Comment Permalink

Ishaan10y00

I suppose I'd be willing to say that those are "facets" of a single property if presented with evidence that they correlated really rightly with each other, but, no, I wouldn't a-priori consider, say, sensation seeking and intellectual open mindedness to be different facets of the same quality.

They might have superficial behavioral similarities. To use Openness as the example, both sensation seekers and the intellectually open minded might go on a roller coaster - one for the sheer thrill and the other out of curiosity as to how it feels- but I wouldn't predict those two qualities to be particularly more correlated than say, Extroversion and Openness.

Or, to use Agreeableness as the example, I wouldn't expect conformists to be kinder -, even if the two are superficially similar in that they are less likely to get into fights. Conformity and kindness aren't two facets of the same diamond, they're just two mostly separate variables which influence how likely you are to object to things.

My criticism is that the Big Five lumps together qualities that aren't even particularly similar into one variable.

(Although, I suppose it is fair to say I'd expect high inter-correlation between industriousness, dependability, and extrinsic motivatability, so I guess you can chalk those up as facets of a single thing)

ChristianKl10y00

I suppose I'd be willing to say that those are "facets" of a single property if presented with evidence that they correlated really rightly with each other, but, no, I wouldn't a-priori consider, say, sensation seeking and intellectual open mindedness to be different facets of the same quality.

The big five aren't a-priori categories. They are supposed to be factors that come out of principle component analysis.

4gwern10y

I think we may have a confusion of terminology here. The Big Five is the result of a statistical procedure which boils down thousands of peoples' responses to thousands of questions (based on personality words) into a set of 5 statistical abstractions which are as non-redundant and predict as many responses on average as possible; why 5? Well, 5 abstractions were chosen because if you boil it down further to 4/3/2/1, what's left doesn't seem particularly meaningful (there may be a 1 'J' healthy-personality factor akin to intelligence's G factor, but last I saw it mentioned it was still highly doubtful), and more than 5 would be harder to use since 5 is a nice easily-handled number which seems to map pretty well onto gross aspects of personalities & can be easily measured in a short questionnaire. However, there is no reason you could not have a factorization instead as a Less-Big 6 (something like HEXACO) or a Even-Less-Big 7 or some larger number. And specifically, each of the Big 5 are themselves the correlation of 6 more specific factors or 'facets' for a total of 30 factors available (assuming you used the longer Big 5 tests which allow reasonably accurate per-facet measurements instead of lumping them together and only giving you say Openness); for example, Openness is extracted from the facets/factors dubbed "Fantasy", "Aesthetics", "Feelings", "Actions", "Ideas", & "Values". You might not consider the "Ideas" facet (whatever that is) to have anything to do with the "Fantasy" facet, but apparently they do correlate in the real-world samples of people taking the long more in-depth personality surveys which were used to construct the Big 5. Often all you need are the top-level factors and so the details of how many factors doesn't matter, but sometimes it is important to go down to the more fine-grained factors and make use of the 30 facets. For example, males/females do not show up as very different on Big 5 inventories, despite the universal belief that men

See in context

20 [POLL] LessWrong group on YourMorals.org (2015)

by gwern

3rd Mar 2015

1 min read

20

In 2011, InquilineKea posted a Discussion topic on YourMorals.org, a psychology research website which provides scores of psychology scales/inventories/surveys/tests to the general public to gather large samples. Niftily, YourMorals lets users sign up for particular groups, and then when you take tests, you can see your own results alongside group averages of liberals/conservatives/libertarians & $GROUP. A lot of time has passed and I think most LWers don't know about it, so I'm reposting so people can use it.

The regular research has had interesting results like showing a distinct pattern of cognitive traits and values associated with libertarian politics, but there's no reason one can't use it for investigating LWers in more detail; for example, going through the results, "we can see that many of us consider purity/respect to be far less morally significant than most", and we collectively seem to have Conscientiousness issues. (I also drew on it recently for a gay marriage comment.) If there were more data, it might be interesting to look at the results and see where LWers diverge the most from libertarians (the mainstream group we seem most psychologically similar to), but unfortunately for a lot of the tests, there's too little to bother with (LW n<10). Maybe more people could take it.

