Monday 19 February 2018

Is a degree worth it in 2018?

It has long been a bugbear of mine that the UK Government, oddly led by Blair at the time, took some sharp steps towards privatising the University sector, but then took fright and left it as a bodge job.


Blair rightly brought in student fees. Students would now pay for their courses, albeit with cheap government loans and this would shine a light onto the excesses of the state sector. And lo, so it came to pass that degrees in Madonna Studies and Kite Flying from ex-polytechnics quickly fell out of favour now that some responsibility was imposed.


Equally though, remained the challenge of differing value for courses. Your author hear studied for a degree in history which required occasionally focus for an hour or two a couple of times a week; friends studying medicine or pharmacy worked very hard indeed - in expensive labs with expensive equipment. The move to fees mean all these courses still cost the same, despite the input costs from the university's being very different indeed. Overall, arts degrees subsidise STEM subjects massively.


Now, Theresa May has recognised this and has moved out of the jobs a couple of ministers (Greening and Jo Johnson) who were firmly captured by the sector and repeated the line of how ending this subsidy would be the end of university courses.


To me something way more fundamental has happened in the last 20 years. The internet has revolutionised learning like nothing before. Apps can teach you a language for free, any information can be gleaned from google for free. Research projects now are either laughably easy or arduous in the extreme if, in the latter case, you actually have to find out something new.


Moreover, the big shortage in the economy is in tech jobs, learning tech and programming is just not a university level of study - it is far more akin to A-levels in terms of depth and length. You can't just learn on programming language anyway.


So now is absolutely the right time to review what is being taught at Universities, how much it costs and what subsidies (as ever!) should be directed where. One point is key, people should not be burdened with costs for useless degrees to which they were guided as kids by their betters.

20 comments:

Charlie said...

I don't really see what the problem is with arts subjects subsidising STEM. The latter provide a much larger benefit to the economy. Media studies? Not so much.

You may well be able to learn how to program using information freely available on the internet (indeed, I did), however that's only possible because the info is free and you can do it on the computer you're using to access the internet; equipment costs are essentially zero. Try learning something that requires some real kit, or where the reading material isn't free, on the web - it's much, much more difficult.

The main obstacle to changing our education system such that it actually produces useful output (the blob aside) is the continued insistence of companies on hiring graduates. A lot of the time, the actual subject of the degree is immaterial, which demonstrates the actual value of the educational content of most degrees. That's starting to change in the technology sector, but only in the startup space. Fancy a job at a large corporate? Get ready to show your expensive, yet somehow also completely worthless, degree certificate, or you won't even get your foot in the door.

dearieme said...

The distinction between public and private universities is American: I imagine it works well there. I don't think it does here. The state doesn't own our universities in the way that it owns most of our schools. it simply uses the power of the purse, and parliamentary supremacy, to bully the universities endlessly. (Some crusty old dons predicted this when the University Grants Committee was set up after WWI.) So to talk of "privatisation" is wide of the mark.

Mind you, even in the US the private universities depend on a flood of money from arms of the federal government to fund research. Some day they will pay for that dependence.

CityUnslicker said...

Charlie - good point re companies, but that is because, Blair again (Major hated the idea, after all he barely had an education!), upped participation to 50%> now companies higher whoever has the highest level of education available to do a job. Degree to be a Costa manager - sure, the "market" provides.

My point about privatisation is that half-arsed reform has produced this. Lots of grads that are unnecessary, a big state funded sector which is not productive and a lack of core skills which the economy needs.

Desperately in need of reform, the uni's will hate it as they have built their current models to siphon off the most funds possible - as you would. So change will be painful.

I don't think arts subjects should support STEM. STEM, we need, so the Govt should indeed invest in people, but Arts grads should not bear this burden anymore than a dustman surely?

Anonymous said...

"learning tech and programming is just not a university level of study - it is far more akin to A-levels in terms of depth and length"

My twenty-something year IT career was the result of a six month government-funded course (a very good one run by a blue chip company - there were a lot of cowboys out there). We were recruited via "aptitude test" (basically IQ test) plus interview. Course attendees were mostly 20-and 30-something uni dropouts, we worked hard because we were all in dead-end jobs and wanted out.]

Six months as a trainee with a City financial services company, lo, a new-minted 'IT professional'.

Nick Drew said...

