Be True To Your CS School: LinkedIn Ranks US Schools For Job-Seeking Programmers
theodp writes "The Motley Fool reports that the Data Scientists at LinkedIn have been playing with their Big Data, ranking schools based on how successful recent grads have been at landing desirable software development jobs. Here's their Top 25: CMU, Caltech, Cornell, MIT, Princeton, Berkeley, Univ. of Washington, Duke, Michigan, Stanford, UCLA, Illinois, UT Austin, Brown, UCSD, Harvard, Rice, Penn, Univ. of Arizona, Harvey Mudd, UT Dallas, San Jose State, USC, Washington University, RIT. There's also a shorter list for the best schools for software developers at startups, which draws a dozen schools from the previously mentioned schools, and adds Columbia, Univ. of Virginia, and Univ. of Maryland College Park. If you're in a position to actually hire new graduates, how much do you care about applicants' alma maters?
In theory, schools can act as a crap filter for workers.
"The difference between theory and practice is greater in practice than in theory."--some C++ Users Journal article
Get thee glass eyes, and, like a scurvy politician, seem to see things thou dost not.--King Lear
A huge number of software development jobs don't require a CS degree, including many highly paid positions. In fact, having a CS degree may reduce the odds of being hired for some positions. It seems the trend of misunderstanding the term "computer science" hasn't lost any momentum.
Write failed: Broken pipe
Some industries tend to hire only from "top" schools because it's a marker for social class, background, or other characteristic desired in the industry.
If your business is looking for a "connection" back to some specific person at that school, then the new grad might serve as a conduit.
However, for straight up technical work, I'm more interested in what experience the new grad has. Are they a specialist in a narrow field?(great if I need that field, not so great otherwise) Does their work show that they are adaptable? (the project you get hired for is often not the project you'll be working on next year) Do they have communication skills, both written and spoken? (these days, fewer jobs are "head down in a cubicle grinding out code", and you ALWAYS need to be able describe what it is you're doing) Does the person have a big picture understanding of how what they do fits into the overall scheme of things? (or, are they a "I process data I receive according to the algorithm I receive and produce data, and I do not care where it comes from or where it goes") I'm always interested to know whether they have independent ideas. Did they just do the homework and assigned projects, or did they come up with something interesting on their own. What other projects have they done, and in what fields. Are they a "everything I do and breathe is in one language because, truly it is the one true language", or do they dabble in other areas.
Are they doing what they do because the like doing it, or because some sort of analysis said "this is the most profitable investment of your time" or "my parents said I should do this" or "I have analyzed it, and Stanford provided the best ROI even though I hate California"
I'm surprised Georgia Tech isn't on the list. The CS school there usually ties UT Austin to rank in the top 10 CS schools.
In our company I am responsible for deciding from a technical standpoint if we give applicants a chance.
In my opinion the school doesn't matter. Having a degree just proves that you can work to reach a set goal.
What counts is what you learned on your own beyond what is taught in school.
If your coding skills are restricted to what schools teach these days, you are useless.
Though there tends to be an irrational tribal loyalty to one's alma mater, hiring properly should primarily consider how the individual's talents (& weaknesses) fit the opening and the firm.
Happiness in intelligent people is the rarest thing I know.
Ernest Hemingway
"f you're in a position to actually hire new graduates, how much do you care about applicants' alma maters?
Let's put this another way. If that is a priority, I doubt I'm going to be interested in working for them.
Clearly they're focusing on the wrong thing, usually brought on by elitism.
Skills are what is important after college, and I'm not talking about how you earned your way into the Kegga-Chugga-Puk frat house.