You can sign up using http://www.yourmorals.org/setgraphgroup.php?grp=623d5410f705f6a1f92c83565a3cfffc

All quizzes: http://www.yourmorals.org/all_morality_values_quizzes.php

Big 5: http://www.yourmorals.org/bigfive_process.php

(You can see some of my results at http://www.gwern.net/Links#profile )

Personal Blog

20

New Comment

33 comments, sorted by

top scoring

Click to highlight new comments since: Today at 7:02 AM

[-]William_Quixote10y60

If you check the survey results you will find that the large pluralality of less wrong types range from socialist / progressive / center left rather than libertarian.

[-]gwern10y120

Maybe that's what they self-identify as, but the non-explicitly-political results seem to usually more similar to libertarians.

[-]Manfred10y120

The best explanation, I think, is that there's some subgroup of the population with certain psychological traits, and only a fraction of these people are libertarians, but most libertarians are these sorts of people.

[-]Transfuturist10y10

Libertarian socialism FTW.

[-]BenLowell10y50

Are there any known groups which have high conscientiousness? I would be especially curious to know about groups with high conscientiousness and openness to experience.

[-]Salemicus10y30

High conscientiousness is strongly associated with political conservatism, although there is dispute as to whether this is a direct effect of personality on ideology or a societally mediated effect (i.e. whether it is conscientiousness -> conservatism or conscientiousness -> life success -> conservatism).

(On the first point see e.g. Hirsh, J.B., DeYoung, C.G., Xu, X., & Peterson, J.B. (2010) - Compassionate liberals and polite conservatives. Personality and Social Psychology Bulletin, 655-664. On the second point see e.g. Morton, R., Tyran, J. R., & Wengström, E. (2011). Income and ideology: How personality traits, cognitive abilities, and education shape political attitudes. Univ. of Copenhagen Dept. of Economics Discussion Paper, (11-08).

However, conservatism is negatively associated with openness to experience. I am not aware of any notable political cluster that combines the two.

[-]seer10y00

I am not aware of any notable political cluster that combines the two.

I suspect there is a reason for that. Namely maintaining high conscientiousness on a group level requires a common set of social assumptions and punishing defectors. Both of which are hard to do if you constantly have new people (especially from different cultures) coming to your society. I suspect the only groups that successfully combine openness and conscientiousness are living in isolated locations where natural barriers limit the amount of outsiders without the people having to "get their hands dirty" by lowering openness.

[-]bramflakes10y00

Extreme case: the Amish

[-]kavec10y00

I scored high on both (4.5 openness and 4.2 conscientiousness), so, err... good question!

(AKA I really, really wish I had an answer for this.)

In the few minutes I had before posting this, I managed to scare up this 2003 meta-analysis that tried to correlate the big-five with Holland's RIASEC occupational types. Before reading I was expecting to see (85%) a correlation between conscientiousness and artistic/openness and artistic, with the prior that I attending an art school and successful art seems to hang on both qualities.

http://www.ibrarian.net/navon/paper/PERSONNEL_PSYCHOLOGY_2003_56.pdf?paperid=20534921

Spoilers: I was wrong. The analysis concludes that openness correlates with Artistic and Investigative types (Table 5) and, of those, conscientiousness correlates very weakly to Investigative (Table 3).

Heading to Wikipedia, you can find sample careers for RIASEC codes below, but I'm not getting a sense that many of these careers really combo well with conscientiousness and openness.

http://en.wikipedia.org/wiki/Holland_Codes

Noted exceptions: Technical writer, "Counselor", Psychology/Psychologist

[-]Ishaan10y00

Of interest: https://www.psychologytoday.com/blog/beautiful-minds/201108/is-conscientiousness-compatible-creativity

(I can't say I'm really a fan of the big 5 questionnaire in general, though. Agreeableness mixes together kindness and conformity. Conscientiousness mixes together industriousness, dependability and extrinsic motivatability. Openness confounds sensation seeking and open-mindedness. I know it's the most respected assessment, but I think it's really messy.)

[-]gwern10y00

Agreeableness mixes together kindness and conformity. Conscientiousness mixes together industriousness, dependability and extrinsic motivatability. Openness confounds sensation seeking and open-mindedness.

Don't those map onto facets, though? You can hardly blame the Big 5 there when you've named no less than 7 possible factors.

