It's that time of year again. Actually, a little earlier than that time of year, but I'm pushing it ahead a little to match when Ozy and I expect to have more free time to process the results.
The first draft of the 2014 Less Wrong Census/Survey is complete (see 2013 results here) .
You can see the survey below if you promise not to try to take the survey because it's not done yet and this is just an example!
2014 Less Wrong Census/Survey Draft
I want two things from you.
First, please critique this draft (it's much the same as last year's). Tell me if any questions are unclear, misleading, offensive, confusing, or stupid. Tell me if the survey is so unbearably long that you would never possibly take it. Tell me if anything needs to be rephrased.
Second, I am willing to include any question you want in the Super Extra Bonus Questions section, as long as it is not offensive, super-long-and-involved, or really dumb. Please post any questions you want there. Please be specific - not "Ask something about taxes" but give the exact question you want me to ask as well as all answer choices.
Try not to add more than a few questions per person, unless you're sure yours are really interesting. Please also don't add any questions that aren't very easily sort-able by a computer program like SPSS unless you can commit to sorting the answers yourself.
I will probably post the survey to Main and officially open it for responses sometime early next week.
- I suggest a basilisk question, as usual.
- I'd also like OCEAN/Big Five restored. The results from last time weren't very definitive and more data would be helpful.
- Given the recent SSC debate on how effective MIRI is, it might be interesting to ask something about opinions of MIRI's effectiveness in pursuing its mission: just asking how much people donated is insufficient since one might donate faute de mieux, and IIRC the modal donation is $0 in the first place.
- Perhaps the AI projection questions could be replaced by the same question-set used in "Future progress in artificial intelligence: A poll among experts", VC Müller, N Bostrom? Buys one comparability with their survey responses; it would be very interesting and intriguing if LWers turned out to be less extreme.
The "currently looking for more relationship partners" feels to me ambiguous -- what if someone is open to having more relationship partners, but not actually looking for more? Perhaps make this a three option issue "actively looking" vs "open to having more relationship partners but not actively looking for more" vs "not open to having more relationship partners"?
I'd add: "Do you use/have you used a Spaced Repetition program, like Anki or Supermemo?"
- No
- Yes, regularly
- I tried it, but stopped
In the politics I'd add a "how strongly do you care about this political affiliation" (scaled on 1-5, from "not at all" to "it's an important part of my identity")
In the politics I'd add a "how strongly do you care about this political affiliation" (scaled on 1-5, from "not at all" to "it's an important part of my identity")
A slightly longer option is to ask people to rate every political affiliation from 1 to 5. This lets us identify clusters better, as well as total political identification (if someone gives 1s to everything, that's the 'apolitical' option).
A couple (more) questions I'd find interesting:
How many times do you exercise per week, on average?
How many nonfiction books do you read per month, on average?
How knowledgeable would you consider yourself in the following fields? (on a scale of 0 to 5, where 3 is about "studied it at university" and 5 "I'm a publically recognized expert")
Psychology
Economics
Artificial Intelligence
Statistics
The goal is to give reference points for the jumerical scale so the numbers can be more meaningfully compared, not necessarily to have a detailed reference that will annoy some people (too much to read) and confuse others (don't fit in).
I was thinking of something intermediate, like "took specialized classes, not 101", but I'm afraid being too specific about education means people will not take into account other things like how much they forgot or how much they learnt in the meantime. Someone who took econ 101 but works in a bank and argues about economic policy on the internet everyday and read a few good books is probably more knowledgeale than someone who majored in economics and then said "screw it all" and became an actor.
I would personally be interested in more detailed drug use questions. Examplae. In the Past year have you taken:
-Modafinil
-Amphetamine (of any sort including Adderall)
-Heroine/Other Opiates (not prescribed by a doctor as a painkiller) -Marijuanna
-Hallucinogens (LSD/DMT/Psychobillin)
-Testosterone/HGH (if you took testosterone because you are transexual do not click yes)
Maybe there is a better list of drugs?
it was bad to make people admit to illegal activities
You could use randomized response method for this question or any other controversial question (for the type of questions that are technically feasible, of course).