Most success with a school name depends on how many jobs are near your school. Once you get too far away, the name doesn't have as much clout.
is basically "if a lot of people leave companies for Company A then A is a desirable place to work; what are the most prevalent schools that A's employees attended." So basically if your goal is to work at A you have a statistically better chance of landing a job if you work at one of the top schools in its list. That says nothing about the quality of the school nor that their grads do any better in terms of percent employees or starting salaries than other schools; nor does it seem to address the experience level of those hired. I also think the results are biased to companies that hire a lot of people each year so you might expect more people to go to them, thus increasing their desirability in Linkedin's model; yet a small company that hires very selectively and is a place many graduates want to work would not impact the school rankings.
As for the question of the impact of an alma mater on hiring, having a highly regarded school on a resume would at least get it a second look; as would having my alma mater on the resume. However, I'd hire the top person at a "lesser" school over someone at the bottom of a more prestigious one. The former shows they have a work ethic and keep their eyes on the prize while they are in college; the latter comes across as someone who just gets by.
As for the "person at the bottom may just be so smart they are bored to tears by their classes and thus basically ignored them while doing something really brilliant" argument; fine, but most work at many companies is routine and boring at some point but you still need it done well. If i really need someone very good at some specific then I can find them; what I don't need is someone who checks out overtime they get asked to d something they find uninteresting.
I'm a consultant - I convert gibberish into cash-flow.
Is it on purpose that there are only American schools? I see nothing from Europe and Japan, for example. It seems terribly unbalanced.
Next company in USA, to justify corruption type behavior. Statistics like this are useless. It only encourages corruption, "white boys club" and "I scratch your back, you scratch mine"
No wonder there are so many losers working as presidents, directors and in other executive positions. Like old saying "who you know and who you blow"
This is not statistics done on those doing CS degrees, and their employment outcomes.
It's statistics done on linkedin members who have done CS degrees, and their employment outcomes.
Perhaps there is a substantial slice of CS grads who don't do linkedin, and that population is not representative of the sampled one.
A candidates school(s) definitely come into my hiring considerations. Especially as a tie breaker or when their is little other information to go on.
This is not because I think the top schools teach you so much more than other schools, The big difference is in who gets accepted in the first place.
Top schools screening process are reasonably correlated with qualities I look for in a candidate and therefor are valuable input to my hiring decision.
In the early days of Amazon, they didn't teach PERL, Linux, Apache at schools. We just picked up books and did it. True learning the fundamentals is good but if you have the desire, you will teach yourself much faster than a University and you will do it on CURRENT skills rather than ones that are 5 years old.
I still believe in the University system but I believe it needs to severely be rethought for continuing education rather than a 2 yr, 4 yr, 6yr etc system and include apprenticeships as well as creditid continued training as part of its curriculum much like lawyers and doctors do.
Pedigree comes up a lot in professional sports. Players drafted in the first two rounds of the NFL draft are almost never cut in training camp the first year, even if they perform poorly. Even years later, when they are traded or signed as a free agent, the round in which they were drafted (if any) comes up early in the discussion.
I'm sure those schools do have really good programs, and I know my school had a mediocre computer science department (graduated about a year ago), but what I've noticed really separates good and bad developers is their drive to continue learning and making things because they enjoy it. If people like programming, they'll be really damn good at it, no matter where they went to school. Hell, I would say some of the best developers I've met either didn't go to school or majored in unrelated fields and just learned at work (mind you this was back in the 90s/early 2000s when they did it) and their free time.
A serious study would fold in far more information, which ironically is accessible.
It would be interesting if I could enter my transcripts for college as well as grades which could be used for "future predictions". Even things like certificates and "specific job skills" could be added to really drill down to estimation. Moreover, there are "test companies" which could evaluate one's knowledge of Java or C#, which could be tied into a knowledge management and evaluation system. The problem is to get enough statistics entered as well as insuring that the data is correct. The other issue is to make sure that one is not comparing apples to oranges. An intro CS course at MIT is not the same as one at Harvard Extension School taught by U. Lowell's finest who could not get a real day job and have time "to teach".
