The 2017 Stack Overflow Developer Survey is Now Live
We’re back for another edition of the Stack Overflow Developer Survey, the largest and most comprehensive survey of software developers on earth. Every year, we include questions about your favorite technologies, coding habits, and work preferences, as well as how you learn, share, and level up as a developer.
Want to see how you stack up against the world’s developers? Take the survey now.
We’ll publish initial results for you to peruse in March. A few weeks after that, you’ll be able to download and analyze anonymized results of the survey under the Open Database License (ODbL). We look forward to seeing what you find!
Why We Do This
We estimate that responses to the 2016 survey cost 14.6 years of total developer time. That’s a lot of missed opportunities to squash bugs and ship product. So why do we do this? And more importantly, how do we pay developers back for their time spent?
It starts with our shared, company-wide mission: Everything we do at Stack Overflow is done to make developers’ lives better. Whether it’s through Q&A, Documentation, or finding developers better jobs, we’re single-mindedly focused on doing what’s best for the developer community, and how we can promote developers’ interests. The first step to doing this well requires knowing in detail what different developers want, what their pain points are, and how they want to share and improve their skills.
Developers are generally misunderstood by non-technical managers, recruiters, and policy makers. This has led to lousy hiring practices, poor environments for development (open office concept, anyone?), and misallocation of resources. Having reliable, comprehensive survey results we can point to helps us educate companies. It also empowers developers who need more data to support their cause, whether it’s in asking for a raise to match industry standards, or because they’re pushing for a change with internal development or HR practices.
We also of course use the survey to improve our own policies at Stack Overflow, as well as our products for developers. We want to know if what we’re doing is working, and what you want to see improved.
See How You Stack Up
Last year, we fielded responses from over 50,000 developers in 173 countries. We’re looking to break our record again this year, and we need your help. Take the survey now.
Got questions of your own or feedback? Post in the comments below or on Meta.
39 Comments
For some reason, sometimes when I press the ‘Next’ button, I go back to the first page of the survey instead of the next question. This is a bit annoying because I have to start all over.
Update: Finally managed to get through without this happening.
It’s quite frustrating that there is no back button on the survey.
I came here to post this. I realized too late that the “worked with in the last year” and “want to work with in the next year” didn’t have to be either/or. Really wish I could have gone back and fixed it.
So,
13.[…] When you have completed your studies, what annual salary do you expect to earn in your first job after graduation? […]
Which currency? (would have answered in dollars, but the previous question asked about currency I use day-to-day, which definitely isn’t $)
I said it was too long because it said I was only 55% complete at that point.
Even worse for me. It said 44% when I completed the survey.
In case anyone is interested in the results of last year’s survey, here are the survey results from 2016.
Thanks for finally adding Agriculture to the list of industries!
I cracked up at question 11: “How do you pronounce ‘GIF’?”
I think it’s about which language you learned it. For me, it’s a hard G (like gift).
It’s not as bad as previous years, but there’s still a strong bias towards employees. Some questions don’t have suitable options for entrepreneurs or freelancers to answer.
Also, it’s funny that the question “why haven’t you made a CV/Developer story” doesn’t have an option “I’m not looking for a new job.”
Done! As per my level best.
I realize you guys didn’t code the survey platform yourselves, but wow. Around question 10 I accidentally clicked next before I’d filled it out, and there was no way to go back; even going back to the start page and clicking “start survey” again took me to question 10, and I was too lazy to start clearing cookies just to fix it. Then around question 55 I clicked “next” and ended up back at question 1; fortunately it remembers all my answers, so I’m just spamming “next” until I get back where I was. At least I can do the skipped question along the way now I guess
I’m certainly not a professional but 90% of the work I do to earn my bread is software development. So I’m stuck on Q1.
What a narrow-minded question!
In what sense do you consider yourself “certainly not a professional,” then?
I am not paid to write software. I’m paid for something else. But most of my time is spent writing software that I use in my job. I hope my answer to carrie helps explain this difference.
Isn’t the definition of a professional a person engaged in a specified activity as one’s main paid occupation rather than as a pastime? So, if 90% of your work is paid work, you sir, are a professional 🙂
A professional provides skilled services (in this case, the development of software) to an employer who pays the professional in return for those services provided. Moreover, a professional, as opposed to a skilled or unskilled worker, is trusted when making decisions to put the interests of the employer ahead of other interests including her own, the employer’s clients, the common good, etc.
I don’t do any of that. I do not provide software development to anyone. I am not employed by anyone. I am not paid for developing software. I write software that my business uses internally.
Imagine you own a business that sells a niche financial accounting and audit service and your unique value derives from proprietary software that you developed yourself. With respect to its clients, the business provides professional service but it is not software development. Nor is “software developer” your profession.
