This isn't even close to an apples-to-apples comparison and looks like mostly an advertisement. You're looking at broad academic institutions preparing students for a huge range of careers vs. an institution that takes advantage of quickly ramping up a student for a specific career that currently has very good average salaries.
Besides that, I don't get this:
"Part of it is that the cost of delivering that education are very reasonable."
He then preaches about how higher ed's cost structures are terrible. But when you look at the cost per week of education, Flatiron is over 3x higher than a 4-year college. That doesn't suggest to me that higher ed's costs structures are all that ridiculous compared to Flatiron.
Overall, I get the impression that Flatiron is providing a valuable educational program to its students. However, if you really want to impress me or inspire others, show me how this model can be successfully applied to careers outside of software engineering.
It would seem a better argument would be made for time spent rather than dollars spent.
> However, if you really want to impress me or inspire others, show me how this model can be successfully applied to careers outside of software engineering.
My opinion is that most jobs don't require any of the skills taught in college (sure you probably need to write well, but you can learn that in high school or just by reading a lot... I'm sure almost no one needs any math aside from how to work a calculator).
The two primary benefits of college are: First as a signal to employers to say, "hey, I did something moderately difficult so I'm a relatively safe bet." And second, as a place to develop social skills and the ability to interact with people different from yourself.
I once read an interesting article comparing a college education to high heeled shoes.
High heeled shoes were once a utilitarian device: They made it easier to keep a solid grip on the stirrups when riding a horse. As footwear, the feature embodies a tradeoff that's only practical for a particular class of people: Riding a horse is easier, at the expense of making walking a bit less practical. But since this class of people happened to be wealthy, and it's common for folks to want to mimic the habits of those wealthier than themselves, eventually a fashion for high-heeled shoes developed among non-equestrian classes. That development also divorced the shoes from their practical roots, which allowed fashion to take over and push the shoes to even more impractical extremes, which is where we get modern incarnations such as women's pumps. To some extent, the only real purpose served by such shoes is to signal to others that your day-to-day life doesn't involve doing anything that would render them too impractical, and therefore you're of a higher social status than anyone who doesn't wear them.
Higher education seems to be starting down a similar path. It used to be that it was a thing for the upper classes, who were expected for various reasons to have a well-rounded liberal arts education. Because of this it became desirable in part because wealthy people do it. Nowadays that seems to be a major driving force. How else can you explain why a time-consuming and expensive BA is still considered more desirable than one of the trades, many of which offer comparable or even better income prospects?
> How else can you explain why a time-consuming and expensive BA is still considered more desirable than one of the trades, many of which offer comparable or even better income prospects?
I disagree. If you look at non-technical white collar jobs, they all involve mainly analytical and persuasive writing. It doesn't really matter if you're writing essays picking apart Foucalt, or writing a memorandum explaining why the company should buy this kind of server versus this other kind of server. It's the same skills.
But then you'd be able to reduce things down to a relatively short writing curriculum. Many university courses involve writing, but they are not primarily designed to improve students writing.
Some universities force students who are behind in writing to take courses to catch up before they can even enroll in the normal liberal arts classes. You have to be decent at writing before they'll even let you read Foucalt.
The US government has proposed a draft framework on ranking higher ed institutions based on Outcome. This gets closer to the ROI model for evaluating any program/college.
Access
-Percentage of students receiving Pell grants
- Expected Family Contribution Gap.
- Family Income Quintiles: number of students from low and moderate income families.
- Percentage of First-Generation Students
Affordability
- Net Price: what a student actually pays (cost of attendance minus financial aid).
- Net Price by Income Quintile: what a student actually pay by income class.
Performance
- Completion Rates.
- Transfer Rates
- Labor Market Success: The most controversial metric by far on the list.
- Graduate School Attendance of Former Students
- Loan Performance Outcomes: A metric to be determined that will examine whether or not students are able to repay their loans after graduation.
Also, I'd like to see statistics about how much of Flatiron-like schools and traditional schools graduates manage to get above-avarage salaries in their field. Comparing with other fields makes little sense.
Part of a VC's job is advertisement (for their portfolio companies) ;-) However, in this case I couldn't find a past association between Flatiron school & Fred Wilson and/or Union Square ventures, so this article seems to be expressing genuine appreciation for the model. Maybe he likes them so much that he will invest in them in the next round :D
To understand the cost of education argument, notice the catch lies here: "teaching students skills that are directly related to job requirements." How much time of a traditional 4 year college is spent on skills that a grad would actually need or use? A lot of it is just archaic legacy. If you take those out, then the cost of education might actually compare favorably for bootcamps.
An additional (or alternate perspective) is that bootcamps are a supplement to, rather than a replacement for traditional education. Bootcamps teach hard job skills, whereas college teaches soft & social skills. Either way, they are a welcome addition!
