[cperciva]

In certain fields the "go into debt to pay for an education which will help you get a good job" strategy works well. Unfortunately, governments and educational institutions have taken "college graduates earn X% more and have a Y% lower unemployment rate" statistics and pretended that they apply to all college graduates, regardless of their field.

Only a fool would think that a degree in underwater basket weaving would provide the same employment prospects as a degree in any of the STEM fields, but fools are precisely the target market for institutions providing such courses.

[nextstep]

It hasn't been too long since I was starting college. For freshman (in the US at least), colleges encourage students to try lots of classes and take their time picking a major. Many schools don't require students to declare a track until they are in their sophomore year -- halfway through a four-year program! Many can be completed in 2-3 years (even "hard" majors, like physics), and the schools design the degree programs with this buffer so students can wait to make a decision. Of course, the schools don't mind recieving four years of tuition for an education that could be completed in three.

But it's easy to blame the schools. In my experience, students don't really make career choices (like picking a major) by weighing the economic costs and benefits. Our current culture in college encourages students to take on massive amounts of debt regardless of their field. For many liberal arts degrees, this is probably a foolish financial decision. But for many college freshman, the apparent difficulty of a subject is considered thoroughly, while the job prospects of a career path are a secondary consideration.

[cperciva]

for many college freshman, the apparent difficulty of a subject is considered thoroughly, while the job prospects of a career path are a secondary consideration.

Right. My point was that this is partly because of the educational culture which promotes "getting a degree will make you earn more" rather than "getting a degree in subject X, Y, or Z will make you earn more". Students who enter college intending to study science but can't hack it would probably be better off dropping out; instead, most end up in liberal arts and are surprised when their degree turns out to be useless because all the advertising they ever saw promoted "having a degree" as the only relevant flag.

[Raphael]

Difficulty is the deciding factor because the goal is the credential.

[jchrisa]

Economics is the wrong reason to go to college. If you go to college to learn you'll be in much better shape. Even if you study basket weaving.

[barik]

But you can be smart about this by hedging your bets and still achieving the same effect. A degree in basket weaving and a degree in Engineering will both serve to expand your abilities by teaching you critical-thinking skills, teaching you how to communicate more effectively, and, in short, providing you with a framework for "learning how to learn". But one will make it far easier to get a job than the other.

My rule of thumb has always been this: Write down all of the things that you're interested in and would enjoy spending four years or so learning about. Then, from that list, pick the ones that are also employable to some extent. At the same time, don't simply pick a degree solely because of employment prospects, since markets are cyclical, even in classical STEM fields.

[Mvandenbergh]

Right, STEM fields.

Except Chemistry, biology, and biochemistry (massive R&D cuts by the pharmaceutical industry has flooded the market with experienced scientists).

Except CivEng (oops, not much new construction).

STEM degrees are often considered harder by employers, so for many jobs that don't strictly require them, they're used as signalling mechanisms in exactly the way that having a degree at all used to be a signal. That means there's a problem with encouraging more people to get such degrees because you flood the market with STEM degrees and repeat the process of devaluing the signal.

What does that leave? CS degrees are currently very much in demand, but will that remain the case for the next 10-20 years? Will it even be true next year? There were a lot of people who had employment trouble after the first bubble burst, I would hate to see that happen again. Not to mention that this is hardly a solution that will work on a large scale. The industry does not require a million new software engineers every year.

Would anyone have predicted ten years ago that going to law school would be a disastrous financial and career decision for many thousands of people? That was supposed to be one of those "safe" direct-to-career educations. Hasn't worked out that way.