when you catch a candidate lying in an interview

by Alison Green on November 5, 2012

A reader writes:

I work in software development. As a part of our interview process, I ask a series of technical questions. In the past 5 years of interviewing, I’ve managed to catch two applicants bold-face lying/cheating (by looking up the answers online) during a phone interview. The first time, I didn’t say anything to the applicant and finished the interview like normal. The second time, during a short coding test we conduct with a little internet application, the applicant looked up the answer online and copy/pasted someone else’s code. It was pretty obvious he didn’t produce it and, before I could call him out on it, his phone died. I discussed it with some colleagues and they suggested sending him the link to the site he copied the code from to keep him from trying to bother to reschedule an interview.

I guess my question is, what do you or others do when you catch someone lying during an interview? Do you call them out on it? Or do you let it, and the candidate, pass?

There are two schools of thought on this, and both are legitimate:

1. Don’t call the person out, but just remove them from the running. The thinking here is, why bother? You’re not going to hire the person, and it’s not your job to explain why or teach them a lesson.

2. Raise it. Not in an “aha, gotcha!” tone, but simply be direct about why you’re questioning their answer. For instance, in the situation you described, you could say in a neutral tone, “Hmmm, it looks like that came from XYZ.com…?” When I’ve caught people plagiarizing on the written exercises I give when hiring, I’ve generally pointed it out by saying something like, “This answer appears to be taken word-for-word from XYZ online. As a result, I won’t be able to consider your candidacy further.”

I generally do #2 rather than #1 because I like to err on the side of being transparent, but either approach is fine. (Although if you’re going with #2, you’ve got to keep it matter-of-fact and not punitive, since punitive veers too close to unprofessional.)

By the way, I’m assuming here that you know for sure that the person cheated. If you only suspect it but aren’t sure, then I think the right thing to do is to ask about it before simply taking the person out of the running, in case you’re wrong. In that case, I’d say something like, “This is awkward, but your answer seems to be based on XYZ. Can you tell me more about how you reached that answer?”  Or, for a different type of possible lie, “This is awkward, but no one at ABC is able to confirm you worked there.” Or so forth. Occasionally something looks like it’s a lie but there’s an innocent explanation, so if the picture is hazy, I’d ask about it if you can.

You may also like:
{ 115 comments }

{ 115 comments… read them below }

  1. Mike*

    For programming questions I like to take questions that have answers easily found on the internet and then modify them in subtle ways. For example, taking the fizz buzz question and changing what the output should be.

  2. Neeta*

    I remember something like this happening to a colleague of mine who was conduction Javascript/CSS interviews via Skype. Voice call, not video.

    Basically, there were candidates who answered correctly, but took a suspiciously long time to do so.
    Eg:
    Interviewer: What does X mean?
    Candidate: um… long pause… answer

    He didn’t say anything to the candidate, but just told HR that the person cheated. They probably just quietly took him out of the running.
    Basically I’d do 1) in your case. Sure giving the candidate a lecture might sound tempting, but then they’ll probably try to come up with excuses and the like… In other words: why bother?

    1. Ask a Manager* Post author

      Yeah, in a case like that where you can’t know conclusively, but you also can’t really say, “Are you looking these answers up?” because the person might just be incredibly slow, I’d just move on.

    2. Brad*

      You could probably ask a more open-ended technical question about whatever X is (assuming you’re not testing their vocabulary). “If I asked you to build Y, how would you use X?” The answer could be not to use X at all, but then you can ask why to see if they grasp the concept. I’ve done it a few times in the past and each time gotten some good answers.

  3. Anonymous*

    For programming and technical writing tasks, sometimes the applicant has already created content that’s pretty much answered the question and posted it online. Copying and pasting from one’s previous work is shady, and I avoid it, but keep in mind that candidates could be basing their answers on their own old work. The best thing to do, of course, would be to say “I answered X on this date at this site; do you want to take a look?” But this might not always be possible when the work is posted somewhere anonymously or under a pseudonym. Not sure if that’s the case here; just putting it out there.

      1. Anonymous*

        If you produced the work for a conference or official event, there’s an expectation that it’s original. If you’re expected to come up with work on the spot, or in a short timeframe, and deliver something that you had much longer to polish, then you’re giving a potential employer the wrong impression of your work pace/skills.

        1. Mike C.*

          In what sort of context? If asked to present at a conference, one is usually speaking/presenting on work which was previously done, or building on top of it. Both require a ton of using previous work.

