Why? There is no doubt in my mind that if you're the type of student who can get into MIT/Stanford/CMU/Berkeley/etc (at least Berkeley Engineering), you will earn more by not going to college (at least in tech hubs like SV and NY). Assuming you have a marketable skill (iPhone, Android, front-end, Django, Rails, etc), entry level (including internships) starts at $80k. If you don't have such a skill, you can teach yourself over a summer. In four years, you could be making $110k+ (very conservative). A CS grad straight out of a top college will probably make $93k (EDIT: revised to $95k, which is average for a Stanford/CMU/Cal CS grad) [1], depending on the market climate.
You factor in the anywhere between $0-250k tuition you're spending on college (financial aid depending), and the ~$380k you'd earn in 4 years, and, at least on the financial side, you come out way ahead if you don't go to college, anywhere between $380k to $630k ahead, just from those 4 years alone†.
In fact, this probably applies to most colleges, not just top-tier ones, but in which case the salary numbers are more variable.
You also won't be behind in terms of knowledge as long as you're proactive - the resources for self-teaching theoretical CS have never been more numerous than they are today.
Now, that being said, the reasons I'm attending college instead of dropping out (I did take a year off to work at a YC startup): I value the unique social experience of college, my parents are paying for everything (huge mitigating factor - if I had to pay for everything I would not be going), and I see the value in pushing myself out of my social, academic, and extracurricular comfort zones - hanging out with business students, doing debate, taking classes in literature, etc.
[1] https://studentaffairs.stanford.edu/cdc/jobs/salary-grads, http://www.cmu.edu/career/salaries-and-destinations/2012-sur..., https://career.berkeley.edu/Major/CompSci.stm
†Tax not included (subtract ~22%). The effect of having a higher salary due to work experience not included. Number varies depending on your housing costs. Number varies depending on student loan interest. Number varies depending on yearly tuition increases. The $250k figure for college tuition usually includes housing, your employment take-home money will depend on where you want to live in Silicon Valley (SF = big hit; East Bay = fairly affordable).
1. Those social connections? They actually have monetary value. Unless you already come from an extremely well connected background, it's very likely that knowing a few extra people in disparate fields can come in handy for your future career.
2. Without a college degree, you'll sooner or later hit a glass ceiling. I know, I know, it's superficial of people to prefer that guy with a masters degree for promotion to management. But they do it anyway.
3. Last but not least, in college you learn to learn. Most schools these days force you to take a range of classes from history to foreign languages. (Some of these can come in handy on their own, see Steve Jobs' experience with calligraphy.) Learning to adapt to entirely different fields will be quite nice when, ten years from now, that hot new computer skill that's netting you extra cash right now inevitably becomes obsolete.
Let's not mention the fact that there's a bit more to life than raw earning power; such considerations might seem a bit indecent to some of the folks here...
> do not go to college just to earn more money. You need to have better reasons than that.
> Now, that being said, the reasons I'm attending college instead of dropping out (I did take a year off to work at a YC startup): I value the unique social experience of college
Connections are overrated in SV. As a software engineer it is extremely easy to network, since everyone wants to have coffee with you (cough recruit cough). From my year at a YC startup, I made more valuable connections than I've made in college - angel investors (trying to recruit me for their portfolio companies), startup founders, etc. I still see these people, though I do have to make an effort to keep the connections warm. Friendships are a different matter; see the beginning of my post.
Steve Jobs dropped out of college and audited those calligraphy classes for fun as a non-student. So that's kind of a bad example, since if you really wanted to follow in his footsteps, you should drop out and then audit classes at Stanford or something.
Let me give you my experience with "learning to learn:" Most students at top-tier colleges do not really care to "learn to learn." Students are more concerned with taking easy classes to fulfill their requirements than actually learning about subjects they have no interest in. If anything, it is people who are self-taught who have truly learned how to adapt. Who is going to have more developed autodidactic skills? A college grad or someone who had to teach themselves for 4 years?
And while I am incredibly enthusiastic to see how online learning helps to curb the ever-growing debt crisis of the "traditional" university education, I do wonder how will this affect the social aspect of the college years? (Meaning the maturing, the friendships, and the non-classroom learning.)
