HN Theater @HNTheaterMonth

Well hiring for software engineers should involve fields outside their own while de-emphasizing leetcode. For example, there are psychometric studies comparing personality and job success.

Some of JP's lectures talk on this point quite a bit.

Workplace Performance, Politics & Faulty Myers-Briggs: https://youtu.be/GXHj7eZ23gk

IQ and The Job Market: https://www.youtube.com/watch?v=fjs2gPa5sD0

The reason they give is based on self evaluation.

People are particularly bad at self evaluation.

Keep in mind, most people don't know what their IQ is. And most don't consider themselves to posses less intelligence as they don't know what more intelligence feels like.

See this video for relationship between job expectation and IQ: https://www.youtube.com/watch?v=fjs2gPa5sD0

There is an interesting clip from Jordan Peterson on IQ and jobs. Sure intelligent people want something meaningful from their jobs, but at the lower ends people are probably happier with something less challenging.

https://www.youtube.com/watch?v=fjs2gPa5sD0

Personally I wouldn't mind cycling around all day and staying fit, but keeping software running pays a fair bit better.

A good lecture that regards IQ:

https://www.youtube.com/watch?v=jSo5v5t4OQM

https://www.youtube.com/watch?v=fjs2gPa5sD0

This is a decent article on a true problem, but there are two things I would mention that add context to the situation:

Genetics do matter. Adult intelligence appears to be somewhere between 50 and 80% determined by genetics. Thus we should expect over time in a meritocracy for smart kids to come from the same families. This is compounded by assortative mating by IQ, which has become far more widespread in the last forty years. So things are going to look nepotistic (and they always will be to some extent, see my second point below) when, in fact, this is a natural phenomena.

So, what? Just let the people who lose the genetic lottery rot? No. Instead, we might want to focus on having a more balanced economy where, rather than everyone striving to become part of the elite, a working class person can have a perfectly reasonable life. Jordan Peterson discusses this problem in this video:

https://www.youtube.com/watch?v=fjs2gPa5sD0

Secondly, it is not immoral for parents to make large investments in their own children's lives. This behavior is no doubt somewhat driven by genetics (nearly everything appears to be at least 40% genetic) but it is also a learned behavior. Government programs that replace parental investment have been catastrophic for minorities in the united states: for example 73% of non-hispanic blacks are born out of wedlock. So delegating parental investment in children to the state, in the name of ending nepotism, may have some very detrimental effects that are difficult to predict.

All of this is to add context and suggest a different solution: making middle class non-elite life paths more attractive, not to deny that there is a problem.

⬐ icebraining

assortative mating by IQ, which has become far more widespread in the last forty years

That's curious. Is there any explanation why? Naively, I would assume the opposite, although I can imagine some factors that would point in that direction.

for example 73% of non-hispanic blacks are born out of wedlock.

Why is that catastrophic? Is there good evidence¹ that the kids would have been better off otherwise?

¹ (by good evidence, I'm discounting simple comparisons between children born into and out of wedlock, since it's pretty easy to see why many confounding variables might play into that.)

⬐ oculusthrift

women have started going to college in higher numbers than men. the vast majority of women will only date a man who is equal or higher education level than they are.

⬐ icebraining

That seems to be outdated information: "For the first time in 50 years, the educational balance among married couples has tipped towards women. Wives are more likely to be the better educated partner than the other way around. The trend is particularly sharp among newlyweds; in 2012 almost 40% of college educated women were married to a guy without a degree."

http://time.com/7442/wives-are-now-more-educated-than-husban...

⬐ forapurpose

> Is there any explanation why? Naively, I would assume the opposite

You may not be the naive one, as we don't have any evidence

⬐ None

None

⬐ danieltillett

Greater choice. Men and women have more choice and longer times (at least at the upper middle end) to find a spouse. People tend to want to marry people with similar attributes to themselves. The end result is people tend to marry people of similar IQ to themselves.