        2. Vicki*

          Nothing I( do in a job is ever “on the spot”. I look things up all the time. I keep an online notebook of code I’ve written. I look up examples. I have _used_ examples.

          That’s how I work.

          Why should an interview require me to pretend to have memorized things that I haven’t memorized )_because I know I can look them up_. Why should I be on the spot in an interview when I won’t be on the spot in my job.

          If you don;t want something to be an open book test, don;t run it over the phone.

      2. BCW*

        I’m not in tech, but I assume its similar to a lot of Universities. I know at my school if I used a paper I wrote for one class and submitted it for another, it was still considered plagiarism. Now to many, including myself, it is ridiculous that you can plagiarize yourself, but the logic is that you are supposed to produce an original work.

        1. Esra*

          That’s not really the same as tech at all. It would be strange to come up with an entirely new way of doing something you’ve already done. Unless you can clean up the code in some way, you really shouldn’t be doing the same thing in different ways.

        2. Another Ellie*

          It’s not really ridiculous, it’s related to academic fairness. The reasoning behind that is basically that you’re defrauding classmates and deceiving the professor in the realm of grading if you don’t state that it’s your work beforehand and get it cleared. For example, say everybody in the class has to spend 15 hours researching and writing a short paper, but you have to spend just 2 hours fixing some things that a previous professor identified as needing improvement. Then you get an A and set the standard for the class with your excellent paper, but in fact other students did much more work than you at the same level, only they were at a severe disadvantage because they were starting from scratch and didn’t have the benefit of feedback. The professor should have known that you didn’t do the equivalent amount of work so that they could factor that into grading, but instead you lied and maybe a couple of students deserving of As got A-s, then they knocked a couple of kids down to B+, etc. (and, yes, your professors grade you against other students in the class. It’s inevitable in the grading process.)

          This continues to be an issue in professional academia. For example, there’s a particular professor in my field who was recently exposed as having published the same article 3 times under different titles. It wouldn’t have been a problem if she had just noted in each publication that it was a re-print of previous work and then made that clear on her CV. Instead, she made it look like she had published three times the amount of work as she had actually done. Since academics are partially judged by volume of output, the inflation would have affected the way that hiring and tenure committees looked at her applications. She may have been hired/promoted over other people who were just as qualified but more honest. It’s not a firing offense, but everybody definitely thinks about the fact that she did that when they hear her name.

          1. danr*

            I saw many examples of this when I was an indexer for a library publisher. Sometimes the article would be tweaked a bit, but many times it was identical. When an article looked familiar, we would check the author for previous entries… and then give the new article the exact same subject headings. It’s a lot easier to do this with online databases. And probably a lot easier to catch folks at it.

          2. BCW*

            I think being paid to get an article published as a professor is very different than being a student and being able to use a paper twice. But its just difference of opinion I guess.

            1. Rana*

              Strictly speaking, you shouldn’t be able to reuse papers, as no professor is going to ask the same question as another professor (or even the same question in a different class). So whatever you had to say about the rhetorical devices in Plato’s Republic, it’s not going to be that helpful when you need to write a paper about the gender dynamics of the California gold rush.

              The reason it’s academically dishonest, though, is not because of this transferability issue. It’s because you’re trying to get twice the credit for half the work, or, to put it another way, you’re trying to have that paper count double what it should.

              So there are two reasons not to reuse old work: one, it means you’ve cheated yourself of the opportunity to learn something new and improve your skills; two, it means you’re stealing extra credit that you did not earn.

              Don’t do it.

              1. Going Anon*

                I took classes through University of Phoenix’s program for associate’s degress, called Axia. I’m not sure whether the instructors write their own courses or if it’s given to them by the college, but I flunked a class once, retook it the next session from the same instructor, and every assignment was EXACTLY the same. And I’m sorry, but I don’t think it’s even POSSIBLE, and it’s certainly not worthwhile to write the exacts same assigned papers from scratch. I took the necessary exams and quizzes without referring to my old coursework, based on my reading and understanding of the material (both from the course before, which was, of course, not necessarily accurate knowledge, since I’d flunked and all).

                But when a paper was assigned that was exactly the same as the paper before it (and we’re not talking creative writing or even a research paper you can choose a new topic for or at least new sources–we’re talking fact-based writing like read a chapter or five and write a reaction paper, or take this case study and write a SWOT analysis or whatever), I’m sorry, but I took my old papers and updated them based on my better understanding the second time around and the feedback I’d received from the instructor. It was all MY work, so I don’t see how doing that can be considered cheating any more than writing a first draft and asking someone to critique it then writing a second draft would be, though of course there was far little first-draft-writing to be done when I took the class the second time around.