On the subject of degree - is it worth it?: for the whole of my *ahem* several decades in the world of work, (some) people have been saying - you'll neeed an MBA for any career in business

well I never succumbed to that (and it's worked out OK for me) - so that's me voting with my feet. But I still wonder...

what do others think? (we are talking UK business careers here, I don't really care what Americans feel the need to do)

Electro-Kevin said...

Far too many people going to university.

A degree is not special anymore but seems a minimum for many careers that didn't used to need them.

Contrary to popular belief boomers did not get their degrees free. They were not allowed to take degrees at all in the main.

The BBC counters accusations of low representation for Leave experts on TV that 90% of academics are for Remain.

Well 90% of academics should not be academics.

Universities for all increased numbers in the seats of indoctrination - it was all part of the plan to change the electorate leftwards... and also to bond them to the state for good (whilst keeping youth unemployment figures hidden.)

STEM students could easily be supported, there are so few of them relatively.

Y Ddraig Goch said...

RE: CU @ 1:30

"I don't think arts subjects should support STEM. STEM, we need, so the Govt should indeed invest in people, but Arts grads should not bear this burden"

As a previous discussion thread demonstrated, the current state of Universities
is not my specialist subject - but I'm not following where you are going with
this. Degrees exist on a continuum. Some, like medicine, physics and
engineering are extremely valuable and also expensive to deliver. At the other
extreme there is sociology - which is totally useless and cheap to deliver. If
sociology students pay fees that subsidise worthwhile subjects then that is a
good thing (and in fact the only good thing that might ever come from a
sociology degree). If these subjects were priced based on what it costs to
deliver them, sociology would be extremely cheap and engineering very
expensive. Why on earth would we need more sociology degrees (or any at all,
really)?

James Higham said...

"So now is absolutely the right time to review what is being taught"

For who to review? In which form and with what clout?

Anonymous said...

but Arts grads should not bear this burden anymore than a dustman surely?

Why can't a dustman have a degree?

https://www.veolia.co.uk/graduates/

Anonymous said...

I think there are several things going on here.

Firstly the age old arts versus sciences debate, nice refined sophisticated thinkers against crude and rough reductionists. False and not useful, but snobbery abounds.

Then the uni or not debate. Sure, you can train as an accountant or programmer or business consultant on the job and do very well. The problem comes ten years down the pike when you go for that finance directors job. If you are up against a few degree candidates then you come up against 'policy' or 'future planning' or HR. Thank you for attending but ......

Then there is the cost of course/salary question. A proper science degree is expensive but the wages are not always too good. To put a good physicist to work requires a lab connected to some sort of factory and a whole pile of rescources, so the physicist is not paid all that much. A lawyer or accountant puts in much the same effort to get their degree, once qualified give them a laptop and a chair in a Costa and lo and behold they are productive and get to trouser a bit more of the action.

The internet is hugely valuable and free, I just checked my 1978 Kernighan & Ritchie, £17.95 - about £103 in today's money. Money is made at the front end and the back end of any technology.

Anonymous said...

The current move in the direction of the misnomered "serverless" means a pile of coding jobs will disappear over the next 15 years, so anyone looking to move into dev needs to take that into account... Flexibility is going to end up being a valuable skill.

Anonymous said...

@Roger - if HR people are really making finance director positions on the basis of a history degree taken ten years earlier, it certainly explains Carillion.

Anonymous said...

anon 8.33 - do you mean cloud-hosted stuff, or software as a service (SaaS)? I'd not heard servers were vanishing, but may be out of touch. What does a "serverless" architecture look like?

Charlie said...

Serverless doesn't actually mean "no servers". It just means that you don't have dedicated servers running the backend of your app - it's basically cloud computing but with dynamically allocated backend resources.

I'm not sure why Anon thinks a move to serverless architecture means a pile of coding jobs will disappear though... ops jobs, definitely. But, someone still needs to write the code. Care to elaborate Anon?

Anonymous said...

@Charlie and Anon 9:57 - like I said, serverless is a misnomer, the likes of various functions/microservices means that around 80% of functionality will become commidotised. The results will be less efficient than custom code, but the cost of memory/processing/space on scale will work out cheaper than paying for a dev team. Greenfield projects are already slowly moving there, brownfield ones are harder to migrate mainly due to lifting and shifting data whilst maintaining uptime, but that's being worked on.

We've been abstracting functionality along the chain for ages, the final step was always going to be some kind of hypercard type system requiring little technical know-how.

