The Atlantic article: At Work, Expertise Is Falling Out of Favor

[polly_mer]

https://www.theatlantic.com/magazine/archive/2019/07/future-of-work-expertise-navy/590647/

High in fluid intelligence, low in experience, not terribly conscientious, open to potential distraction—this is not the classic profile of a winning job candidate. But what if it is the profile of the winning job candidate of the future? If that's the case, some important implications would arise.

...

In the stable environments Duckworth and Gladwell draw from (chess, tennis, piano, higher education), a rigid adherence to routine can no doubt serve you well. But in situations with rapidly changing rules and roles, a small but growing body of evidence now suggests that it can leave you ill-equipped.

Paul Bartone, a retired Army colonel, seemed to find as much when he studied West Point students and graduates. Traditional measures such as SAT scores and high-school class rank "predicted leader performance in the stable, highly regulated environment of West Point" itself. But once cadets got into actual command environments, which tend to be fluid and full of surprises, a different picture emerged. "Psychological hardiness"—a construct that includes, among other things, a willingness to explore "multiple possible response alternatives," a tendency to "see all experience as interesting and meaningful," and a strong sense of self-confidence—was a better predictor of leadership ability in officers after three years in the field. Thus, Bartone and his co-authors wrote, "traditional predictors [of performance] appear not to hold in the fast-paced and unpredictable operational environment in which military officers are working today."

https://www.chronicle.com/forums/index.php/topic,264128.msg3641578.html#msg3641578 has a great post from Watermarkup with an example of how someone who was well on the track to a great career would be angry and bitter when the rules change.

And then if after a decade playing against NBA pros as a point guard with a deadly 3-point shot you can't get a job as a pro basketball player because the number of teams in the league gets dropped from 32 to 24
...and a rule change reduces the roster size to 9
...and another rule change eliminates the 3-point shot
...and another rule change replaces the point guard with a robotic clown that tosses the ball out into the court at random

then I'd say yes, there are good reasons to feel some bitterness and resentment.

One of the standard arguments for a liberal arts education over mere job training is the flexibility for a lifetime, not just the first job.  However, I do have to wonder if some people get mentally stuck in adjuncting mode because they're sure that clown will go away and the rosters will go back up as soon as people realize their mistake in making the rules changes.

[marshwiggle]

https://www.theatlantic.com/magazine/archive/2019/07/future-of-work-expertise-navy/590647/

Also from the article:

a world in which mental agility and raw cognitive speed eclipse hard-won expertise is a world of greater exclusion: of older workers, slower learners, and the less socially adept. "This sounds absurd," retired Vice Admiral Pete Daly (now head of the U.S. Naval Institute) told me, "but if you keep going down this road, you end up with one really expensive ship with just a few people on it who are geniuses"

The more adaptability you need, the fewer people there are with the "fluid intelligence" that can handle it. This isn't an argument for some sort of general education over learning specific skills; rather it's an indication that the most valuable people to society will be the very small group of brilliant people who are quick studies. Something which NO education can produce.

[quasihumanist]

Something which NO education can produce.

Is there evidence one way or the other for this?

Since I've been a grad student, my primary pedagogical goal has been to improve the fluid intelligence of students.  Have I always been attempting the impossible?

[mamselle]

Yes, I figure I'm helping those with some degree of potential locate it within themselves and learn how to actualize it with joy.

I don't think there's an intellectual oligarchy, just a lot of folks who hold back on themselves and others because of--oh, I don't know, fear, uncertainty, lack of encouragement in their home or social setting, the seduction of pleasant alternatives that drain vital energy away from ones gifts, etc.

If one has been lucky enough to have experienced the kind of encouragement and habilitation that free ones potentialities, one teaches those things along with subject matter as a matter of course.

And if others are wise, they'll be grateful, not cripplingly competitive, with those who can add value to their lives as well.

M.

[marshwiggle]

Something which NO education can produce.

Is there evidence one way or the other for this?

Since I've been a grad student, my primary pedagogical goal has been to improve the fluid intelligence of students.  Have I always been attempting the impossible?

