Hacker News Reader

Comment by logifail | Hacker News Reader

(Full disclosure, I have a science PhD)

> the number of awarded PhDs is larger, and therefore the number of Nobel prizes awarded per PhD is lower [..] I don't know what that tells about an average PhD vs average PhD then or 'standards of the field'.

I don't think it's about Nobel prizes per PhD, the article mentions neither the word 'Nobel' nor 'prize'.

> "From 1900-1999, the U.S.’s production of academic doctorates increased a hundredfold, while its population increased by 3.5 times—that is, almost thirty times more PhDs per capita"

That's the issue: Why are we dishing out far more PhDs - and I'm guessing also all other types of academic qualification - per capita than we were before?

Hypothesis #1: We're all getting smarter!

Hypothesis #2: The value of the qualification is dropping...

In the immortal words of so many professors: "discuss" ...

View on HN

Bearing in mind the relatively large number of people that simply weren't the right demographic or lacked the earlier education or money to pursue a PHd in 1900 irrespective of their aptitude, you'd expect that per capita number to massively increase even before we get into universities' incentives to enroll more students

That of course can lead to both 1 and 2 (why not both, and let's not forget that smart high school kids today know things which the very best of the PhD candidates hadn't done the work of figuring out yet in 1900)

> Why are we dishing out far more PhDs - and I'm guessing also all other types of academic qualification - per capita than we were before?

On one hand, the sphere of human knowledge is expanding, so the outermost edges (the surface area, so to speak) are longer, and therefore we need more PhDs who are the people at those outermost edges. It also explains the extreme specialization of those PhDs, whose theses are incomprehensible to normal people.

On the other, there's been a systematic push to put everyone through college, and of course some of those people go on to more advanced degrees. So given the incentives involved it's expected that more people per capita come out with a PhD.

It's a well known saying, but my mandatory HN pedantry must point out that it doesn't work because the surface area grows slower than the volume.

> we need more PhDs who are the people at those outermost edges

<spoiler> Apologies if I'm the bearer of bad news, but most PhDs stop working in science and end up working for corporations.

> there's been a systematic push to put everyone through college

Q: Why is that?

> but most PhDs stop working in science and end up working for corporations.

The bills don't pay themselves, I suppose! Especially if a PhD comes with a hefty loan to pay off.

> Q: Why is that?

I don't know if any single person could fully answer this question (and it certainly won't be me), but my guess is that those well-intentioned folks in power saw that college-educated people earn more money, and applied faulty causation. Put more people through college and they will earn more and make the country wealthier! Put everyone through college and everyone will be better off.

Disclosure: also have PhD and leading a lab for of PhD students

I think not the PhDs are the problem, but the amount of papers. We just need all the PhD students to produce those papers to keep the industry going (much like a snow ball system). There is no incentive to scale down, so we keep scaling up.

The problem IMHO ist that we have scaled our methods to cope with enormous amount of mostly empirical results. P-value hacking is implicit in this system. IMHO we can throw away two thirds of all published research in many fields because it is not reliably reproducible/generalizable. If we could/would use all those PhD students to put science on better grounds that would be great. However, we expect all those PhDs to produce new results, so we get what we wished for.

> We just need all the PhD students to produce those papers to keep the industry going (much like a snow ball system).

I'm not sure we agree on the definition of "we just need" in that!

> There is no incentive to scale down, so we keep scaling up.

Scaling up waffle just produces more waffle, and no-one actually wants that. We seem unable to agree on how to judge progress in academia, so we end up counting papers. Which means everyone's trying to publish waffle.

Judging progress is easy enough. Trusting the judgment is hard, because the world is small and conflicts of interest are inevitable. People who are established experts on a topic generally know each other. Because taxpayers don't want to fund someone's research just because the applicant's friends say so, they need ways of judging the merits without the actual expertise to do so.

Hypothesis #3: Because of the Internet more democratization of information on what a PhD entails, how to apply, and how to be successful has increased the number of smart people who previously would have been interested but didn’t have enough readily available information to fully pique their interest.

Could explain the shift up since the internet.

Hypothesis #4: The work at Los Alamos to win the war on the back of PhDs popularized the notion that smart people could change the world.

Could explain the strong shift since 1940.

> Because of the Internet more democratization of information on what a PhD entails, how to apply, and how to be successful

I'd love to see a working definition of "successful" as regards to getting a PhD.

I spent the best part of 20 years recovering from getting mine, and I'm not kidding.

Sorry, my implied addition there was “how to be successful at getting accepted to a PhD.” Sounds like you were successful at that, at least!

We train 3.5 times more PhD's. What we discover isn't more physics, it's that the "Mythical Man Month" is more broadly applicable than it's writer realised.

Or we’re doing more research. A PhD is not about grading the population on a curve.

Hypothesis #3: More people for professors to offload work to.

> Hypothesis #2: The value of the qualification is dropping...

This is Goodhart's Law.

When a degree is something you have to love the subject in order to get, having the degree is a strong signal. Once the government starts handing out student loans to the average person so they can get the degree, having the degree can no longer prove you're exceptional.

The average person can't be above average. You can open up a paper mill and give everyone a piece of paper, but all that does is dilute the value of the paper.

Reply on news.ycombinator.com