We're almost there now. Security is pretty much the last line of defence for maintaining an in-house setup.

Anonymous said...

"the likes of various functions/microservices means that around 80% of functionality will become commoditised"

I've been hearing this ever since object-oriented coding came in, I suppose it might become true one day. Even on mainframes people tried to make stuff reuseable.

"Security is pretty much the last line of defence for maintaining an in-house setup."

Which is odd in a way. If Megacorp are hosting your system, which consists of a large number of third-party or Megacorp modules/addons bolted together, and you have "little technical know-how" and no dev team apart from bus analyst, designer, a systems integrator and a tester, how do you develop in-house security? Why not hand it to Megacorp and pray it's not offshored?

Raedwald said...

There are layers upon layers of issues here ..

First, social equity and the graduate premium. One guy starts as a trainee chippie at 16 and by 19 is hanging 8 doors a day, with a fat wedge each week. By 55 he's down to 3 doors a day and by 60 he's out of the labour market. His chum does As, takes a first degree and in his mid 20s as a graduate trainee earns just a third of the chippie's wedge, but income lines cross in their 30s and the graduate's income keeps increasing until 70 or more. The lifetime difference in income (the graduate premium) used to be £250k discounted back to year 0. This is why it's fair to charge for going to university.

Second, decline in standards. As has been said, a first degree now is what A levels were 40 years ago - evidence of just-above-basic mental competence, but at a total cost now out of all proportion to worth, except in STEM disciplines.

Thirdly, Universities educating 50% of school leavers instead of the 5% when I took my first degree are ponzi schemes built on sand. Second rate academics making as much money as they can for third rate teaching of fourth rate students and no-one but no-one believes a 2:1 from Durham is equal to a 2:1 from Thamesmarsh University (founded 1991). Also fleecing foreigners for what in effect are 3 year EFL courses is not a sustainable business model. And don't get me started on VC's wedges.

So absolutely time for a review. Current system is just a mammoth waste of national resource.

Charlie said...

Anon: "various functions/microservices means that around 80% of functionality will become commidotised. The results will be less efficient than custom code, but the cost of memory/processing/space on scale will work out cheaper than paying for a dev team"

I just don't buy it. You're always going to need dev to glue the various bits together. Your average application built today already relies on a bunch of off-the-shelf libraries, enough that you're probably already at your 80% figure already. But the old adage - that the last 20% of the work takes 80% of the time - applies here.

Disclosure - I'm not a dev.

Anonymous said...

@charlie - why do you need a dev to glue things together? You have push/pull messaging intermediaries. All you need is a front-end in order to drop things on, and to train the person up.

Open a new diagram, drop in a website item, add a page to the website item, define a form and any validation. Drop in a database item, link the form to the database (future changes can be handled via DACPACs or similar) link the page submission to the database. Drop in a new emailer item, link database INSERT to sending a mail to the address.

I can knock that up in around a week *now* for noddy sites, with a dropdown of site layouts for someone. Give me 12 months I can chuck that out on steroids, and with version control.

And people smarter than me, better devs than me, are working on these now.

@Anon 9:05 - OOP was never really the silver bullet because repos of reusable development code were never that popular until recently. The likes of NuGet and npm have run with that now, but have been shown to be fragile and sources of security issues.

We're actually going to be using functions, no need to look at the code. It's like the concept of MS' webforms, or WWF, but not as shit and without the need for codebehind. You have discrete concepts that can be chained together via communication pipes... It's scary.

As for security, what I mean is that security is the only reason not to outsource to the cloud. You're trusting your data to said Megacorp and removing control from yourselves. How can you be sure you're not sharing hardware, and so opening up yourself to SPECTRE/Meltdown classes of issue? Or someone has uploaded an AWS key (again) to the source control and your data is now open to all?

Anomalous Cowshed said...

Anon : "Flexibility is going to end up being a valuable skill."

Unfortunately, the GCSE / A- level / degree paradigm doesn't provide this. It swaps the general for the specific at each stage, that is, increases specialisations.

Second, there's an interesting effect in IT, when deep specialist knowledge runs slap bang into economics. Essentially, trying to fix a fault on a £300 machine is basically pointless, if it takes more than about 3-4 hours, so no-one wants to pay for that knowledge.

This whole thing isn't going to end well, if similar effects arise for degree holders. They've got a lot of deep knowledge, that nobody wants to pay for.