I'm not a psychologist, but everything I've read is that crystallized IQ, i.e. skill at a specific task, can be improved by practice but fluid IQ cannot. All of those websites offering quizzes to keep your brain active are, from what I've read, overselling since exercises will only improve your ability in similar tasks.

Also, the article mentioned the "ideal recruits" being high in Openness and low in Conscientiousness; from all that I've read on personality, those traits are also pretty much static over time, or at least they don't really change by any sort of training. So both fluid IQ and personality are things which education can't really change.

[dr_codex]

There's a bitter irony that the navy is trumpeting its new training systems. Several recent collisions with tankers have been at least partly attributable to throwing Ensigns onto bridge watch positions almost the day that they are commissioned. It seems that adaptability may not be more important that actually looking out the port side window every once in a while.

Those of you with teenage drivers in your households might want to reflect upon whether you'd rather ride with the "conscientious" kid or the "open to new experiences =~ distractible" one.

Those of you without teen driver might ask the same question of the pilot on board your next 737.

Bluntly, US Navy training is a mess, and I would be really cautious about emulating any part of its experiment.

[polly_mer]

Bluntly, US Navy training is a mess, and I would be really cautious about emulating any part of its experiment.

The article used the Navy as an example of the current state of employment in many places.  To the Navy's credit, they purposefully are experimenting.  Many other sectors are just doing without any experimental design or claims at all.

The median length of time in a job for an individual is 3-5 years, depending on how you want to slice the data by age, gender, and race (https://www.bls.gov/news.release/tenure.nr0.htm).  Even looking at the Boomers in their 60s, only about half the employees had 10 years in the same job.  Professional class folks, those who will almost all have college degrees if not graduate degrees, were at 5.5 years.

At my current employer, we new people are pointing out that assuming that the norm is to spend 20 years in a job ladder akin to assistant/associate/full professor is going to fail now that the pension doesn't provide a very, very good carrot for doing so with a stick for leaving too soon.  Now that we're all on 401k plans so the penalty for leaving is negligible, the traditional practices assuming that people will find a niche and then spend 10-15 years getting real depth aren't working.  This is especially true since the way projects are assigned (too much work, too few people with the relevant depth) actually precludes the newcomers from developing that depth as we get 4-6 new projects every year based on deliverables promised instead of some purposeful development in 1-3 areas with all projects helping develop those areas.

Yes, that's worrying as a cultural shift, but that's the reality on the ground, even in some parts of academia.

[marshwiggle]

Bluntly, US Navy training is a mess, and I would be really cautious about emulating any part of its experiment.

The article used the Navy as an example of the current state of employment in many places.  To the Navy's credit, they purposefully are experimenting.  Many other sectors are just doing without any experimental design or claims at all.

From my read of the article, one of the problems the navy had with people wearing multiple hats was about who was responsible for a given task, rather than about how well it was done. This makes sense. To be honest small organizations have always worked with people sharing tasks like this; what's novel is applying it in a much more formal setting like the military.

(Anyone remember the Asimov short story about sometime in the future, when they realized they could make spaceships require vastly less computer resources if they crewed them with humans who could do math?)

[polly_mer]

(Anyone remember the Asimov short story about sometime in the future, when they realized they could make spaceships require vastly less computer resources if they crewed them with humans who could do math?)

One of the huge changes in my adultlife time is automating many repetitive tasks and distributing other repetitive tasks to eliminate the need for secretarial and other pools of diligent humans of average intelligence.

The bar for humans who can do math has gone substantially up because the computers are good, fast, and cheap enough that a cell phone has more computing power than we had to put humans on the moon in the 1960s.  We still need humans who can do math, but being fast with algebra and trig is no longer a solid lock on anything.  My elementary schooler can use a spreadsheet to collect and analyze basic data because he doesn't have to know all the math underlying the algorithms; he just has to know the name of the function he wants and find it on the list.

Computer programs that do the symbolic manipulation of equations have become amazing; no more nearly daily paging through Abramowitz and Stegun for anyone.  The hard part remains figuring out which equations to write to describe the situation, how to figure out all the necessary constants, boundary conditions, and initial conditions, and then how to solve all those coupled equations (often partial differential equations and more complicated) perhaps millions of times to get a good simulation.  There's a reason we use programs like Abaqus in engineering after students know the basics of conservation equations.