                The way I figure it, the time saved writing a first draft is what gave me time to really buckle down and understand the material, and pass the class the second time around.

                I also see it as similar to an assignment for, say, a persuasive essay. I could pick a topic about which I know nothing, do a ton of research, pick an argument, and start persuading. Or I can pick a topic I’m already knowledgeable in and on which I have a strong opinion, and all I have to do is find a few sources that support that strong opinion and write a paper from the gut with a few references thrown in. It’s not cheating–it’s all my own thoughts and feelings, I’m just giving myself a head start.

              2. Hari*

                So there are two reasons not to reuse old work: one, it means you’ve cheated yourself of the opportunity to learn something new and improve your skills;

                I disagree. Researchers themselves often build off of previous work, case studies, and articles. No one starts from scratch every time, but usually every time you are trying to do something a bit different. Like you mentioned no professor is going to ask the same question as another, however they may overlap in areas or have similar characteristics. By using your first paper as a starting point you have more room to grow with your paper as you have already researched the basics and can now spend more time going in depth about the specifics the new professor is really looking for. You are saving yourself time and now have more opportunity to really learn something new instead of spending time re-researching materials you already have. It’s called being resourceful not cheating.

                “two, it means you’re stealing extra credit that you did not earn.”

                You did the original research in the first place? Then you earned it. An example of how that applied to me is my senior seminar class. I took it in the winter although I was graduating in the spring. I wrote a 25 page paper on freedom of speech/expression in all mediums, history to now, court cases, the philosophy behind it, and my personal stance, took me days. My final quarter I took a Poli Sci class where for our final paper was 5 pages and was also on freedom of speech but how our laws contrast with foreign policy. The only real research I had to do was on the contrasting country but do you think I was going to re-research or write anything that had to do with the US policies when I pretty much wrote a mini book on it? Sorry but no lol. Also I cheated no one by doing it cause my original paper was literally 5 times longer and a lot more in depth.

                1. Rana*

                  Actually, neither of you has done what I’m talking about. Both of you re-approached the topic, adjusted your original argument to reflect your new thinking, adapted prior research to new purposes, and turned in (it sounds like) papers that were different in thrust and intent than earlier ones.

                  What I’m talking about is the case where a student changes little beyond the name of the class at the top and the date, and submits identical or near-identical papers. And, yes, believe it or not, some people do do this, and are surprised when those papers are rejected.

                  I hope you can appreciate the difference? And why it would be unfair to people like yourselves to give that sort of lazy duplication the same grade and credit as the hard work you did?

            2. Another Ellie*

              You don’t quite grasp the distinction here. There’s nothing wrong with using a paper, article, whatever, twice. Happens all the time, in fact. The issue is using it twice but not telling anybody beforehand, ie lying about it. Students are allowed to use past work if they clear it with their professor ahead of time. Usually if the professor allows this they will ask the student to do new research in order to expand the paper. This is fair to other students who are preparing fresh work for the class, and it also furthers the goal of assigning papers in the first place, which is causing the student to continue practicing and expanding their skill-set. If the student doesn’t enter into an agreement about re-use of work with the professor beforehand, the student is defrauding the professor and other students by passing their work off as new. This is problematic for the reasons I outlined above. The same issue is really what’s going on with the professors. It’s not the duplication (which happens openly quite often), it’s the lying which then tricks people into evaluating them based on false accomplishments.

            3. Ellie H.*

              No, it’s exactly the same. Honor is honor, regardless of whether or not you are paid for your cheating.

          3. Hari*

            I don’t really see the issue with reusing a paper. Like others have mentioned its highly unlikely especially in a university setting that 2 professors will have the exact same prompt enough where you will get away will a straight copy and paste. I have totally re-used sections or ideas from other papers of my own. I put the proper citations and quotes in. Nearly every single sentence in a research paper not in the hypothesis or conclusion (and even there too) has a citation on it. The research isn’t yours to begin with, the only thing you get credit for is how you arrange the words. If I wrote a paper about the history of Chocolate Teapots in one class and in my new class my professor wants a paper on how Chocolate Teapots came to influence Chocolate Coffee Carafes then theres nothing wrong with referring back to previous research of my own for ideas or if I have a few relevant lines that I could transfer that would fit in my paper. You aren’t cheating anyone because you have already done the research yourself once, no need to do it over again if you don’t have to.