It remains that in the average case dropping out makes more financial sense.
Internship pay (assuming one position as a rising Freshman, Sophomore, and Junior) will add up to 60-70k, and contracting could be another 5-10k for only a few hours of work every few months. The benefits are, as you yourself mentioned, being able to indulge your passions and dive into esoteric areas without the pressures of the real world. Had I dropped out of university I never would've been able to pursue my interest in quantitative literature, my ex-roommate would never have taken the dive into distributed systems, both decisions which have had a net return over not going, even accounting for lost salary.
> the anywhere between $0-250k tuition you're spending on college (financial aid depending)
Internship pay does assume you get positions all three years.
The BigCos are at the top of the margin and are not representative of the average case.
And if we're not? Do go to college then?
The problem is that your situation is a lot more variable. I know people who dropped out of community college and are make 100k+ in SV right now. I know people in better colleges studying CS who will probably never make that much. It's a lot more variable.
People who drop out of high school are more likely to be mentally incapable of doing high paid engineering work than those who are also able to graduate from an engineering program. The people capable of graduating from an engineering program were capable before graduating. Those who are high school drop outs are more likely to be incapable.
There's a cool study (can't find the cite) where they looked at lifetime wages of people who were ACCEPTED to a given university and then compared those who actually went to college and graduated to those who decided not to go to college after being accepted. The lifetime wages were the same, showing that it was the selection process of getting admitted to college that was simply choosing for talent and not anything that college does to anyone. Those who benefit from burdening the next generation with massive education debt don't discuss this though and claim that it is not selecting for talent. To show that is the case, take people who are in that illiterate high school dropout category, send them to college, and then compare their lifetime salaries to the ones who were not illiterate. This study has not been done. Yes, I am aware there are hundreds of studies proving that college graduates make more than non-graduates, I am not contesting that there are such studies or that they are the majority of studies on this topic.
In any case regardless of whether you believe in the existence of the unnamed study I regret I can't name, I hope it is obvious that the set of people who are high school drop outs and the set of people who have graduated with an engineering degree are not randomly selected people differing only in if a machine called college molded them into a higher earning resource asset. Since they are not randomly selected people differing only if a engineering education has been applied to them as a black box function, it is not correct to assume that salary differences between the two are solely attributable to whether or not they have been run through this black box.
Also think we are shooting a bit behind the bird to try to make current/future value conclusions on any present data as the value of a degree relative to other options is changing so rapidly.
This is precisely what universities do and argue that buying into the massive education bubble is still a "good investment" since you earn more.
I have no doubt there is value to classroom lecturing and the "soft" features like interaction with a wide range of your peers (assuming you choose to leave your dorm room), but the value of college can't be estimated by earnings differentials between college graduates and non-college graduates.
First, I wouldn't assume the same career duration for both GEDs and College Grads - generally, a dropout will start working earlier.
A general "GED average" also pretty highly weights the comparison for other reasons - poverty, area (areas with higher education and wealth levels have higher average earnings, even among those with GEDs), career type. It's an okay generalization to make in this case, but it does bias the results somewhat. People with similar levels of education/wealth tend to group together, which means a lower salary in a poorer area may have similar buying power to what seems like a higher salary in a more expensive area.
Unfortunately what might be the most telling metric is much harder to define: how much would person X make in her lifetime if she went to college, and how much would she make without? Even simplifying to the same field (their strengths would be similar in both cases), it's extremely difficult to plot her career trajectory.
Of course, this may all be due to my personal bias - college dropout, doing fine, and making well above the average starting salary for a CompSci grad by age 25. Decent savings, no debt - it was a much harder road, but I'm not sure I'd make the tradeoff knowing what I do now.
Personally, though, I can't discount the value I've felt in not having the average $27,000 in student debt hanging over my head [1]. It's enabled me to take risks, try new things, change career twice, and follow my passions.
[1] http://money.cnn.com/2012/10/18/pf/college/student-loan-debt...
It's not a big deal because most of the arguments here are relative and not dependent on specific numbers but rather differences. I also understand HN began as a place to discuss Silicon Valley startups. But still I think it would be nice if HN were generally more cognizant of salary variations across location.