I have to laugh every time I remember a humanities colleague who indignantly exclaimed, "Your physics class isn't that hard; all you do is solve a couple equations".  Yep, the hard part of intro physics is not the math; it's figuring out which equations to solve.  The hard part of my current research isn't even figuring out which equations to solve.  The hard part of my current research is

(a) figuring out which of the hundreds of variables are really physical and getting good values for them from the experiments that often were not designed to isolate those parameters

(b) figuring out which variables are purely numerical and figuring out which values for them will promote stability enough that the equations can actually be solved numerically

and

(c) sighing heavily upon realizing that the equation as written in the paper isn't really the equation in the computer because some human made assumptions to make their coding job easier without actually having expertise in either the physics/chemistry/engineering or the computer science numerics to make a good choice.

One human can work through 5-10 pages of deriving the mathematical equations.  One human is much more daunted upon facing the results of teams of developers working for at least a decade on a million lines of code written in multiple languages with various wrappers with hundreds of individual files linked into dozens of libraries and fed through a couple compilers so that running "the same code" on different machines doesn't necessarily give the same-enough answer.

The machines can't take over everything yet, but we're way past the stage that learning a little math is a big improvement in employability for the average person.

[marshwiggle]

I have to laugh every time I remember a humanities colleague who indignantly exclaimed, "Your physics class isn't that hard; all you do is solve a couple equations".  Yep, the hard part of intro physics is not the math; it's figuring out which equations to solve.

Or what boundary conditions to use.

I realized part way through graduate level Electromagnetic Theory; the whole course is just Maxwell's equations with a bazillion different sets of boundary conditions.

The machines can't take over everything yet, but we're way past the stage that learning a little math is a big improvement in employability for the average person.

I agree. The point that resonated with me is that human adaptability is really valuable. However, the more adaptable they are, the rarer they are as well.

[spork]

The expertise of shipboard Navy personnel sucks because of https://www.navytimes.com/news/your-navy/2017/08/27/navy-swos-a-culture-in-crisis/. It's "learn how to drive a ship with CD-ROMs and never speak up when you see a problem so you can climb the ranks." If you don't know that you need to look out the window, it really doesn't matter if you have deep expertise in one specific skill area or intellectual flexibility to have superficial expertise in many skill areas.

A very small example of automation: professors used to order up photocopied sets of assigned readings. The photocopies were made by either a human department secretary or other humans in a copy center. When I started teaching online I sent a photocopied material to a librarian for scanning and placement in the library's catalog of digital course reserves. I copied the chapter/article myself, because departments no longer had secretaries. Now I don't even bother going through the library, I put the digital version of the reading in the course's LMS website myself. So no need for a human librarian to do that work. And soon the three other human library staff at the circulation desk will be reduced to one (if that), because library patrons can now check books out themselves by scanning their university IDs at kiosks.

[biop_grad]

The question therefore is: with increased automation, what is the face of education?  And how would one ensure that those who are not lucky enough to have one of the newer, more flexible jobs can have a reasonable standard of living?

[polly_mer]

The question therefore is: with increased automation, what is the face of education?  And how would one ensure that those who are not lucky enough to have one of the newer, more flexible jobs can have a reasonable standard of living?

What to do with people who can't/won't/don't get the few remaining jobs really is the interesting and more pressing-every-day question, isn't it?

I don't know the answers, but I did read a pretty good, albeit rambling, trio of blog posts years ago where one recurring question was: should we stop doing student loans and just use that money for lifelong food stamps and rent subsidies?  https://thelastpsychiatrist.com/2012/11/hipsters_on_food_stamps.html is part one of that trio.

[ciao_yall]

The question therefore is: with increased automation, what is the face of education?  And how would one ensure that those who are not lucky enough to have one of the newer, more flexible jobs can have a reasonable standard of living?

Programming all those robots to do the work. Kind of like during the last century.