAT scores, which ought to be impossible.

      • anonymousskimmer says:

        GRE math is typically easier than SAT math. GRE verbal and SAT verbal are typically equivalent.

        At least this was the case when they both used 200-800 point scales.

    • AM says:

      Related concern, though less of an issue: I prepped for the SAT a substantial amount when I took it at age 12, did minimal prep when I took it again at age 17. (I did list myself as 10-50 hours prep.)

  110. Anteros says:

    I’m uncertain about how I should answer the question concerning how many children I have. I have 2 step- children (who very much feel part of my family) and I recently discovered that I have 20 biological children from sperm donation 25 years ago. 19 of those I have never even heard from and perhaps who don’t even know that they were donor conceived.
    I don’t want to skew the results such that the category ‘SSC readers who are Englishmen living in the South of France’ has a population with an average number of 22 children. Unless that’s absolutely fine 😀

  111. sty_silver says:

    I love the idea of having prisoner-dilemma type questions. But there is a problem: the prize money comes from you rather than from Omega, and I don’t see any reason to prefer a random SSC reader having money rather than you having money. This made me “defect” vs the random opponent, but I don’t really consider it defecting. I considered cooperating anyway to represent the spirit of the question but then decided not to.

    • svalbardcaretaker says:

      Isn’t it the other way around, average SCC survery takers are more likely to be EA type people and you’d want them to have more money?

    • ShemTealeaf says:

      This was my exact thought process as well.

    • Lambert says:

      I thought that, but then realised that Scott was doing this for a reason and that he probably figures he wants marginal survey responses more than money.

    • googolplexbyte says:

      I just assumed the $1000 is pre-commited and the share not going to the winner is gone.

    • Scott Alexander says:

      Good point, next year I’ll clarify that I’ll burn any of the prize money that I don’t distribute.

      (am I serious? I think no, but I guess that’s for next-year-me to decide)

      • googolplexbyte says:

        You could just save it for the next survey prize pool.

        • Dacyn says:

          No, that’s still an example of Scott using the money. Whatever money he would have used for the next survey could then be used for something else.

          • Michael Watts says:

            Only if “saving it for the next prize pool” means the next prize pool stays the same size and the money that would have gone into it instead goes elsewhere. It could mean the next prize pool gets bigger.

          • Dacyn says:

            @Michael Watts: But then there is more money for whoever wins the prize.

            I guess for people who value Scott having money different from random SSCers having it, this could still be a good idea.

      • Deiseach says:

        I’ll burn any of the prize money that I don’t distribute

        Just like the KLF, eh? Remember to make a bangin’ music video with Tammy Wynette (or the present-day equivalent) first, though! 😀

    • Dacyn says:

      I don’t see any reason to prefer a random SSC reader having money rather than you having money

      Doesn’t that make it more like a true prisoner’s dilemma, since you are supposed to be indifferent to the utility functions of others? The reason why “game theory says to cooperate anyway” (according to some people at least) is that the random SSCer has leverage over you that Scott does not. (Although arguably Scott has counterfactual leverage since he chose how to set up the problem.)

      • Tarpitz says:

        What if you don’t care about game theory but generally favour cooperation on principle, for reasons which are not consequentialist but do prefer Scott keeping the money over giving it to a random?

        • Dacyn says:

          I think in this context “cooperation” could mean one of at least three different things:
          1. The most prosocial action.
          2. The one which is most analogous to the “cooperate” option in the standard Prisoner’s Dilemma.
          3. The one which is labeled “cooperate”.

          So if you want to say you favor cooperation on principle, you should specify which of those you mean.

          In any case, preferring Scott > random SSCer is certainly a consideration in favor of the option marked “Defect”.

    • delta says:

      Strongly agree. I would really like to have answered ‘co-operate’ but I am fairly certain that Scott would better use money than a random blog-reader of his.