[-]Ishaan10y00

My criticism is that the Big Five lumps together qualities that aren't even particularly similar into one variable.

[-]gwern10y40

I suppose I'd be willing to say that those are "facets" of a single property if presented with evidence that they correlated really rightly with each other, but, no, I wouldn't a-priori consider, say, sensation seeking and intellectual open mindedness to be different facets of the same quality.

However, there is no reason you could not have a factorization instead as a Less-Big 6 (something like HEXACO) or a Even-Less-Big 7 or some larger number. And specifically, each of the Big 5 are themselves the correlation of 6 more specific factors or 'facets' for a total of 30 factors available (assuming you used the longer Big 5 tests which allow reasonably accurate per-facet measurements instead of lumping them together and only giving you say Openness); for example, Openness is extracted from the facets/factors dubbed "Fantasy", "Aesthetics", "Feelings", "Actions", "Ideas", & "Values". You might not consider the "Ideas" facet (whatever that is) to have anything to do with the "Fantasy" facet, but apparently they do correlate in the real-world samples of people taking the long more in-depth personality surveys which were used to construct the Big 5.

Often all you need are the top-level factors and so the details of how many factors doesn't matter, but sometimes it is important to go down to the more fine-grained factors and make use of the 30 facets. For example, males/females do not show up as very different on Big 5 inventories, despite the universal belief that men & women have very different personalities; but this is because the male/female difference is on the facets and that gets hidden when it's boiled down to the 5: "The Distance Between Mars and Venus: Measuring Global Sex Differences in Personality".

And the facets themselves can be factorized even more narrowly if one wants. (The most extreme example I've seen was a paper I can't seem to refind at the moment, but where they had hundreds of people fill out thousands of test items online and with this huge bulk of data, they extracted what they called the Tiny 100 or the Little 100 or something like that.)

but I wouldn't predict those two qualities to be particularly more correlated than say, Extroversion and Openness.

I think your prediction would just be wrong. Interest in new ideas does seem to correlate with sensation-seeking (I'd guess those are the Openness facets "Ideas" and "Action", respectively) and that's why they help give rise to the Openness factor.

Conformity and kindness aren't two facets of the same diamond

I'd agree on that one. Conformity I would expect to fall under Conscientiousness ("Order"?) and kindness is definitely under Agreeableness ("Altruism"). Hence conformism and kindness are going to be mostly independent traits.

[-]Ishaan10y00

So as I said before, you changed my mind but - now that I've had some time to mull it over, I'll articulate why instinctively I felt something wrong about the big-five...because everything you say is true, but I still have a nagging feeling of wrongness. I've put the more important parts in >quotation format, since these are unverbalized thoughts still in formation and tend to over-ramble and I don't have time to properly pare it down at the moment.

Back when I first looked at the questions to formulate the Big 5 (my memory indicates a much longer questionnaire than what see when I google it now) I was bothered by questions such as "I tend to tell people when I disagree with them" and "I like to make people comfortable" both being held under the same variable (agreeableness).

It seemed to me like there were many underlying factors at play in how someone might answer such a question: "aversion to causing negative emotions in others" (niceness) and "aversion to drawing anger towards oneself" (conformity) competing with "aversion to white lies", such that an honest-to-a-fault person would be behaviorally similar to a jerk, such that kindness is behaviorally similar to weakness. I had a similar pattern of objection for conscientiousness (sense of responsibility, preference for order, attention to detail, willpower), extroversion (Preference for companionship, social skills, lack of social anxiety, general energy level).

So then, I think the "general form" of my objection is:

"Factors X and Y are unrelated, but X and Y both contribute to factor P. You're creating a Construct C, measuring X, Y, and P separately, and tossing them all in the C bucket, and that annoys me."

Why did it annoy me? Well, it's mostly because in some cases, we called Construct C by a name that I felt aught to be reserved for Factor X. For example, I intuitively feel "Extrovert" aught to be reserved for "the degree to which one prefers comfortable social situations to comfortable solo situations" and should not be entangled with social skills.