How much time of a traditional 4 year college is spent on skills that a grad would actually need or use? A lot of it is just archaic legacy.
I thought it was moe that universities are meant to promote general enlightenment as much as (or, primarily rather than) useful trade skills. With a fair bit of confusion coming in where some skills (engineering, S/W development) depend heavily on (parts of) said enlightenment (math, how people think).
That was always a big argument when I was in and around a university computer science department: how much time you spend on "general enlightenment" topics like linear algebra and how much to spend on specific, valuable skills, like x86 assembly.
If there would be 1000x flatirons pumping out graduates, startups would have greater access to lower cost developers, which i'm sure would make VC's job more profitable,and easier.
Why are you comparing cost per week? You can't spend an equivalent time in college to flatiron and gain anything meaningful. The value is in the unit of a degree or the completion of the program. You can't partially invest in college and expect to get proportionate consideration or results. Likewise, you can't really do anything with 1/3rd completion of flatiron.
That's the value you get at the largest scale. There are several smaller scales at which education delivers value. There's the individual course, and there are the individual skills you learn.
I'm going to go ahead and guess that the flatiron school is not teaching much in the way of linear algebra or discrete mathematics, for example. That may not be a hindrance in many software jobs, probably most, but both are absolutely essential to the things I do day-to-day. I have met precious few developers who majored in something other than CS and can really grok a lot of the stuff I work on, simply because they don't have the education.
The exception is people who've taken the time to flesh out their knowledge with online classes from sources like Coursera. This gave them nothing comparable to a degree or completion of an educational program. They just took one class, and it wasn't even 12 weeks long. But that one class still provided enormous value.
The other exception is people who did attend a college-level CS program, but didn't complete it for whatever reason - dropped out, switched majors, whatever. They didn't get that BS in CS, but they still got a lot of value and were able to do something with it.
Because the article was trying to say that Flatiron's costs were lower, yet they're charging more.
Granted, my calculations were dumb. I took the 4.3 years the article used as time to finish a 4-year degree and multiplied it by 52. Obviously, the number of actual weeks in-class is different, but I'd imagine most universities costs during the various breaks and lighter summer sessions aren't substantially less than in the middle of a fall or spring semester.
Because what people charge is based on what their costs are. If you have to provide staff, location, etc. for a longer period of time it drives up your costs which in turn drives up your prices.
Or you could look at the traditional universities teaching a "whole range of careers" for people who may or may not need most of them as being hugely inefficient.
Maybe a better model is to spend $15,000 on a crash-course for a new career, do that for a while and if you're unhappy with the career choice, pay another $15,000 for another crash-course, and so on.
One issue with this proposal is that I still find $15,000 way too expensive. It may be "much cheaper" than paying $200,000 for a bundle of courses, but it seems no cheaper on a per-course basis. Surely we can have education in the future that's much cheaper than that? We should look for disruption in that as well.
A huge problem with this comparison is that the field of software engineering is really, really, really hot right now. So you can get away with studying up for 16 weeks, and get a $70k job. We've been here before -- http://www.defmacro.org/2013/12/09/learn-to-code.html. When the field encounters a short-term cool-off (as it inevitably will), these hires will be the first to go because 16 weeks is absolutely not enough to get a strong base of fundamentals. You could (barely) learn Rails and JavaScript in that time, but you really aren't getting a strong enough engineering skillset to build a resilient career.
The bigger problem is that high school seniors don't understand basic economics of supply and demand. The job market is becoming polarized -- top people in their fields are doing better than ever, and the market doesn't really need anything else (with software engineering currently being an anomaly). So I'd say it isn't worth going to college (from an ROI perspective) if you aren't positioned to be in the top 10% of your field. This used to not be the case, but the world is very rapidly going in this direction and I'm not sure if there is a solution to this problem, any more than there was a solution to manual labor jobs disappearing post industrial revolution.
> EDIT: When you always add and never subtract, you get cost structures that are not sustainable.
I started college at the start of the dot com bubble. I remember network engineering was the hot field of the time and there were places offering fast turnaround and relatively cheap Cisco certification training.
These days, I wonder how many companies are hiring CCNA's with no other higher education.
Exploiting a short-term market condition is revolutionary, even if it looks different than the status-quo while the market condition exists.
This blog post is extremely misleading: the three featured alumni [0] on the Flatiron School's web site all have 4 years of college at good schools (one even has an MFA)
Saron Yitbarek - University of Maryland College Park
B.A. English, B.S. Psychology
2007 – 2011 [1]
Justin Belmont - Tufts University
BA, English
1999 – 2003
Columbia University in the City of New York
MFA, Nonfiction Writing
2005 – 2008 [2]
Danny Olinsky
University of North Carolina at Chapel Hill - Kenan-Flagler Business School
BSBA, Business Administration; Entrepreneurship [3]
This, of course, is the irony of the higher education debate. Those having the debate very likely did not go through the voc-ed tracks (which is effectively the "revolutionary" model the Flatiron School is providing).