  112. Milo Minderbinder says:

    Reading the responses, I accidentally got my OSRI numbers mixed. A value of 123 should be switched from the female set to the male set, and a value of 80 from the male to female. Apologies for the mix-up.

  113. I think I mentioned Coffee as an “I don’t know” effect – I don’t drink coffee and I was genuinely unsure if you wanted the data specifically on coffee (e.g. I don’t know if there are maybe secondary chemicals to coffee that might indeed make one more sleepy).

    In case that’s not what you wanted, know that I do drink other caffeinated drinks (cola), and they make me more awake.

    Hope you get lots of interesting answers! 🙂 I’m looking forward to seeing what you do with the data.

    • Tarpitz says:

      I don’t drink coffee because it has the effect of simulating a rotten hangover – headache, nausea, the works. Even smelling it is mildly unpleasant on this axis. Tea, cola, Red Bull etc. work as intended. Coffee flavoured things – coffee cake, for example – are fine. I have no idea what this is about.

    • imoimo says:

      Cola might make you more awake just due to the sugar. Coffee is a better test of caffeine reaction.

  114. Michael Watts says:

    How often do you experience typical nasal allergy-like symptoms (runny nose, stuffy nose, sore throat, sneezing, coughing, etc) outside the context of a cold?

    All right, you’ve annoyed me.

    I have phlegm constantly collecting in my throat. Have had for three years. Back when this condition was more recent, I didn’t cough, with the result that my airway filled up, I wasn’t able to sleep through the night, I wasn’t able to stay awake at work, and I couldn’t get enough air when walking uphill.

    I now cough all the time. I cough every few waking minutes. I do this volitionally, because I appreciate being able to breathe normally. I don’t have what most people would consider “a cough”, a sensation in the throat that forces me to cough. I’ve become greatly annoyed by medicine’s apparent refusal to distinguish coughing, “the behavior I voluntarily engage in because the alternative is worse”, from coughing, “the reflex I wish I could suppress”.

    So I can only answer “How often do you experience coughing outside of the context of a cold?” with a 5, or more accurately something off the scale like a 20.

    But I can’t provide that answer under the heading “Allergies”. I don’t have an allergy.

    (Obviously, this has had an extremely negative impact on my quality of life. I will take advice. Details: Claritin had no perceptible effect. Flonase had a minor effect. I tested positive for H. Pylori, but eliminating it had no effect. A large battery of allergy blood tests all came back negative. I have frequent and severe heartburn. I have a hiatal hernia which showed up on a barium swallow X-ray. Omeprazole (I take 80mg / day) helps with the pain of heartburn, but not with the accumulation of phlegm in the throat.)

    • Sagar Apte says:

      Some things you can try that might help you (usual disclaimer about taking medical advice from internet strangers):
      Are you able to sleep in a reclining position rather than completely flat? Try it and see if it helps with the symptoms. It should also alleviate the heartburn.
      Time your meals such that you don’t eat anything for 4 to 5 hours before going to bed.
      If you’re taking calcium channel blockers for hypertension, discuss with your physician whether they can be replaced by drugs that don’t relax the lower esophageal sphincter, and the same goes for nitrates.
      You could also try Domperidone, which increases the tone of the lower esophageal sphincter as well as gastrointestinal motility.

    • Scott Alexander says:

      This is fair, and part of what I was doing was trying to hide the actual thing I was looking into. I should change it, but I can’t change variable names midstream, so you will have to suffer, sorry.

    • Simulated Knave says:

      It’s the heartburn. Start taking antacids before you go to bed. Worked for me. I was taking omeprazole in the morning, and it wasn’t helping with the phlegm. Started popping a zantac before bed and it cleared up in days. Zantac got recalled, and am now back to being all phlegmy.

      I need to remember to take my damn omeprazole at night and see if that works, but I keep forgetting.

    • Elementaldex says:

      I had a similar problem for years and have controlled it reasonably well by taking NAC before bed. It is a mucus thinner and makes things much more manageable. Reducing dairy and sugar also seems to help but even complete elimination does not resolve the issue completely. Hope something helps!