I think this "tangling" of variables which intuitively bugs me is why all the factors correlate (except for neuroticism, which inversely correlates). To give one example, someone who is simultaneously extroverted (as defined by me, above) and neurotic will have depressed extroversion scores because they're nervous and unskilled in social situations. They're not introverted, they're just isolated, and the Big 5 can't tell the difference. The human reader will go on to conclude that introverts are inherently more neurotic. (I can give several other examples in this format for the other constructs, if you don't agree with this particular one).

Here's an example of this happening "in the wild": Here's someone talking about the dark side of creativity, and they reference this study showing that creatives are lower on the honest-humility factor on the HEXACO.

Now, what do you think a highly creative, original thinker would answer on a HEXACO question like "People sometimes tell me that I am too critical of others"? Do they voice criticism because they are arrogant meanies who don't care about hurting others, or is it because they're nonconformists who value honesty over the warmth of mutual agreement and are simply treating others as they want to be treated themselves? Obviously the latter, right? But in the data, that's going to come out as "Scoring low on Honesty-Humility". (This sort of thing is why I suggested that Agreeableness ends up collapsing both kindness and conformity).

Anyway, my previous dismissal of the Big 5 was wrong. It's very good at what it claims to do, which is separate out factors which lead to certain patterns of responses. From the algorithms point of view, it doesn't matter if X and Y are uncorrelated, they still cluster together because of the relationship to P and fold into Construct C. And it is useful to separate out broad patterns of responses into broad constructs even if those constructs contain totally uncorrelated factors, because there is no way you could separate out all the uncorrelated factors anyhow.

Still, that's why it bugs me. I feel intuitively more satisfied when I look at tests which attempt to myopically measure uncorrelated factors individually and totally ignore the other factors, even though those tests probably won't be as good at catching broad swathes of behavior like the Big 5 can. People do actually feel like creative smart folks are arrogant and narcissistic meanies because of the behavioral overlap (which is why Lesswrong-ish types are often accused of arrogance and write posts delineating proper humility from "false humility"), and the Big 5 does capture things like that. That instinctively bugs me because it's conflating things in the same way human beings do, but on the other hand it's conflating things in the same pattern that humans conflate things, which is actually a pretty big achievement for an algorithm now that I've spent more time thinking about it.

Considering all of the above, I think it would stop bugging me if we had more precise labeling for the names of the five factors. The trouble I have is that it's just too easy to conflate here, the researchers are human to begin with and already prone to conflation, and then when we imprecisely label the Big 5 factors we only facilitate that conflation. (And maybe the original researchers had this danger in mind, when they labeled it Agreeableness instead of "Niceness", hoping that people would realize that compliance is a factor as much as altruism, but...)

Does that make sense / still sound wrong to you? (If you get to it, sorry it's not more concise.)

[-]Ishaan10y00

I have now changed my mind about the Big Five. Thanks!

[-]ChristianKl10y00

I suppose I'd be willing to say that those are "facets" of a single property if presented with evidence that they correlated really rightly with each other, but, no, I wouldn't a-priori consider, say, sensation seeking and intellectual open mindedness to be different facets of the same quality.

The big five aren't a-priori categories. They are supposed to be factors that come out of principle component analysis.

[-]Adele_L10y-20

I would be especially curious to know about groups with high conscientiousness and openness to experience.

Possibly startup-culture people?

[-]TobyBartels10y00

Well, I'm more agreeable than average, while LessWrongers are less agreeable; while LWers are about as neurotic as most people, I am less. OTOH, LWers are less extraverted and conscientious than average, and I am if anything more extreme. (We're all average at openness.)

[-][anonymous]10y-20

Interesting to compare this with the Lesswrong annual survey

[This comment is no longer endorsed by its author]Reply

[-][anonymous]10y-20

I really think it should measure masculine / feminine thinking, however I also think the way some researchers define that is weird. For example sometimes m. is seen as "autistic", emotionless thinking (why? how is a Rambo-rage not an emotion?), in some other research e.g. Hofstede, it is "live to work", work-oriented attitudes... frankly this is IMHO not what the term means. I would measure masculine thinking like:

Do you think fraud is sometimes worse than force?
Do you think it can be right to avenge verbal insults by force?
Do you think not responding to force with force is dishonorable?
Do you think breaking your word is dishonorable even if it leads to the best or least bad result for all?

I think this would be very interesting! See how it overlaps (or not) with conservatism, loyalty, authority, purity etc.