That doesn't mean there's no value in providing education that leads directly to career opportunities or salary improvement. As an American I know our culture loves to yell "But!" and provide an example of a VC or entrepreneur who pulled themselves up by their bootstraps, but these are the exceptions rather than the rule.
It's not an apples to apples comparison. As a Flatiron grad myself I can tell you that most my classmates already had college degrees. They do have a program that serves as a substitute for college, but it's not clear that the main program has been filtered from the data.
That said, I thought Flatiron was an exceptional investment, much higher ROI on my salary than college, but also served a fundamentally different purpose and role in my education than college.
You can't discount your traditional education from your ROI on Flatiron. They aren't exclusive investments. In cases like yours, Flatiron is complementary to other education.
If Flatiron wants to make their ROI case stronger, they should look at people who have Flatiron as their exclusive post-secondary education.
Despite the obvious flaws of comparing a coding "bootcamp" school to a 4 year higher ed institution, I think there is a very valid argument in thinking about ROI when picking degree choices. While not everyone can be coders, (if everyone could, then the salary/demand would drop and this solution would be invalid) everyone should seriously think about ROI. Counselors almost never discuss ROI when advising 17-18 year olds on college choices. Questions like, can you pay off a 6 figure debt to pursue a comparative literature degree?
If you pick a program that has high salary/demand, you will have better ROI. Even if you spend lots of money on it, you will be able to pay it back. That is ROI. (Not just cost by itself)
Here is a list of top ranking salary schools. They are almost all dominated by Ivies, tech/engineering (and interestingly, Military) schools.
> While not everyone can be coders, (if everyone could, then the salary/demand would drop and this solution would be invalid)
I think we really can't know how much the supply of programmers would rise if we gave everyone access to decent education. The state of programming education is abysmal.
Today almost every good programmer is also an autodidact. That's pretty weird when compared with other professions, and I predict that things won't stay this way forever.
Education doesn't guarantee a higher salary - instead it opens up opportunities and raises each person's potential ceiling. This makes using averages and ROIs very difficult.
I could have gone to med school (historically a very high ROI degree) but I'm terrible at memorization/studying and probably wouldn't have thrived i.e. wouldn't have reached anywhere near its ceiling. Obviously, it would have been a poor degree for me.
Or, I could have gotten an MBA (another very high ROI degree), but my salary probably wouldn't have immediately been higher than what I have as a software engineer/data scientist. But an MBA would enable me to enter an executive track that is mostly closed off to me now. Is it worth it? Honestly I don't know - its an extremely complicated calculation.
Instead of ROI, I think it's worth thinking about how much each degree will cost you (both time and money), what it will enable you to do, whether you are capable of doing that, whether you want to do that, and what the payoff will be. If I were a counselor, I would go through this exercise with my students.
ROI is loosely all that. Outcomes (salaries, happiness, growth..etc) have many different metrics. I'm agreeing with what you're saying even though they may not be "ROI" to you.
Counselors do go through a lot of the non-salary career choice exercises (if you had a million dollars what would you do scenarios) but rarely go over the "money" talk.
"Instead of ROI, I think it's worth thinking about how much each degree will cost you (both time and money), what it will enable you to do, whether you are capable of doing that, whether you want to do that, and what the payoff will be."
Um, are you quite clear on what "ROI" is? You just described it there, while making it sound like you disagree. As with any future-looking thing, it's all probabilities in the end. There aren't any guarantees in life.
> As with any future-looking thing, it's all probabilities in the end.
Actually, this was the exact opposite of my point and my own personal experience. On average, a medical degree would have paid out more than the option I went with, but I ignored this simple ROI calculation. Instead, I did a deeper analysis, mixed with some intuition about myself, and went with what was at the time a fairly low ROI option (masters in AI/CS - at a time when the average MS in CS was paying 5k more than a bachelors). I did this because I knew it would open certain doors for me, not for any monetary gain. I'm confident it's served me better than the "just go to med school" advice I was repeatedly given when I was 20 - advice that was strongly backed by every monetary statistic you could find.
So, maybe that's still an ROI estimate, but it's a complex one that explicitly rejects averages. It's not what people typically mean when they refer to "ROI".
"So, maybe that's still an ROI estimate, but it's a complex one that explicitly rejects averages. It's not what people typically mean when they refer to "ROI"."
I still say you've got a seriously impoverished view of "ROI". You pretty much never just get to look it up on a table, you certainly never get a guarantee, and you are absolutely supposed to consider everything in a ROI calculation... if you value personal satisfaction than be sure to factor that in.