⬐ empath75

Even assuming that there is a genetic component to iq, you’re assuming that iq is strongly correlated with success and that the iq difference from genetic differences is meaningful.

⬐ Robotbeat

> Adult intelligence appears to be somewhere between 50 and 80% determined by genetics.

Citation desperately needed. An IQ test, if you've ever taken one, measures a whole bunch of things but not necessarily intelligence. It's a written test which a native speaker benefits tremendously from due to some of the sort of convoluted language used for some of the questions. And many of the mental tricks it tests are things which can be (and are) taught.

Also, even if you somehow proved that some aspects of intelligence were biologically inherited, that still doesn't prove genetics. Current understanding of epigenetics shows that you can pass on traits caused by your environment through mechanisms not in the DNA itself.

⬐ a_puppy

Whether IQ tests measure intelligence depends on how you define "intelligence". IQ tests measure ability to solve visual and/or verbal puzzles, and it's reasonable to call that a type of intelligence. I'm not saying that's the only form of intelligence, but it's clear that IQ scores have _something_ to do with intelligence.

But that question is ultimately a red herring. IQ scores are partially biologically heritable. IQ scores are correlated with job performance. So there exists a biologically heritable factor that's correlated with job performance. That's very relevant for discussions about social mobility, whether or not you want to call that factor "intelligence".

You asked for a citation. The strongest evidence for the claim that IQ is biologically heritable comes from studies of twins. A meta-analysis of 111 separate studies (https://sci-hub.la/10.1126/science.7195071) found that when identical twins were separated and raised in different families, their IQs were correlated at 0.72. Fraternal twins raised together had IQs correlated at only 0.60. Biologically unrelated children raised together (an adopted child and a biological child) had IQs correlated at 0.29. (Side note: Before you demand someone provide a citation, try googling it for yourself. This result isn't hard to find.)

You have a good point that the mechanism could be epigenetic rather than genetic. Here's a study that found a hint of a link between epigentic factors and intelligence: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2892514/. Here's a study finding that epigenetics are correlated strongly between identical twins: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3063335/ On the other hand, the epigenetic correlation decreases as the twin pairs get older, but the IQ correlation actually increases as twin pairs get older. So it's not clear.

It's also worth noting that IQ correlates with income at only 0.23 according to http://www.emilkirkegaard.dk/en/wp-content/uploads/Intellige....

In general, I agree it's difficult to draw unambiguous conclusions from the data we have. I think the data are consistent with the possibility that genetic inheritance of intelligence plays a major role in reducing social mobility, and also consistent with the opposite possibility. But you are demanding a much higher standard of evidence for the claim that intelligence is genetic than for the claim that it isn't, and I don't think that's fair.

⬐ dorchadas

> Side note: Before you demand someone provide a citation, try googling it for yourself. This result isn't hard to find.

Or the person making a claim can actually provide a citation for their claim. It's not on a person challenging a claim to prove it, it's on the person making it.

⬐ Graziano_M

Whatever IQ measures, it's correlated strongly with success.

There are 'cultural bias free' IQ tests that has no words in it at all, and the correlation holds.

⬐ Robotbeat

It should be obvious that the fact that IQ score correlates with success doesn't mean it measures genetic intelligence.

This video explains well the job problem for people who are unable to learn fast:

https://www.youtube.com/watch?v=fjs2gPa5sD0&t=8s

⬐ alternize

that was a great 10 min watch, thanks!

while there are several good points made in the video, one really made me think: if you're in an IQ segment that gives you lot of opportunities - as the prof said: "not just opening word, but programming" -, it is really easy to forget that many people might struggle with "reasonable simple" tasks.

Modern advanced society already requires a high minimum intelligence for people to function in especially to hold a respectable job or be a respectable member of a community. The minimum is increasing when more automation is used for most kinds of automation.