  115. Aztonarra says:

    Your security scheme doesn’t work. At least of the answers so far, everyone is asking questions in the public section and giving witty answers in the private section. All of these answers are listed in the “see previous responses” page.

    • Michael Watts says:

      All of these answers are listed in the “see previous responses” page.

      It’s worse than that — the challenges and responses are both being listed in the order they were received, making it trivial to learn the response to any challenge.

    • Rana Dexsin says:

      Wow, yes. That seems like a large oversight.

      Important edit and also @Scott: the “No, please remove my answers before release” option re the public dataset seems to not be applying to the “See previous responses” link, meaning that anyone who wants their text answers to be private should consider holding off for now (moderators, please feel free to strike out the bolded text for accuracy (and mention that you’ve done so) if the problem becomes fixed (or if I’ve drastically misinterpreted the situation)).

      Disclosure on how I determined the above: I submitted two nearly-blank submissions, each with “Yes” to taken surveys before, “No, please remove my answers before release”, Defect/Defect/Defect (it seemed appropriate), and test text in one text field, once using the private keyword field and once using the sibling explanation field (the latter was because the behavior of the private keyword field made it harder to tell what was going on and I had to try again). The latter clearly showed up in the “previous responses” page, which makes me think the whole condition is probably broken. These two rows should probably be manually deleted.

      • Rana Dexsin says:

        Update for anyone reading the above comment: the Google Forms display of previous responses seems to have been disabled per below.

    • sharps030 says:

      Can confirm, this is broken and the answers are visible. I have the first 265 minus 130 public/private pairs cached in an open tab.

      At this point, I don’t see how any of those people can win. You’ve leaked the private answer, so you can’t know that it’s them claiming if they win.

      So unless you’re gonna fix it and then exclude/punish early entrants, I don’t think the prize can be given anymore.

    • HarmlessFrog says:

      This is hilarious. Might as well just use the participants’ google identity to send them an email or something, that’s not public, at least.

      • Radu Floricica says:

        I just put my gmail as password, cause it seemed the thing to do just in case. I guess I was right.

    • Scott Alexander says:

      Normally I wouldn’t care, because I can just ask the winner their age and number of siblings (or something) and only the real winner would know.

      But someone said they put their standard password in there, which seems bad, so I’ve disabled ability to see all responses for everyone (I can’t disable it for specific questions).

      • the verbiage ecstatic says:

        If it makes you feel better, the fact that they use the same “standard password” everywhere, INCLUDING RANDOM GOOGLE FORMS SURVEYS, means they are almost certainly already as completely hacked as it is possible to be.

        • sty_silver says:

          Totally random question assuming you know more about this than I do: what if you use a standard password plus a postfix that you change depending on the website? Totally hypothetical^TM.

          • the verbiage ecstatic says:

            Haha I’m not an expert, but my informed guess is that if your postfix is truly different on each site, and it’s not something totally obvious (like the website name), that’s enough to protect you against anyone just trolling for random identities to hijack. If they’re out to get you personally, that’s another story, but that’s a much rarer situation and unless you’re a particularly interesting person, it probably doesn’t apply to you.

            That said, for someone starting fresh, best advice is to use a password manager like 1 password

  116. Unirt says:

    What’s ‘spiritual’? Is it defined as a) existential or otherwise intellectual interests, or b) a preference for the mysterious to stay mysterious?

    • chaosmage says:

      There is no consensus definition.

      One suggested definition that I like is “actively searching for an appropriate stance towards existence as a whole”. Paraphrasing from Olaf Stapledon’s “Star Maker”.

    • Ragged Clown says:

      I interpreted it to mean somewhere between “I feel special sometimes” and “I have the Holy Spirit coursing through my veins”.