Another example might be an independent security analyst who wrote all their own testing tools. That person wrote a lot of software but not professionally.
Maybe some will dismiss the distinction with “Bah! Semantics!” Fair enough, but I think meanings can be important. Jackie writes software for someone else who makes money with it. Pat makes money by selling something useful that involves software he or she wrote. Pat’s position in the world is different from Jackie’s in ways that I think are important.
I was with you until you said “my business” and “I write software that my business uses internally.” If your coworkers use your software, then *your coworkers are your clients*. This is a fairly common situation.
But you also state “I am not employed by anyone.” So what do you mean by “my business”? Are you self-employed in a non-software industry, simply using your software for your own personal benefit to improve your own ability to do your job? (My realtor does something similar.) If so, then okay, I suppose I see why you would not consider yourself a professional software developer.
> Are you self-employed in a non-software industry, simply using your
software for your own personal benefit to improve your own ability to do
your job?
It a partnership but otherwise yes, that’s pretty much it.
The word
professional describes a transactional relationship in which the
professional renders a service to an employer in return for payment. My
co-workers may be clients in some sense but not in this sense. They are
not an employer who pays me to write software for them. Some of our staff are professional software developers but I’ve never been employed like that. We have a brilliant, very young developer just now who I’m trying to steer away from software as profession for basically the same reason Zed A Shaw’s “Advice from an Old Programmer”.
Believe me, I get that surveys are hard to build – please take this for the gentile feedback it is intended to be.
Somewhere between “Some college or university study, without receiving a bachelor’s degree” and “Bachelor’s Degree” is an Associate’s Degree. The local community college is turning out a fair number of people that do get *a* degree, it’s just not one of the ones on the list – and it’s different than others I know, who just completed “Some college or university study, without receiving a degree”…
Also, you might want to forward some of the feedback to the company that hosts the survey software as I too couldn’t move back.
For me it was when they asked for an email address, but wouldn’t take a tagged (e.g. me+tag@example.com) email address. I wasn’t going to give one without the tag, it wouldn’t let me move forward with out providing one, and I couldn’t move back to change the option that prompted for the email address in the first place.
Thank you guys for the work you do, here’s to making 2017 better than 2016. 🙂
Damn, too bad you can’t go back. I thought of the “which technologies did you work with over the last year and which technologies do you want to work with the next year” questions as somewhat exclusive… I picked JavaScript as “worked with over the last year” but of course I’ll continue to use it, so I should’ve picked both…
I hated the granularity on the questions about how long. They increment in years to 19, then everything else is lumped together. Why not do two year groupings so that I can put in that I first learned to program 40 years ago and have been working in programming for 31 years?
I understood that as an artifact of biases of the questions’ authors.
Irked me as well, but what can you do? And I work for SO.
Thanks for the survey. I have some feedback to (hopefully) help make the next one even better.
Question #32 lacked what I consider the most essential criteria for hiring, namely *what* they have done and their *ability/aptitude*.
Question #52 was deficient in choices, since “I have no idea” was not a choice for “They’re only in it for the money” and “I hardly noticed there were ads” was not a choice for whether ads were distracting.
Question #59 seems incomplete in that I found some questions simply too limiting.
For all the people who, like me, lament about no back button. If you don’t mind spending the time, close out of the survey early, open it again in an incognito/private window and take it again. But yeah, it’s 2017.. we should have a back button on surveys.
No preferred desktop OS question? Noooooooooooooooooooooooooooooo!
Spaces or tabs? Both.
On Reddit, people are saying that some questions are poorly worded or missing choices they’d like to have, and that there are too many questions.
Best fictional programmer: BPS: Battle Programmer Shirase. “His signature finishing move is the Double Compile.”
How could I forget about that guy…
Way too long … didn’t finish it
How many questions are there? I know it’s long, but *how* long??
EDIT: The answer is 67. There are 67 questions! Although there may be slightly more or less depending on the answers given.
Somewhere between “Some college or university study, without receiving a bachelor’s degree” and “Bachelor’s Degree” is an Associate’s Degree. The local community college is turning out a fair number of people that do get *a* degree, it’s just not one of the ones on the list – and it’s different than others I know, who just completed “Some college or university study, without receiving a degree”…
antalya transfer
For me it was when they asked for an email address, but wouldn’t take a tagged (e.g. me+tag@example.com) email address. I wasn’t going to give one without the tag, it wouldn’t let me move forward with out providing one, and I couldn’t move back to change the option that prompted for the email address in the first place.
Sorry to rant, but did I misread or you have forgot to include Delphi/Pascal this year? (which is TIOBE #11 atm)
Great survey. Not too long – covered all the bases.