A simple opinion based on personal observations having passed via EE, Math and CS degrees and heaving thought core CS (math) subjects. CS degree is not something I would recommend for someone to get/invest into. It is some kind of morph between (extreamly soft) EE and (soft) Math degrees. If people are interested in ether EE,Physics or Math they should go for it, for the sake of loving the subject and experienencing the hard problems they model from the real world. If for the sake of carier safety they need to learn the CRAFT of coding, so be it, for It is highly likely that at the end 50% - 60% of these people will work in software development field anyway, But their experience and aquired knowledge of solving problems will be more valuable in the software development field when NOT coming form, what is considered, CS background.
I interview almost 50 new grads a years, about half are CMU grads. CMU grads can pass a technical interview about 50% of the time. When interviewing students from other schools the I am looking for the rare individual who has the truly excelles, they exist but it is maybe 1 in 10. I only have so much time to interview so I would prefer to spend it with students who can actually answer hard technical questions. CMU gets over 10,000 applications for the 125 slots in its undergrad CS class. Why would I spend my time trying to find the best and the brightest when they have done so much of the effort for me?
What makes a job "desirable"? I've never had any desire to move to the most crowded, highest cost of living cities in America. I could probably make more money, but I have different criteria.
The whole "talent shortage" thing would be a lot easier to believe if a small number of companies weren't poaching a small number of employees in a small geographic area where not too many people want to live. The talent pool would be a lot wider somewhere besides the DC metro area and San Francisco/Silicon Valley, places that attract only people who want to live in them.
Poorly designed projects provide more job security and require more labor to maintain. They are managed by people who don't understand that their projects are sinking or do understand it, but don't understand why that's happening. And, of course, they require more employees. The people who manage them select people who are willing to suffer the pain of being inefficient because those doing the selection don't understand the benefits of increased efficiency. The result is that more jobs is not an indication of a better education. More jobs going to graduates of these schools can very well be an indication of their alumni having been educated in how things were done before better methods were invented.
Any guest worker system is indistinguishable from indentured servitude.
The conflict in modern programming is between code monkeys and math brains. Both are dismissive of the other. The code monkeys think the math brains overcomplicate things. The math brains think the code monkeys don't understand the problems they are solving.
Any guest worker system is indistinguishable from indentured servitude.
I have had interviews where the questions asked were not in my forte. I studied math, but worked as programmer for most of my life. At my university, I remember seeing certain classes taught for different audiences. Numerical analysis taught for CS majors was completely different than that taught for math majors. CS people devoted a lot of time to computational complexity formulations for various sort algorithms and approximations to those. Math numerical analysis chose different problems to study. Even year to year, the math problems studied varied based on the professor.
Take differential equations. DE for engineers emphasized recognizing and memorizing certain route formulations for solving "standard classical problems." DE for mathematicians involved heavy theorem proving and esoteric questions like "existence solutions". Engineers would laugh, saying why the hell would we need to know that? Well, the answer is simple. Most of the easy differential equations that "you can solve" can be solved by looking it up in a book or these days, via Google. However, in the real world, you will most likely encounter fresh ill-posed "one of a kind" DEs. Another example happened when an engineer asked me help with solving sets of DE when the Space Shuttle's robotic arm hit an object. The way that we formulated the solution then was completely wrong. To deal with the impulse event, we switched from a high order RK method so a lower order approximation after impact.
When I took my GREs in math, I bombed my DEs part. Well, the reason is that all of the questions asked focused on the Advanced Nonsense taught to EEs and not on stuff which I was taught.
Years later, I had the exact same problem during an interview with Microsoft. I didn't do well because the sort of questions asked came from the CS school rather than the math school. I don't mind that, but I resented him because he just thought that the interviewer dismissed me as an idiot or fraud because I did not know the answers to his questions. Had he said, "Well, you just MIGHT know your shit, but it is not the shit that we do here, and I do not have a way of evaluating your skill set," I would have accepted that compromise.