            The professor example isn’t the same because she published the same article 3 different times on the same subject and presented all of them on her CV as different. A professor from one class wont know what you have written for another. It would only be the same if a professor asked for lets say 3 papers on chocolate tea pot factories and you submitted the same paper 3 times.

        3. Mike*

          I know this is late, but I am a recent computer science graduate. It is actually encouraged to reuse your work from previous assignments and classes if you can. In fact, most of my curriculum for many courses were set up so that you would do so. There are several reasons for this.

          One reason is that writing code is not like writing a paper or book. While they are both protected under the same copyright laws, you cannot “cite” code like you can another person’s words. Code is the vehicle through which ideas of systemization are expressed by humans and interpreted by computers. Many different people might have the same idea and express it themselves. This is one area that code and written word are the same. However, if you take the ideas I mentioned (no citing, communicating to a computer, and expressing an idea), there is not a legitimate reason to cite another person’s code. If an author quotes another person’s words to emphasize or reinforce his own new or similar point, that is an expression to other humans–it is a substantive device to say, “Here is someone else saying the same thing.” Computers are not skeptical like humans. They are incredibly dumb and totally reliant on the instructions given to them. It takes what you tell it at face value, so it doesn’t need evidence. The point here is that there is no need to practice citing, so it’s not there.

          Another point is that you will still reuse code when you get out of school, so it’s good to get used to thinking in a manner of “oh, I bet I could either use or modify A to work with B because that would solve the problem I am having with/make it easier to implement/save me time with B.” The saving time part is especially important. Writing code is not like writing a paper. If you already have something that you know works well, you don’t have to spend precious time debugging that particular part.

          The last point, which is closely related to saving time, is that reusing code keeps you from introducing new bugs into your program. Every time you write code, you’re creating bugs. The exceptions to that rule are few and far between and generally the code isn’t that useful. Ask any programmer: if they finish writing a module, it compiles on the first try, and when they test it, it works on the first try, they will be scared. That sounds like a really good thing, but it could be very bad.

          The issue with this that pertains to the question above has more to do with copyright and corporate espionage. Let’s say the interviewee was asked a question that they had code from their current employer they had written to implement a similar problem. Let’s say the interviewee copy/pastes the code he had written to answer the question. That’s a copyright violation. Now let’s say the interviewee’s current employer is in direct competition with the company he is applying to. That could violate federal trade secret law.

          Let’s say that said interviewee had an assignment in college that had him implement something similar, so he uses that. No problem. None. It might subvert an interviewer’s intentions for the question–for example, thinking on your feet–but there’s nothing illegal about it, and the interviewee might not be aware of the interviewer’s intentions. At any rate, in that case, the interviewer should make more clear that the point of the exercise is to see how the candidate thinks on his feet and have him redo the exercise (at least if you ask me).

  4. Joey*

    Alison,
    Your plaigerizing example isn’t asking anything. You’re just taking them out of the running and being transparent about it. So is it fair to say you only ask about it if there’s some doubt?

    1. Ask a Manager* Post author

      Ah, my wording in #2 was confusing. I suppose I should have said “raise it” rather than “ask about it.” You can ask about it — i.e., “it looks like this came from X site….?” with a question in your voice and wait to hear their response. Or you can just explain your findings, like in my second example there.

  5. anon-2*

    Yes – it’s good to screen candidates. But, just remember, there’s an old Chinese proverb = “A fool can ask more questions than any wise man can answer.”

    If, as an interviewing manager, you ask twenty questions — and the interviewee trips up on #18 or #19 … you haven’t “caught” him in anything.

    Think about that. In my younger days I went to one interview where a clown manager did that to me, and two others who applied for the same position. A year later the position was still open, and I was asked to re-interview there.

    Hey, I held my nose and ran outa there once. Not gonna go back….

    2. Josh S*

      It’s not about tripping up on a question as in ‘not being sure of the answer’ or getting it half-wrong due to stress/whatever. It’s about LYING or CHEATING or PLAGIARIZING the answer. I don’t care if I ask you 1000 questions. If you cheat/lie/plagiarize on ANY of them, that brings into question your work ethic, your integrity, and whether I’d ever be able to trust what you tell me. And because of that, I don’t want to work with you.

