David McClelland published an influential article (1973) claiming that IQ tests have no value, because they do not correlate with success and it is not clear that they measure anything other than social status. McClelland opened up a new discussion of whether tests predict career success, and whether the purpose of education is social investment or social reformation (why would we even want to single out children with high IQs if those are the children we want not to educate, in order to level the playing field?)

This work is controversial, maybe even more so today than in the 1980s. (Barrett & Depinet 1991) accused McClelland of simply lying, by not mentioning most studies that disagreed with his conclusions and misrepresenting the results of those he did quote.

But in all this time, no one has asked the most-important question: Should we try to make (other people's) children more successful? And should we deliberately promote children because they're likely to be successful?

(If the answer is yes, perhaps we should focus on giving more opportunities to children of the wealthy, since parental wealth is the strongest correlate with career success.)

A close look at (Barrett & Depinet 1991) suggests that, when social class is factored out, IQ correlates well with objective measures of performance, such as employee evaluations, ratings of work samples, and production quantity, but poorly with measures of career success such as job title and salary. Social intelligence is thus the stuff that improves your career but not your performance. That sounds suspiciously like it's skills that help you put one over on your co-workers.

Success is a zero-sum game. It's measured by your position and wealth relative to other people. It makes sense for a prep school or college to advertise that they will make you more successful. It doesn't make sense for a taxpayer-funded school system to do so. Public school is funded by the public in order to benefit the public. The public wants performance, not career success, from you.

It's no paradox that IQ correlates more with performance than with success. Social intelligence does wonders for your career success. People with high social intelligence are able to drive their (often stupid) ideas through committees by using coalition-building and hate-mongering, as well as sarcasm, dismissive humor, emotionally-laden jargon ("death tax"), distraction, and a fine sense of when they can use argument by assumption. They are the people who get grants by schmoozing, playing off the prejudices of the review panel, and snappy data-free PowerPoint presentations. They are the artists who paint a canvas black and then publish a three-page explanation of how that is a criticism of art consumerism. They are good at getting raises, bonuses, and promotions, and at taking credit for other people's work. They are the people who are ruining science and art.

Think that a boss with high social intelligence will make your work more pleasant and resolve conflicts with your co-workers? Maybe. Or maybe that boss will strategically create conflicts to foster competition, and use their superior social intelligence to make you work harder and longer for less pay.

(There is an underlying assumption behind how all this testing is applied that the same skills make a person a good worker and a good manager. I'm not even going to touch that question, especially since behind it lies the even harder question, "A manager good for whom, the company or the worker?")

It can make sense to teach social skills to people who lack them, but it doesn't make a lot of sense to fast-track people for having competitive skills at zero-sum contests. Teaching everyone skills that would maximize their individual competitiveness if no one else has those skills may have no net effect. Putting people into gifted programs or admitting them into more-elite colleges because they have high social skills might mean that people with higher intelligence (and better ideas) will have a harder time getting their views heard. Give me a workplace full of stuttering nerds with pocket protectors, not conniving manipulators.

Social skills may be an important and overlooked part of education. But we shouldn't uncritically overhaul our educational system without looking carefully at what we're maximizing for.


References

Gerald Barrett, Robert Depinet (1991). A reconsideration of testing for competence rather than for intelligence. American Psychologist 46(10), Oct 1991, 1012-1024.

David C. McClelland (1973). Testing for competence rather than for "intelligence". American Psychologist 28(1), Jan 1973, 1-14. doi: 10.1037/h0034092.

David Payne, Patrick Kyllonen (2012). The role of noncognitive skills in academic success. Presented at 21st Century knowledge and skills: the new curriculum and the future of assessment. Los Angeles, California, January 11-13, 2012.

New to LessWrong?

New Comment
31 comments, sorted by Click to highlight new comments since: Today at 5:30 AM

And should we deliberately promote children because they're likely to be successful?

(If the answer is yes, we should focus on giving more opportunities to children of the wealthy, since parental wealth is the strongest correlate with career success.)

Bad economics. You should invest where the marginal utility of each dollar is highest, which is especially unlikely for the children of the wealthy, who will usually already have great amounts on investment made in their future, such that the marginal utility of the next dollar is extremely low.

You don't invest in the likely most successful, you invest where your dollar gives you the most bang for the buck.