[-]RowanE10y00

I think at least the first one of those is okay, but caring about honour seems to be culture-specific and tradition-specific enough to be a separate thing from masculinity, and asking whether something is "dishonourable" in place of whether it's bad is going to get false positives from anyone who thinks "well, I'm not sure if honour is even a real thing, but as I understand it, refusing to break your word even when it's obviously a bad idea is the kind of thing "honourable" people do".

[-][anonymous]10y40

How easy is it to test for things without being culture-specific and tradition-specific?

Haidt's authority foundation, for example: how would you test for that? You could ask people to identify authorities in their lives, but maybe they don't make the connection between the word and the concept -- how many people who believe the NYT, the "scientific consensus", etc. on faith would recognize that they're doing so? You can't test that by asking about the military; the people who think the NYT on authority are probably less likely than the people who don't to think that the military is a legitimate authority whose commands can be trusted.

And what about the purity foundation? Haidt has written a few times about liberal purity, but it hasn't been incorporated into the test yet. It's not hard to see that it exists: concerns about the purity of food, disgust-evoking metaphors for illiberal positions and the people who hold them, and so on. Inorganic food is full of toxins and nationalism is a revolting disease. But how would you test for liberal and conservative purity at the same time? You'd have to go meta enough to capture both (as well as the concepts of purity that exist in every other cultures) in such a way that the people taking the test would not only know what you're talking about but also make the connection between the concepts you're asking about and the things they're doing.

Given the memetic success of Haidt's test (a college professor I had referenced it in a lecture a few times, and said he'd heard about it from NPR), it doesn't appear obviously wrong to a lot of people. This could be because they legitimately haven't seen evidence of liberal concern for the other foundations (which would imply social distance from at least certain strands of what Haidt calls liberalism), but it could also be because they don't make the connections between that evidence and the concepts.

It's not hard to miss connections. I've ordered fast food innumerable times in my life, so I've heard phrases along the lines of "the meal, or just the sandwich?" after ordering a burger. But I still didn't think burgers counted as sandwiches until someone pointed it out a year or so ago. (It's probably relevant that I'd never seen a sandwich on a roll [circular, bunlike] until this year -- burgers are nowhere near my image of the prototypical sandwich.) And this is a case where there's no major political movement with an interest in making people miss those connections! Part of the memetic success of the five-foundations theory is probably that Haidt provided academic evidence for what liberals already liked to think about themselves.

It may be possible to find a way to test for these things culture-neutrally, but it's much harder a problem than you think. The question of masculinity would probably also need separate tests tailored to separate cultures -- or you could have a test measuring the extent to which the test-taker follows some number of different forms of masculinity. Some cultures are honor cultures, and some cultures aren't, but in honor cultures, honor tends to be in the male domain.

[-][anonymous]10y00

More general: fairness and caring are pretty universal, but purity, loyalty and authority are not only culture-specific, also political tribe-specific, and there are huge mistake potentials here, what is authority for one is a proper expert for another, what is purity for one is understandable revulsion over an immoral act for the other, what one sees as disloyalty can be loyalty to a non-standard group and so on. In fact, my prior would be that loyaly, authority and purity will not predict major political tribes at all when the questions are truly properly set. The lack of them will predict a small number of really smart people. Then there is a larger bunch of people who imitate that small number, follow them as authority, loyal to their causes and feel revulsion when their ideas are dragged in the mud, but still use the non-authoritarian, non-loyalist, non-purist language of their leaders. (This tribe would be called "liberal" in the American terminology. In many Easter European cultures too, in Western Europe just called "normal".) The vast majority of that tribe majority will have loyalist, authoritarian, and purist instincts, just not towards the common targets, and wrapped into a language that denies it. Loyalty to the group that identifies as disloyal individualists. Using Dawkins quotes as authoritarian discussion-stoppers, yet many of those quotes will contain funny, irreverent, anti-authoritarian bits. Conforming to non-conformism. Rebellion as a mass fashion item. These are not new ideas.

[-][anonymous]10y20

The fairness foundation isn't universal. I know people who test low for it, but that may just be a testing artifact. "Whether or not some people were treated differently than others" -- people are different, so of course there will be circumstances where it's right to treat them differently. There are some cultures where it's probably legitimately absent.