      A slip-up on a question tells me, “This person is human, and I understand where their weaknesses are.” An incident of lying, cheating, or plagiarism tells me, “This person looks for ways to avoid doing the right thing, especially if it might make him/her look weak/bad.” That’s not a person who gets hired. At least not in my book, anyway.

  6. Mike*

    One thing that confuses me: Since when is doing a search cheating? As a programmer I find the answers to a lot of things by searching. Now, I’m not going to copy someone else’s code and call it my own but I’ve got no problem doing some quick research.

    When putting together a set of technical questions, to be answered at home, I did it with the expectation that the person will use the internet. If the person can’t read the documentation or tutorials online to figure it out then I really don’t want them.

    When I did my recent job searching and was doing a phone interview I had the internet up and doing searches as needed. I didn’t need it to answer the questions but used it to further the discussions. For example, the interviewer mentioned a term that I wasn’t familiar with so I Googled it and read a quick description that allowed me to ask some better questions. When I came in for the in person interview I sat down with the interviewer and wrote some code for him right then and there. The questions were ones that a person with my background should be able to code without references. We then moved onto design questions on the whiteboard where the exact code became less important (we did a lot of pseudocode).

    1. Ask a Manager* Post author

      Doing a search isn’t cheating, and it’s very common to do on the job. The problem here is that the candidate was asked to create code and copied someone else’s instead. It would be similarly problematic if, for instance, you asked a candidate, “Tell me about what you’ve found to be the advantages of Technology A over Technology B” and they looked that up instead of speaking from their own experience. In a work setting, it’s reasonable to look that up — when someone is asking you an interview question that’s clearly designed to suss out your knowledge on something, it’s not.

      1. Anonymous*

        There’s something important that needs to be noted about coding, that I’m not entirely sure whether the OP understands and I wouldn’t expect a non-coder to care about.

        Coding is not the same as writing an essay. In coding, copying can be a good thing, under many circumstances. In contrast, copying is a bad thing in most other fields in most circumstances (like writing blogs).

        A whole lot of people make their living from having other people copy their code. Additionally, there are a ton of people who have explicitly put their code out into the world with a neon orange blinking sign that says, “Please, copy this! No charge!” Coding is more like making a brick building than it is like writing a newspaper article. Some people thus employed make the bricks, while others stack the bricks together to make houses (and some do both).

        There are lots of situations where it would be bloody stupid to write your own code for a specific purpose. In these situations, there’s a proper formalism for giving credit where credit is due that should be followed – and that’s something an employer should look for. However, if the OP came up with a bad programming test that has a posted answer on every wiki on the planet, it’s a bit stupid to get angry at the applicant for using this answer. Frankly, if the OP can’t come up with a simple programming problem that doesn’t have a ready-made answer available on the first page of Google results, then the OP doesn’t know what he’s doing and shouldn’t be interviewing programmers.

        1. Anonymous*

          You do have to be a bit careful even with open source – some licences (notably the GPL) can get you into trouble if you try this sort of thing and aren’t careful (whether or not this is good is a separate debate).

        2. Neeta*

          I think you’re looking at the issue in a much more complex manner, than needed.

          The OP was talking about the interview stage, NOT working on an actual project. Of course you’re not required to reinvent the wheel as a programmer… but you’re expected to be able to code without constantly needing to go to Google for help.

          During a technical interview, you’re looking to see how a certain candidate thinks. You’re not asking for rocket science. So it’s really not unreasonable to expect a candidate to be able to code a simple fizz buzz script without needing to do a Google search for it.

    2. Julie*

      I was going to mention something like this as well. I find that a huge part of my skill set as an executive assistant is being able to troubleshoot based on looking things up on Google. (“How do I do X in Excel?” “Why isn’t Adobe printing properly?” “Why is my mail merge hanging?”) Being able to come to answers quickly and effectively is a skill I pride myself on, even if I’m not super-familiar with the program.

      Now, that said, there’s a limit. I wouldn’t say I’m a programmer when all I can do is download macros, or say I’m familiar with Photoshop when I can barely understand the menu layouts. You need to be able to evaluate your answers and have a framework in which to place them. And copy/pasting something whole-cloth on an application is probably sketchy behavior, no matter how you look at it. But if I was in the office, and someone gave me a task to do and I just happened to find the exact code online that would do it… you better believe I’m using it!