It can make sense to teach social skills to people who lack them, but it doesn't make a lot of sense to fast-track people for having competitive skills at zero-sum contests. Putting people into gifted programs or admitting them into more-elite colleges because they have high social skills guarantees that people with higher intelligence (and better ideas) will have a much harder time getting their views heard. Give me a workplace full of stuttering nerds with pocket protectors, not conniving manipulators.

I agree with that.

Better to promote creating value over successfully accumulating value.

The Vikings may have been very successful at marauding and successfully collecting value, but even the Vikings, and perhaps especially the Vikings, would want others to get better at producing value for them to pillage.

A predator can be quite successful, but I'd rather support the creators, and even predators have an interest in that.

Bad economics. You should invest where the marginal utility of each dollar is highest, which is especially unlikely for the children of the wealthy, who will usually already have great amounts on investment made in their future, such that the marginal utility of the next dollar is extremely low.

In the context of state educational initiatives, the investment per student in a gifted+talented program will be the same for every student. So "marginal utility" would be the same as "improved outcome". This means you would like to predict success with or without intervention, and choose kids with the largest difference.

Regardless, the real question here is why you would ever choose "career success" as your metric for a publicly-funded educational program. The assumptions that you can make all kids more successful at the same time, and that this will be good for society, are both dubious.

However, you're assuming that education is viewed as an investment. This isn't necessarily the case. I haven't got the article here, but IIRC McClelland said something to the effect that it would be more fair to single out high-IQ students as kids not to spend money on.

This means you would like to predict success with or without intervention, and choose kids with the largest difference.

That's it. I think it's reasonable to expect that poor children who have had little resources spent on them would see a larger increase in performance that wealthy students who have already received a lot of resource on them.

However, you're assuming that education is viewed as an investment. This isn't necessarily the case.

True. The purpose may be to indoctrinate their minds, to crush souls, or to destroy futures. More likely the purpose of the educational bureaucracy is to enrich itself with an indifference to rival Cthulu and Clippy toward the harm they cause beyond that goal.

But you asked:

Should we try to make (other people's) children more successful?

Not being a moral objectivist, I didn't have a true morality to refer to, but I do have my own, so I provided an answer according to my own.

I haven't got the article here, but IIRC McClelland said something to the effect that it would be more fair to single out high-IQ students as kids not to spend money on.

Even more "fair" to single out smart kids to surgically regress to the mean, if the metric of fairness is equality. That's the problem with equality merchants - destroying value is probably the easiest way to achieve equality.

What about the protective role of social intelligence? You only discuss this very obliquely, if at all - but I'm willing to bet that social skills are ultimately more defensive than offensive. Give me a workplace full of highly socially skilled people - where attempts at manipulation (either by "natural" sociopaths or people trying to apply their skills in opportunistic ways) can be routinely detected and perhaps deterred - and I'll show you a workplace that's better and more polite than one that's full of socially awkward geeks.

Worth looking into, but my sense is it has no protective role. Social intelligence is not a conscious thing. The "manipulators" usually don't know that they're manipulating. Being able to manipulate other people doesn't seem to protect them from being manipulated in the same ways.

The "manipulators" usually don't know that they're manipulating.

Evidence? I think there is an instinctual nature to it, but I think they're aware.

The people who aren't aware are the "intelligent".

Some people have that natural motivation for social dominance, some don't, and both sides tend to model the other guy using themselves as the archtype. This makes the dominators fight harder, and the social pacifists fight not at all, not even perceiving that they are being attacked.

Being able to manipulate other people doesn't seem to protect them from being manipulated in the same ways.

I'd just call that losing to someone who is better.

I suspect that a lot of people who aren't socially dominant in the traditional sense optimize for prestige (higher education to gain income, advertising intellect by getting a Phd, picking up an instrument, etc.) , which can be status enhancing, and thus still a display of dominance.

Of course, since the average person cares about social jockeying, sychophantism, and wit at the expense of others, it may help to learn some of these skills if your goal is to move up in the workplace.

I think the relevant aspect of optimizing for presitige or status is in whose eyes is the status measured. Social intelligence, much like epistemic intelligence, often comes at a price.

My sister and I would talk about compulsive analysis - just having some inconsistency bug you, and feel the need to resolve it. An obsessive compulsive disorder, where you just have to straighten out ideas and figure things out. Naturally, with that compulsion, you will tend to figure out a lot of things that others don't, just as someone with OCD will have a tidy house. (I'm going by the stereotype of OCD as an example, don't know how much the generalization is true).