A: "That's not fair!"

B: "Life isn't fair."

Also, I don't think liberal language hides the purity intuition.

[-]seer10y00

From my, admittedly limited, knowledge of non-Western cultures, I get the impression that the fairness norm is very much a Western Civilization thing.

[-][anonymous]10y00

My bad. I wanted to say universalist.

[-]hairyfigment10y00

purity, loyalty and authority are not only culture-specific, also political tribe-specific, and there are huge mistake potentials here, what is authority for one is a proper expert for another, what is purity for one is understandable revulsion over an immoral act for the other, what one sees as disloyalty can be loyalty to a non-standard group and so on.

Note that you just made an argument for these being terrible bases for public and social policy in a diverse society.

[-][anonymous]10y00

Yes. And you can go two ways from there. Either you can try to eliminate them, but then you meet the issue that for most folks it is incredibly difficult to make a difference between policy and values as both are approached from an "I cheer for X" angle. If succesful, you end up with a society that does not have a culture. From this viewpoint I feel for the conservative case. It is very, very weird to try to build a society without culture, it is as if it was not meant for human brains. And it would be very good if people could see a clear difference between policy and culture but again it is very, very hard, it goes against many instincts.

I am basically a product of that. My parents were always the kinds of secular Euroliberals who don't really believe in many values, and for this reason I alway found it hard to find goals in life: there was nothing they were passionate or judgemental about, so I find it to be passionate about anything. From this kind of upbringing it is just hard to think anything matters as everything was taught as a mere preference, hobby, interest...

The opposite solution is to try to break up into an archipelago, where sub-societies, sub-cultures are forming their own rules. This sounds a bit sci-fi but look at it this way. The Roman Empire was largely monolithical. The Dark Ages afterwards an archipelago. Charlemagne's empire monolithical, then the German-Italian city-states and small kingdoms again an archipelago. It sounds a bit like it ebbs and flows.

[-]seer10y-20

Both the Roman and especially Charlemagne's empires were archipelagos compared to today's states. Both contained many sub-states that where mostly left to govern themselves as long as they acknowledged imperial authority and paid taxes.

[-][anonymous]10y20

Look, if YourMorals wanted to be culture-neutral, the first step would be purging terms like "liberal" and "conservative". (I know countries where "conservatives" are far more anti-capitalistic than "liberals" because their capitalism is largely controlled by foreign firms and they dislike that on a nationalism basis: better our government than foreigner's businesses.)

YourMorals is strongly US-centric, and the rest of the world like myself has to resort to rule-of-thumb heuristics to modify both input given and output received...

My input modifier heuristic is "how would those American movie actors I can identify with most answer". I expect others around the world doing something similar. For example, my list of questions is based on "what would Clint Eastwood and people around the world who easily identify with him reply".

[-]RowanE10y20

I'm not saying it should be culture-neutral, I'm assuming that if "masculine thinking" is a real thing, then it's going to be culture-neutral, and we want to, as much as possible, separate "do you think in a masculine way" from "do you think in the way this culture prescribes for men". Otherwise, it'll just be a poll of gender identity with a lot of noise added.

[-][anonymous]10y30

This is a good point, but is it true for the other metrics? It can get confusing when an atheist who does not happen to personally studied evolution much but still believes in it because the scientists say the evidence is good and he trust they don't lie, so he is taking it more on an authority basis, and a creationist, who is doing the completely opposite based on the basis of religious authority, so they may both think the other person is anti-authoritarian :) Because obviously the only the authorities I trust are real authorities. To construct a test for every possible authority people may trust, well, does not sound easy.

The other option is of course construct very abstract and generic tests which may have their own failure modes. Not everybody will understand questions like "if an authority figure tells you to..." - some people like something more specific, and that is culture-dependent.

[-]RowanE10y10

Well, generally I think "abstract and generic" is the obvious way to go, and I think is usually the done thing - I think the very US-specific tests on yourmorals are from studies that were originally only done in the US and are as abstract and generic as they needed to be in that context. The possibility that tending not to understand abstract, generic questions is a culture-dependent trait seems like a really tricky problem, although in some contexts you might just compromise with e.g. "you pastor, or if you are not religious then someone you trust similarly, tells you..."

Moderation Log