    • Ninety-Three says:

      I generally take “spiritual” to mean “Richard Dawkins-brand atheism is wrong about something”. I don’t think it can be narrowed any further because I’ve seen it used to describe almost every possible disagreement with that view.

      • sty_silver says:

        I don’t think it means that at all. I don’t know anything about Dawkins’ view that I disagree with, I just think meditation is very useful.

  117. ana53294 says:

    The immigration question should specify legal/illegal, as positions may differ between them. For example, increase and simplify legal immigration, drastically reduce illegal immigration.

    • nkurz says:

      Agree, I choose the middle of the scale after averaging my “legal” and “illegal” answers. I think this question is too flawed to use the answers unless there was a followup to distinguish the two cases.

  118. shakeddown says:

    “Country” could really use clarification on whether you mean current living country, country you were born/grew up in, or something else. (even a clarification with grey areas would be helpful).

    • Scott Alexander says:

      The question as currently written says “if multiple possible answers, please choose the one you most identify with”.

      I know I could be more specific, but…every year we go through this. Every question has so many possible edge cases that if I were to do it properly, it would expand into 20 long questions with a bunch of conditionals, and take me an hour to process into machine-readable form. What about dual citizens? What about people who switched citizenships halfway through their lives? What about people who are citizens of one country but live in another? How many months of the year do you have to spend in a foreign country to count as “living” in it? What about digital nomads who don’t stay in any country consistently? Does Taiwan count as a country? Palestine? And then if I want to eg make a list of countries by number of SSC readers, I either have to throw away 90% of the data, or make a bunch of weird choices about how to process it.

      Rather than inflict that on you, I just ask the basic version, ask you to exercise judgment, and every year I choose a few important things to go into in more detail. For example, sibling number started off as two questions (number of older siblings, number of younger siblings), and this year has expanded into a monstrosity with 27 questions divided into Section 19A, Section 19B, and Section 19C, and I just know there’s still going to be someone who’s a genetic chimera or something and complains that there’s no good answer corresponding to his situation.

      I am sorry and I know this is annoying but I think I made the right design decision here.

      • A1987dM says:

        I’d split it into “What country are you a citizen of? (If you have several, pick the one you most identify with)” and “What country are you currently living in?” The former would be useful to interpret “cultural” questions and the latter to estimate e.g. how many people might be interested in attending a meetup somewhere. The number of people who answer the two differently would also be interesting info.

      • Aapje says:

        This makes me wonder whether the number of edge cases is increasing, which I would expect due to globalization and increased individualism. If so, surveys may gradually be getting less accurate, assuming that survey creators are not increasing their granularity to match the increased granularity of survey takers (and even if they would, this would just lead to problems due to low answer counts for each option).

      • Roebuck says:

        I think one question is fine, but could be just tweaked so that it doesn’t say “your country” but “the country you live in” or “the country you are from”. The latter two are very common questions and I (and possibly most people who have moved countries) have no problem answering them, but “your country” was really confusing.

      • kronopath says:

        I still maintain that you are missing important differences by not at least differentiating between “the country you spent most of your life in” (which should mostly account for cultural background) and “the country you currently live in” (which should account for current political views, life situation, and the like).

        Every single immigrant is currently misrepresented in your survey. Since you apparently care about political views around immigration, this is a problem: you don’t even have a way of figuring out who is and is not an immigrant themselves.

        You don’t necessarily have to explode it into a huge twenty-question questionnaire the way you did for birth order, but splitting the one question into two, or at least clarifying the one question you currently have, will make it a lot clearer as to what it actually is you’re looking for.

  119. Said Achmiz says:

    Once you click a box on a Google form, you cannot un-click it – i.e. you can change your answer but you can’t unanswer the question. If you click a box you didn’t mean to, please switch your answer to “Other” if available; if not, then choose the most boring inoffensive answer that is least likely to produce surprising results. I realize how bad this is but there is apparently no way around it.

    Add a “Whoops, I didn’t actually mean to answer this question” option to every question. Problem solved?