I think that's much the same with the social dominators, status seekers, and approval seekers. They just feel the need more. Other's approval counts to them. Their status in other people's eyes matters to them. They're motivated by it. Not surprisingly, with all that motivation, they work on it more, and get better at it.

Alternatively, one could chase status, but in one's own eyes. Wanting to be the kind of person you respect. One can go further, and just do what you want to do.

The phrase "just not that into you" applies to these last two types in ways alien to the social status seekers, just as "I don't need to know" is alien to the epistemically compulsive.

Evidence? I think there is an instinctual nature to it, but I think they're aware.

I have no evidence, sorry. There are some studies of sociopaths, who IIRC are usually thought to be aware of what they're doing, but I don't know how relevant that is (is sociopathology a "spectrum disorder", really just a personality dimension; or is it an organic brain dysfunction?)

People with high social intelligence are able to drive their (often stupid) ideas through committees by using coalition-building and hate-mongering, as well as sarcasm, dismissive humor, emotionally-laden jargon ("death tax"), distraction, and a fine sense of when they can use argument by assumption. They are the people who get grants by schmoozing, playing off the prejudices of the review panel, and snappy data-free PowerPoint presentations.

Talk about emotionally-laden! This seem a bit exagerated to me.

Summarizing, the idea is that:

  • high IQ -> better work performance -> better for society
  • high social intelligence -> better career -> better for individual

and since a better career is a zero-sum game, it makes little sense for society to invest in that.

That makes sense, but what's unknown (afaik) is to what extent high social intelligence has (may have) positive effects not just for the individual, but also for whole organizations, society. Career success may be zero-sum game, but a organization/society with a better understanding of the social factor, may be better at reaching its goals.

That makes sense, but what's unknown (afaik) is to what extent high social intelligence has (may have) positive effects not just for the individual, but also for whole organizations, society.

Well, smart people with limited social skills seem to get into a lot of acrimonious but unproductive disagreements which could be smoothed over with a little more empathy and diplomacy.

Yes, I agree entirely. I wanted to raise the question. Educational testers are using career outcomes as a metric for gauging their tests, and that really requires more thought about what the purpose of the tests are.

How teachable is social intelligence? Would a program designed to make intelligent but socially-awkward kids more socially intelligent be viable?

I'm not sure if you mean viable in terms of getting funded, or actually improving the social intelligence of the children enrolled in them.

They do exist if the lack of social intelligence is considered a diagnosis. I had lessons once or twice a week from a speech therapist as a child, partly for difficulty pronouncing letters, and partly for social skills, for that reason. During elementary school, some of this was with the school's speech therapist. I think I also remember a few "how to make friends" lessons with a school guidance counselor, but I'm not completely sure of that. I went to a public school, but probably an unusually well-funded one - I'm not sure how usual it is for schools to provide this sort of help.

I think something must have worked for me, because my social skills don't feel deficient in most situations I've encountered as an adult, and most of my friends who had similar problems as children do still have some problems like this. This sort of socialization could also be done through an institution, though - maybe make membership to after-school activities a requirement of the gifted program, or include more group work in advanced classes. There's also the risk of causing anxiety problems, though - being in a room full of children is stressful if you can't read them well enough to predict what they're going to do next.

I'm also not sure if what I have counts as increased social intelligence - I might just have done the equivalent of memorizing multiplication tables. (Communicating with new people takes more mental energy than people who I'm used to, as though I need to re-learn some things for each specific person.) Practically, it seems to be enough to get up to the waterline, though.

This sort of socialization could also be done through an institution, though - maybe make membership to after-school activities a requirement of the gifted program, or include more group work in advanced classes.

I had a thought on this subject earlier, mostly in the context of gifted students who are bored as hell with standard classes: Have the students who master the material first help teach the rest. They learn a social skill and there are more teachers to go around. Keeping it in the classroom stops it devolving into either the smart kid doing everyone's homework for them, or everyone else beating up on the smart kid for being uppity. Additionally, it would (I think) improve the social status associated with intelligence and academic success, which is a significant barrier in some places.

Making people perform a social task is not the same as teaching a social skill. If the skill involved isn't taught, or is just too hard, then the solutions of "make them teach" or "make them do highly social extracurriculars" or "make them do group work" are straight-up punishment for success and make the "bored as hell" option look pretty attractive.

Yes - it's more forcing someone to learn social skills, than actually teaching social skills. Practicing a whole lot works as a way to learn, but hopefully there is a more efficient, less unpleasant way.