I'd been interested in seeing how many people participate in systematic self-improvement efforts outside the LW/CFAR space. Some examples in rough increasing order of weirdness: diet, fitness, psychiastric help, life coaching etc., social skills programs of various kinds, nootropics. This covers a lot of ground and there's probably some options I haven't thought of, but these two questions would be a good start:
- Do you follow a limited or unusual diet for health (not ethical) reasons?
(this would probably need "I tried and it didn't work" and "I tried but found it too hard" options. We'd probably also want some way of distinguishing between "I'm allergic to onions" and "I kind of try not to eat red meat" and "I'm full-blown hardcore paleo".)
- How many hours a week do you spend exercising, on average?
For relationship status, a polyamorous person can be married and in a relationship at the same time, which is a problem. Similarly, someone can be living with their partner/spouse and additional roommates. Also, "Liberal" in the Political section should probably be renamed to "Progressive", to avoid collisions with how "liberal" is used in Europe and in political philosophy.
Please change the "referred by a link on another blog" option to "referred by a link on another blog or website". It's been bugging me for years.
Again, your relationship style question conflates very different clusters. There is a big difference between traditional monogamy and serial monogamy, or between the ideal of polyamory preached in rationalist circles and the harem type of polyamory (polygyny, mistresses, etc...) which is traditional for a sufficiently high-status man to have. The obvious way to solve this is to split the "prefer monogamous" and "prefer polyamorous" answers into two answers each, each of which describes the relevant clusters in a short sentence similar to the answer choices in the political question.
You are asking too many calibration questions for me. I might answer 1, 2, or, at a stretch, 3. 10 questions is gonna cause me to leave that whole section blank.
Likewise, I don't think I'm gonna answer the probabilities section unless you switch to radio buttons, like you use in the extra credit political section.
Per your own advice, you may want to add a completely implausible question to the survey so you can measure our Lizardman's constant and subtract it from the other results.
I'd suggest a section asking opinions on famous thought experiments. "Do you flip the switch in the trolley problem?", "do you push the fat man in the modified Trolley problem?", "is the China brain conscious?", "do you one box against Omega?", "does the Star Trek transporter kill you?", etc... maybe some not so famous though experiments, too. Eliezer has that thing in which a primitive culture which is afraid that cameras will steal your soul but does not realize that the eye is basically a camera is used as an analogy for crbcyr jub ner nsenvq bs hcybnqvat, and you asked that question about the angel which offers you reincarnation as a metaphor for pelbcerfreingvba.
My bonus question is "Do you read non-rationalist fanfiction? That is to say, do you read fanfiction other than HPMOR, Luminosity, Friendship is Optimal, etc... and their derivative fanfanfiction?"
(for example, 0 if you are single, 1 if you are in a monogamous relationship, higher numbers for polyamorous relationships)
This implies that you are only poly if you're in more than 1 relationship. I suggest simply
(for example, 0 if you are single, 1 if you are in a monogamous relationship)
because
- If you're poly, the idea of counting relationships will be familiar to you already; it's only those who aren't who might need it set out explicitly
- I hate saying "you have to add more text here to cater to my itch" - if I can fix what I perceive to be a problem by removing text, I much prefer that.
For Super Extra Bonus Questions: (feel free to modify the answer choices)
With which of these metaethical positions do you most identify?
- Non-cognitivism: Moral statements don't express propositions and can neither be true nor false. "Murder is wrong" means something like "Boo murder!".
- Error theory: Moral statements have a truth-value, but attempt to describe features of the world that don't exist. "Murder is wrong" and "Murder is right" are both false statements because moral rightness and wrongness aren't features that exist.
- Subjectivism: Some moral statements are true, but not universally, and the truth of a moral statement is determined by non-universal opinions or prescriptions, and there is no non-attitudinal determinant of rightness and wrongness. "Murder is wrong" means something like "My culture has judged murder to be wrong" or "I've judged murder to be wrong".
- Substantive realism: Some moral statements are true, and the truth of a moral statement is determined by mind-independent moral properties. "Murder is wrong" means that murder has an objective mind-independent property of wrongness that we discover by empirical investigation, intuition, or some other method.