      1. Ask a Manager* Post author

        I agree, but if an interviewer asks you something that’s clearly designed to test your knowledge, not your search skills, then that’s the answer they want, even if you think they should be testing your search skills instead. But you can certainly say, “You know, I don’t know off the top of my head, but in the past when faced with that type of question, I’ve easily found it by XYZ.”

    3. M-C*

      Probably showing my age here, but indeed I also wonder why finding the right answer efficiently on the net isn’t part of what people test for. “Knowing” something, in technology, is a skill with a very short built-in obsolescence. Finding the right answer, as in being able to evaluate what’s out there, is a much better skill in the long term.

      1. Ask a Manager* Post author

        Yes, they should test for that. And perhaps they do in another part of their interview. But that doesn’t make it okay to turn their test of one thing into a test of your search skills, without disclosing that, when that’s not what they asked for.

        1. Mike*

          Without seeing the exact question it is hard to judge what they are asking for. While they might think they are asking for one thing there question could be vague enough to be interpreted another way. This is (IMO) especially true in a phone interview.

          1. Ask a Manager* Post author

            Based on my experience with candidates who do this, I doubt it. I’m really baffled by the responses to this one — candidates do lie and cheat, and the question is about how to handle it when that happens.

            1. Mike C.*

              The issue here is that the use of non-proprietary code is such a common thing that most would never think of it as “cheating” or “lying”.

              1. Elizabeth*

                In a technical interview, though, there’s an implicit understanding that the interviewer wants to see what you can do. So if they ask, “How could you write a program that takes a set of 10,000 customers’ ZIP codes and finds the five most common ones?” they don’t expect a copy-paste from somewhere else.

                1. Vicki*

                  If someone is really stupid enough to ask me that question in an interview, I honestly feel that the best response is: there;’s an easy way to do that, there’s a well-known algorithm, it;s been documented copiously, and I would LOOK IT UP. Because my time (and yours) is too short and you’re not paying me to invent something so trivial off the top of my head.

            2. Anonymous*

              Copying code is the basis for the whole field of programming. It’s not unusual and it is not undesirable. In coding, the guy who decides to spend all day writing a linear algebra library from scratch to see if he can tweak it to go faster is not what you want for 99.97% of programming jobs. The guy who picks out and uses an existing linear algebra library to get his job done quickly is the one you want.

              In management:
              Some managers are realigning the organization’s objectives to improve the ROI for the stakeholders. These guys are like coders who write everything from scratch – they make a lot of work, but they don’t get a lot done.

              Other managers are fixing the business’s problems to keep profits up and keep customers happy. These are the coders who use code other people have already written and tested – they get a lot done, but it doesn’t always make for a good show.

      2. Rob Bird*

        Unless the job requires you to do a lot of searching online, it’s probably not the best way to answer questions. Employers want to make sure you are the best candidate, not Google.

        1. Mike C.*

          That’s a silly thing to say – Google doesn’t work unless you have enough information to ask the right questions and are able to filter large amounts of data efficiently.

    4. Lily*

      Maybe I am just catching the poor plagiarists, but those who I have caught just copy something vaguely similar from the Internet. They can’t figure out how to write it themselves, but they are not able to choose something appropriate, either. Who would want someone who randomly copies and pastes?

  7. Steve Martin*

    I’m actually confused why looking up the answers is a problem, but maybe I just need more context to what the questions and answers are. If something’s already been solved and I know it (and trust the source) I’m not going to try to reinvent the wheel.

    Plagiarism is something to select out based on, but not all finding the answer online is plagiarism. I don’t use a lot of code anymore, but I have access to a bunch of coded solutions written by others that I have permission to use/modify.

    Unless the question/answer specifically makes it clear that use of existing code is disallowed (which to me isn’t clear in the original post) I would dig more into how the candidate arrived at their answer before taking them out of the running.

    1. Ask a Manager* Post author

      The OP says that the candidate copied and pasted someone else’s code, when asked to come up with code himself on the fly. If you’re asked to come up with it yourself, you’re supposed to come up with it yourself. In lots of interviews, it’s very clear from the context that that’s the expectation, and I’m assuming it was the case here.

      1. Anonymous*

        On the job, programmers should re-invent the wheel as little as possible.

        An interview should help determine fit for the job. If the interview question can simply be Googled, then the interviewer isn’t doing his company any service by dinging programmers who are trying to be as efficient with their time as possible.