One of the baseline facts of modern economics is the highly variable nature of individual preferences. Microeconomics 101 is literally an exploration of this fact, and all economics cease to function if you assume all agents in the system have the same preferences. The idea that it's all about money is a slanderous strawman, not how it actually works. And so with ROI... if you are not considering your personal happiness with a ROI investment in your degree, that's your fault for failing to do a proper ROI investment, not basic economic's fault for not taking your personal preferences into account.
If this sounds wrong to you, I'd suggest digging in and learning some microecon 101, because it sounds like you've got a strawman understanding right now. If you're a programmer it's easy enough to pick up by just reading about it.
I'm referring fairly specifically of the context of the parent comment. These are the parts that I was replying to:
>can you pay off a 6 figure debt to pursue a comparative literature degree?
>If you pick a program that has high salary/demand, you will have better ROI. Even if you spend lots of money on it, you will be able to pay it back. That is ROI. (Not just cost by itself)
>Here is a list of top ranking salary schools. They are almost all dominated by Ivies, tech/engineering (and interestingly, Military) schools.
I disagree with the approach of the first two quoted parts, and what the third comment is alluding to.
> One of the baseline facts of modern economics is the highly variable nature of individual preferences. Microeconomics 101 is literally an exploration of this fact
Erm, I studied econ in college (concentrating in micro econ and game theory), this certainly is not what micro econ 101 is about. One of the key assumptions of classic micro economics is simplifying people into homogenous representative agents (usually, perfectly rational agents who seek to maximize gains and minimize costs). The field of properly modeling individual preferences in micro economics is a relatively recent development. Economics 101 is where the simplistic definition of ROI arises, e.g. http://econwiki.wikidot.com/roi. And yes, a lot (probably most) people still mean this when they refer to ROI.
"The Flatiron School started two years ago and teaches students, both high school grads and college grads, how to become software engineers in a twelve week course that costs $15,000."
Wow. It's a good thing that whole "software engineering" thing isn't a profession or anything. I wish this had been available 25 years ago, before I wasted 4 years and roughly 9000 hours (4 years, 2 semesters per year, 15 weeks per semester, 15 class hours per week = 1800 class hours; by their class+lab+deployment numbers, a 5/1 ratio) getting a bachelors.
Plus, even if I had decided to stick to the BA, on their schedule I could have packed all that into a little over 2 years, instead of 4.
Like the old saying goes, if you want to burn down a house, you don't burn it down from the outside, you light the fire inside. An individual who has just invested thousands of dollars in their own self-education and (most likely) heard the negativity and 'you're crazy' from their family and peer group has a lot to prove. Their house is on fire. Your ability to learn has nothing to do with your age - it has more to do with your mindset and your ability to proactively deal with frustration and failure. And let go of your ego (code reviews).
I know programmers with 15+ years of experience who are doing the same thing year after year. After a while, the years of experience for that type of programmer are meaningless. Also, either the company or their peer group restricts their ability to learn new things. Many of them with families and (now) college tuition bills to pay don't mind this restriction on learning. There are many more 9-5 programmers than the people who live and breathe on the bleeding edge.
The reality is that while there are only a couple superb programmers at the level of Linus Torvalds, Bill Joy, Thomas Knoll in the world's history - there are many orders of magnitude more programmers who might not change the world but help keep the increasingly technological world running. There is room for many more programmers and we don't need to restrict the industry to the narrow intake funnel of 4 year CS degree.
It’s been an interesting comparison. However, there are programming Bootcamps that are free of charge, like http://pilot.co/bootcamp (disclaimer: I work for Pilot). The ROI comparison is quite interesting in that case… :)
I thought the comparison of graduation rates between the Flatiron school and universities was very misleading. Most stste systems have open enrollment in community colleges and some four year colleges with low, low standards. The truth is that there are tons of private and public universities that will let anyone in who can fill out a student loan application.
I don't know anything about the Flatiron school but I discussed App Academy with a graduate of the first New York class. They made it sound like the excellent results were all selection effect rather than treatment. A lead mentor who was never there and two assistants who had finished the previous San Francisco cycle. Some of the people doing the course had more programming experience than these teachers. It is very likely that the graduation rate reflects selectivity as much as good teaching and curriculum.
Besides that, I don't get this:
"Part of it is that the cost of delivering that education are very reasonable."
He then preaches about how higher ed's cost structures are terrible. But when you look at the cost per week of education, Flatiron is over 3x higher than a 4-year college. That doesn't suggest to me that higher ed's costs structures are all that ridiculous compared to Flatiron.
Overall, I get the impression that Flatiron is providing a valuable educational program to its students. However, if you really want to impress me or inspire others, show me how this model can be successfully applied to careers outside of software engineering.
It would seem a better argument would be made for time spent rather than dollars spent.