If someone's children will likely have a lower than this minimum intelligence, which you can infer from parent's IQ and other measures, will you encourage or prohibit them to use either embryo selection or genetic engineering? What are the impact on the child's life for each option? Is the 'distaste' more important than giving a better opportunity for the newborns?

Also, what can you do? You cannot put regulations on China or other countries. On this issue, they will not agree on international agreement either. And there will be rich people who fly to other countries to have the procedures done. This will cause even more inequality than allowing it domestically.

Jordan Peterson IQ and Job Market https://www.youtube.com/watch?v=fjs2gPa5sD0

⬐ ajeet_dhaliwal

If someone's children will likely have a lower than this minimum intelligence, which you can infer from parent's IQ and other measures...

Can it really be inferred? Not including conditions or syndromes of course. The parents' environment may have affected their intelligence and the child may turn out entirely different.

⬐ rvern

Yes, it can be inferred: https://en.wikipedia.org/wiki/Heritability_of_IQ.

⬐ cmahler7

intelligence is 50-80% genetic, identical twin studies have shown this many times.

⬐ pcnonpc

With a large number of DNA sequences, the genetic cause can be separated from environmental. That's why embryo selection is possible.

Getting parent's DNAs sequenced might be needed but it's affordable now. See gwern's answers and link for more info.

To flesh out a large range of implications will take at least an effort worthy of a PhD dissertation.

With successful embryo selection, at least we can keep up with the increasingly higher minimum IQ needed to function well in advanced society and avoid a large underclass of unemployable people.

Jordan Peterson - IQ and The Job Market https://www.youtube.com/watch?v=fjs2gPa5sD0

⬐ opportune

I find the underclass argument important and generally ignored. The whole drive for UBI is due in part to the fact that all but the most specialized human professions are being automated. Of course, specialization is not necessarily an indicator of intelligence, but I would guess that they are pretty correlated. Since intelligence is at least in part genetic (and the environmental portion seems to respond very poorly to poverty, which a UBI would likely put people in or close to), it is very likely that a society with UBI would have a very large generational underclass of those unable to sufficiently specialize enough for gainful employment.

If we can make more people capable of becoming specialized, everyone will benefit. Perhaps fields like medicine are limited in the number of people it would make sense to employ, but software engineering and scientific research pretty generally benefit the more people get involved (wrt software I am thinking more in terms of FOSS and startups, not huge teams for simple tasks). The barrier to participation in the future economy might end up essentially becoming IQ. It's important that we give people the tools to help their children be successful as the economy changes.

I agree entirely with you on the urgency of figuring out productive jobs for people with IQs on the left side of the bell curve. Jordan Peterson has a very good video on this[1]. I would couple this with a concern that we end (or, dare we dream, reverse) the dysgenic social policies we currently have in place (IQ is 50-80% heritable).

In the short run, we need to work to move the production chain back to the united states, and that will generate some manufacturing jobs. I suspect more than the corporate-interest controlled media will say, but it almost doesn't matter if we are playing the long game for our kids.

[1] - https://www.youtube.com/watch?v=fjs2gPa5sD0

⬐ sho

This is just a clip from the full lecture here: https://www.youtube.com/watch?v=D7Kn5p7TP_Y

If you like the clip, watch the whole thing. It's great.

⬐ xiphias

Cool, thanks, I'll do that!

Yep.

Jordan Peterson has discussed how important IQ is in the modern economy and how both the left and the right are in denial about it. While the left thinks anyone can be anything, the right thinks lazy people just need to work harder. Both are wrong.

https://www.youtube.com/watch?v=fjs2gPa5sD0

Difficult problem.

⬐ xiphias

Thanks, I loved the video, so I submitted it:

https://news.ycombinator.com/item?id=14173566

⬐ sho

Agreeing with my sibling comment - absolutely fascinating video, thanks. Watched the whole original lecture.

⬐ Spooky23

So what's the "right" answer? The high IQ crowd yucks it up, while the proles live in caves somewhere?