        I have yet to have a positive experience with a company that used sample code exercises or logic tests in its hiring process. What I mean by that is that I’ve passed them with flying colors, been hired, and then realized doing simple tests that have no business relevance shows an utter lack of knowing how to hire. And that was reflecting by teammates. Fool me once, shame on you. Fool me twice, and I no longer will participate in these.

        If an interviewer cannot think of ways to talk to a programmer to determine how much the programmer knows and doesn’t know, that interviewer isn’t a good technical interviewer. You can try to solve this problem by coming up with problems involving manhole covers or numbers of names in a Manhattan phone directory, but those are just band aids to cover up the root cause in the hiring process.

        1. Ask a Manager* Post author

          That’s your prerogative, but it’s still not okay to copy someone else’s code when asked to produce your own. If you think that’s a silly interview method and want to decline to participate, that’s fine, but that’s not the issue here.

          1. NUM*

            Is there more to the original question than what appears here? It hasn’t been clearly established in your comments so far or by the OP that the OP/interviewer asked for original work.

            1. Ask a Manager* Post author

              No. I’m assuming it based on the context. (And my answer ends by noting that the advice only applies if you’re sure there’s cheating involved, not a misunderstanding or mistake.)

              Hopefully the OP will weigh back in to confirm.

          2. Anonymous*

            In programming, it is explicitly OK to copy an awful lot of programming. Like Android – the operating system that Google wrote, the thing that runs Android phones and many other things. It is free to use, in part or in whole. Apple, their archnemisis, could start running their phones off of it tomorrow if they so desired, and hapless programmers taking employment tests can copy parts of it for job interviews. Legally. Encouraged to do so by the writers themselves. There are caveats that must be observed, but copying it is perfectly fine.

            Telling a code-writer not to use other people’s code is like telling a blogger to write articles only in a language they’ve constructed from scratch themselves. Sure, it might end up being wonderful, and it worked for Tolkien and the Star Trek writers (Klingon). But it takes a long time and isn’t really a sane approach for most projects. I know it’s hard to grasp if you haven’t had your hands in programming, especially if you make your living in a field where copying is always equivalent to plagiarism – but that is exactly how nuts it is.

        2. Josh S*

          There’s a significant difference between gimmicky questions/exercises like crossing a bridge or counting manhole covers, and pseudo-realistic exercises that demonstrate you skill in an area.

          The gimmicks really have no place in an interview, and I consider it a red flag if an employer asks them–it tells me that they don’t know how to screen for what they really want in an employee, the office is probably full of marginally qualified people, and they don’t know how to evaluate or measure success based on results. Alison has said similar things in the past.

          The work related exercises, in contrast, show an interviewer that you’re competent at DOING the tasks (ie executing) and not just good at understanding it in theory. Which points to a functional workplace with a lot of people who know their stuff.

          1. Elizabeth*

            Agreed. It can still be a fair interview question even if someone out there has already answered it. To use a non-coding example, when I interviewed for my job as a teacher I was asked to write a sample report card comment for a fictional child. I could probably have Googled “sample report card” and found someone else’s comment, but that wouldn’t have given the interviewer any information about my own writing skills, or any indication of how I’d do when faced with a problem I couldn’t Google.

        3. Dan*

          I actually agree with this. Our company is looking at modifying its technical interview process a bit. While many programmers can’t fizz-buzz their way out of a paper bag, I brought up that interviewing is a two-way street — candidates should be looking for the right fit as well.

          I said that with five years of coding experience, if I were given fizz-buzz during an interview (and nothing more substantive), I would likely decline an offer if it were extended. And if it weren’t, I wouldn’t recommend that my friends work there either. Why? Because giving a mid-level programmer a question suitable for a programming 101 mid-term shows that you have pretty low standards, and don’t really know how to get at what a candidate really knows and can do. I don’t want to work at a place with low standards.

          1. Henning Makholm*

            Your parenthetical “and nothing more substantive” is the key here. Yes, if “warm body and can program FizzBuzz” is all the employer wants, then I’m sure I won’t be happy in that job (never mind the actual work, but what kind of coworkers would I have to deal with? Eek!).

            But on the other hand, I’m cool with their using it as an initial screening tool — that doesn’t tell anything about the height of their standard, just that lots of their applicants who’re not up to a programming 101 midterm. Getting all those who aren’t me screened out quickly gives them more time to discover how wonderful I am, and I’m 100% for that.

      2. Mike*

        I agree that copying and pasting the answer is bad. That said, a lot of weeding questions are so basic there are only a few reasonable approaches to take. So, the person could have done it themselves and have it match code on a website somewhere.