If you're in a position to actually hire new graduates, how much do you care about applicants' alma maters?
depends on the school!
some schools matter, and some do not....and some that matter are a benefit, and some are a drawback!
if an applicant came from a program that distinguishes itself among others by *requiring all grads to make a capstone project* for example...that could be a point in the applicants favor over others who did not do a capstone
Thank you Dave Raggett
I dropped out of college, moved out of state and in the course of 15 months, tripled my salary...I make more than any of my friends...so where do i fall on this chart?
and really - why should I go back and finish? Whats the point, more debt??????
isn't it on this list largely due to its proximity to Silicon Valley? You'd think that the number of applications to work at tech companies in the valley coming from SJ State would be off the charts to begin with due to it being in the middle of the valley... I'm sure Georgia State has a reasonable CS program too, but few if any applications from there would be going to companies in Silicon Valley. Does that make SJ State a meaningful CS job target or just a beneficiary of location?
Though not a perfect measure by any means, I think it would be more interesting to see the CS job acceptance rates coming out these schools and the average starting salary for each.
However, it is highly relevant that GED holders and/or high school or college graduates with degrees completely unrelated to computer science tend to be better programmers.
Where do you get that from?
I have indeed met a lot of great programmers that did not have CS degrees. But I've met even MORE programmers that were not great or even coming close to it, that did not have CS degrees...
I would say that overall a CS major has a base level of competence above non-CS holders. But I've never seen a study or in decades of practical experience seen anything to back up the claim you are making...
For what it's worth I have both a GED *and* a CS degree.
"There is more worth loving than we have strength to love." - Brian Jay Stanley
I frequently do both phone-screens and on-site interviews of recent grads, and I pay almost no attention to the candidate's school. What matters is how he or she does on the interview. Blasphemy, I know :-)
A simple opinion based on personal observations having passed via EE, Math and CS degrees and heaving thought core CS (math) subjects.
But clearly not a grammar class........
"First they came for the slanderers and i said nothing."
Does linked-in do business in those markets, or are you just another self-centered selfish foreigner coming to an American website expecting the red carpet of internationalism to be laid bare before you?
I presume it's just an American list. Waterloo isn't there either, and it's certainly on a hiring par with the 'mericun universities named.
ah selecting people like me you mean :-)
I've been hiring both interns and recent grads from engineering and programming positions since 1981 when I first hired a dozen interns out of Drexel.
While MIT and Rensselaer Polytechnic Institute interns and grads have impressed me as hands-down the most brilliant, Drexel students are usually the most prepared for the challenges of every-day work life with Rowan University (formerly Glassboro State) coming in somewhat behind Drexel. (Based on my experiences, Drexel's 5-year program that includes 4 six-month internships should be adopted EVERYWHERE).
Mind you, I usually see applicants only from North Eastern colleges, but over the years it means I've probably hired more than 200 of these individuals.
[Unfortunately, I've never even gotten a resume from anyone from my alma mater, the University of Bridgeport, in response to a want-ad, so I can't say good or ill of UB products.]
Live Long and Prosper - Thanks Leonard. You are missed.
... to guide us throughout the turmoil of our adolescence. The slashdot crowd (wrong) assumption that as anyone can successfully self-teach himself to a successful career. That is totally wrong. It turns out that the majority of population cannot absorb new ideas and new material in a library setting, one-on-one with a textbook. Most of us need human interaction. That is just the way we are. The best schools happen to employ the most successful teacher who by their skillful enthusiasm rekindle you interest to learn time and again with every lecture so that you can sustain that interest for the first 10,000 hours of endeavor. So, the bottom line, we need more good schools and more good teachers. Youtube, or stack overflow, or and other virtual method of learning would not work for majority of people.
EE deals with hardware. Math deals with abstractions, structures, and all that. CS is a lot more than these two IMO. You don't learn CS theory, algorithms, the coding paradigms, or software engineering principles in either Math or EE fields. I used to work as a sysadmin in a leading math department and I saw no indication that the mathematicians are specially more adept at coding or even using computers. The ones who were experts in computing and coding actually chose to do so.