- Constructivism: Some moral statements are true, and the truth of a moral statement is determined by whether an agent would accept it if they were undergoing a process of rational deliberation. "Murder is wrong" can mean something like "Societal agreement to the rule 'do not murder' is instrumentally rational".
With which ethical position do you most closely identify?
- Utilitarianism
- Egoism
- Contractualism
- Contractarianism
- Other Consequentialism
- Kantianism
- Divine Command
- Other Deontology
- Natural Law
- Aristotelian Virtue Ethics
- Stoic Virtue Ethics
- Epicurean Virtue Ethics
- Other Virtue Ethics
- Intuitionism
With which of these broad political groupings do you most closely identify?
- Progressivism (Includes American progressives, European social democrats, socialists, communists, left-wing anarchists, the social justice movement, etc. Important values include economic and social equality, liberation of oppressed groups, and democracy.)
- Liberalism (Includes European liberals, libertarians, anarcho-capitalists, etc. Important values include freedom of association, individual autonomy, and technological progress.)
- Conservatism (Includes American conservatives, Christian democrats, nationalists, neoreactionaries, etc. Important values include tradition, bonds within communities, and patriotism.)
- Neither of "...and currently looking for more relationship partners" or "...and not currently looking for more relationship partners" seem to capture what I feel is a common poly sentiment, "...and I'll figure out what I want with new people as I get to know them".
- Neither of "Atheist but spiritual" and "Atheist but not spiritual" seem to capture "Atheist who doesn't know what 'spiritual' means in this context and therefore doesn't want to come down either for or against it, because there are both interpretations where it seems nice and interpretations where it seems silly".
- P(Aliens), P(Many Words), etc. -- are these supposed to be conditioned on the fact that we're not in a simulation, etc.? Otherwise, for example, P(Aliens) goes down in proportion to P(Simulation).
- Given that you already ask the Sex question, the Gender question can have just three answers “Man”, “Woman”, and “Neither/Other”.
- As suggested by people on SSC, you might want to add a question “Is your gender an important part of your identity?” or something, to find out how many of us are cis-by-default.
- I'd have three answers to Relationship Goals, “actively looking”, “open but not actively looking”, and “not open”.
- Split the “no” answer to Children into “not now” and “never”.
- Selfishly, I'm glad you didn't split physics in the Profession question into theoretical physics and experimental physics so that I, as a phenomenologist, don't have to choose, but you might still want to consider splitting them.
- Maybe in the Religious Views question “Agnostic” → “Agnostic/Apatheist/Ignostic” (cf. “Don't know/Don't care/Don't understand”). Also, consider specifying what the hell you mean by “spiritual”, as IIRC the previous times different people interpreted it in different ways.
- In the Aliens question, does octopus-level intelligence count? Does pre-50,000 BC human-level intelligence count?
- Wouldn't (say) three calibration questions be enough?
Thanks so much for running this again
Comments
-Items should either be randomized or if sorted, they should be sorted based on last years prevelency. For example in politics libertarian is the top listed item despite not being alphabetically first or being a majorty / plurality view.
-The religious denomination shouldn't ask atheists to skip the question. One is about what you believe the other about what you do. Plenty of atheists fast on Yom Kippur or go to church on Christmas.
-Less wrong use should have two items on comments that differentiate by frequency. Eg commented on a thread and comment at least monthly.
-Instructions for percentages should be more clear. Maybe give an example. As worded I'm unsure if to write"50%" or just "50"
-Not sure if cryonics should be conditional on no catastropic event.
-While we are asking about all the psych issues any reason not to ask about anti social or ADHD or other ones as well.
-The taxes question isn't meaningful. Politicaly the divide isn't lower higher in general. It's typically more / less progressive. Suggestion; "Should taxes on people earning more than 1mm per year be lower / higher"
-Older and younger siblings should be next to each other and not separated by birth month
-More details on vegetarianism. Rather than yes / no it should be a spectrum. Maybe: "vegan / vegetarian / reduced meat (eg meatless Monday) / no vegitarian leaning"