I also know a girl in high school who had asperger's syndrome and was an extreme extrovert (she felt bored if she didn't have an outing or a visit from a friend almost every day if it wasn't a school day), and she still seemed to have significant trouble with social skills. So I guess that might mean that either practice doesn't work for everyone, or doesn't work as well as I had thought.

Hrm. Good point. Now that I think of it, I would have hated my own suggestion at the age to which it applies, although I'm a fairly extreme introvert so I may not be a representative case.

Some of the most well known "social intelligence" classes were held by Dale Carnegie in 1912, and many business schools still make students read his book. Nowadays social training exists in many forms, and there are programs aimed specifically at kids as well, although maybe nothing aimed at making future computer scientists less awkward. Studies show positive results, but mainly treatments for more serious behavior problems are studied.

Is there reason to believe that social skills are more difficult to teach than math, or rationality?

Is there reason to believe that social skills are more difficult to teach than math, or rationality?

Yes. They're very hard to understand. It's hard to teach something you don't understand.

Auxiliarily, I'd expect that common sense would kick in and people would feel confident in contradicting teachers.

(That said, it seems likely to me that they can be taught.)

IQ correlates well with objective measures of performance, such as employee evaluations, ratings of work samples, and production quantity, but poorly with measures of career success such as job title and salary. Social intelligence is thus the stuff that improves your career but not your performance.

The second sentence here contains a slight of hand. Just because IQ doesn't correlate with success, doesn't mean that social intelligence drives success. It could be something else altogether, like conscientiousness or physical attractiveness.

It's by definition, if you define social intelligence as "personal skills not measured by IQ that correlate with career success". I haven't seen this definition explicitly, but that's what you get when you choose career success as your metric and factor out IQ.

I'm just claiming that "social intelligence" is loaded terminology if you haven't yet demonstrated that the things which we normally think of as "social intelligence" (courage in asking for raises, for example) actually are the main driver of career success modulo IQ. If we did a study and it turned out that most career success was driven by physical attractiveness, then your definition of "social intelligence" would still be logically valid, but it would be awful terminology.

Good point.

Let's say IQ test do correlate with success (as measured by conventional standards). What would that prove? That a society that values high IQ rewards people with high IQs. The relevant question is Is IQ a valid measure of intelligence? Well, good luck defining intelligence in a scientifically meaningful way.

"Social intelligence", oh boy... At this point we're just giving common sense wisdoms--flattery gets you everywhere/the socially adept rise higher in social contexts etc--a lacquer of scientistic jargon.

What would that prove? That a society that values high IQ rewards people with high IQs.

You do realise that it's rare for co-workers to know each other's IQs? Obviously there's a third thing that both IQ and success correlate with.

My point was hypothetical. I skeptical a correlation actually exits,--damn lies in all--but that's beside the point. My point is a society that is into boiling complex, difficult to define concepts like intelligence down to a simple metric is liable to have lots of other analogous, oversimplified metrics that are known, if not to coworkers to teachers and whoever else makes the decisions. And I'd wager people who do well on tests are apt to be the same ones who get high marks on Cognos reports--i.e., the same prejudices affect what's deemed valuable for both.

As far as our actual society, there is only partial truth to this, we are metric obsessed of course--but nepotism and, more than anything else, the circumstances one was born into, probably play the biggest role apropos success as conventionally defined.

I'd wager people who do well on tests are apt to be the same ones who get high marks on Cognos reports--i.e., the same prejudices affect what's deemed valuable for both.

Well, fair enough.

i think it's a inseparable trinity! When i was choosing a speciality to study for i relied on those one where credits were provided. I realized that diplomas and what there is written mean nothing for employers. So I've chosen philologist and get a scholarship. But now i've faced with the problem of where to go to work and how to make my resume effective, i solved this problem with this guys https://pro-papers.com/resume-writing-service . I hope i will find the best job for me.

To my knowledge most differential psychologists wouldn't claim that intelligence is the moderating effect per se when it comes to success. After all IQ is merely defined as "what the intelligence test measures". I don't know enough about social intelligence to comment on that really, but I do like your point. I guess we have all seen studies in diagnostics suggesting that incremental validity of most variables is small to non-significant when compared to intelligence. But what does that tell us if we don't have a real understanding of what we actually define as "intelligence"?

Also I think one should differentiate between fluid and crystalline intelligence. While I would expect the former to correlate with basically any measure of success the later one prolly doesn't by too much.