        1. Mike*

          One more thing:

          When I’m evaluating programmers I care less about the code (as long as it works) than why they did what they did. So make part of the question to document WHY they chose their approach. This has an added bonus of seeing who fully reads the question and documents the why instead of the how.

        2. Ask a Manager* Post author

          Maybe the differing viewpoints here are because I’m someone who never asks coding questions in interviews, because I have no expertise there. When I’ve seen this type of thing happen, it’s not involving code — it’s involving outright plagiarism (presenting someone else’s work word-for-word as their own original thoughts). So it’s much more clear-cut.

          If you guys are telling me that this is totally normal with coding-type questions, even in a phone interview, then I can accept that. And I’d ask in that case that you read my answer as applying to non-coding situations.

          (However, if the OP weighs back in and says that he clearly asked the candidate to demonstrate his own work, then I assume you’ll agree the candidate did indeed cheat/lie/mislead.)

          1. Cathy*

            I’d have to say it’s not really normal to ask coding questions in a phone interview. I’ve never asked or been asked anything like this on the phone, so that part struck me as a little odd.

            I might ask how someone would approach a specific problem or what types of problems he’d faced in his recent past positions and how he solved them. I’m not expecting code as a response though. I’m using these questions to test his English communication skills as much as his programming skills. If you can describe a problem and have a discussion about the solutions you considered and why you chose the one you did, I’m not that worried about whether you’ve memorized the syntax for importing a WSDL in Java.

            Writing actual code (or more likely pseudo code on the white board) is for the in-person interview. I’ve also occasionally asked people to send me examples of their work if they have anything that’s not proprietary to their current employer.

        3. EJ*

          As a programmer, I agree. All of AAM’s advice applies if this is confirmed plagiarism (as she points out).

          1. Henning Makholm*

            Thirded, also as a programmer. I don’t think anyone who’s qualified to judge if the test is passed or not would have any difficulty knowing how much or how little room for variation there is in the task, and adjust their expectations of originality accordingly.

      3. Steve Martin*

        It’s not clear to me from the question that the OP asked the candidate to come up with code on the fly. The OP only says they wanted an answer to a technical question, and the coded answer they received was copy/pasted. I agree that the need/desire for original work can be made clear, but I can’t tell whether that was the case here or not.

        My default expectation when I’m asked a technical question is that the person wants an answer, period. I don’t assume caveats to how the answer is arrived at. Other than opinion questions I can think of only one time (post-college) where I was asked a technical question and looking up the answer wasn’t acceptable.

        I don’t try to pass others’ work off as my own, but if an interviewer wouldn’t allow me to use answers/solutions I already had available or could find easily and insisted I instead reinvent the wheel I’d be inclined to self-select out of that position.

        1. EJ*

          I would suggest it’s a given that material produced during an interview is expected to be your own unless you cite your source.

          As a programmer, of course I reuse code all the time in my work products. It would be ridiculous not to. But this is an interview, where I’m being asked to demonstrate my skills and my thought process. I shouldn’t be putting forward reused code.

            1. RG*

              Then say “I’ve run into a similar problem before, and this is the type of code I used to solve it. I would make XYZ tweaks to fit your parameters.”

              You don’t need to reinvent the wheel, but you should be able to demonstrate that you know how the wheel works. If you are using recycled material, cite it, even if you are citing yourself. It’s kind of like showing your work on a math problem, vs just writing down the final answer. I want to know that I’m hiring someone that can think and reason, not just come up with the right answer.

    2. Anon*

      It seems like there’s a side issue here about coming up with the kinds of tasks that someone would be expected to do on their own in the actual job. For instance, I’m a lawyer. There’s any number of types of routine motions I wouldn’t have a candidate draft, because my expectation is that they will be using the same boilerplates that everyone uses; it’s just a waste of time to expect them to reinvent the wheel and I wouldn’t ask them to do it in a hiring situation. (In practice you’ll tweak said boilerplate to reflect the individual situation, but that’s subtler, I think.) Instead, I’d ask for something involving substantive legal analysis of a question they’re not likely to have encountered before. That way they’re doing something I’d want them to do if they were hired rather than something you’d only ever do in a fake practice-type situation.

      1. Ask a Manager* Post author

        I agree. I’ve usually asked legal candidates to do research and analysis. But I don’t think it changes anything about the overall question here: it’s not okay to copy someone else’s work when you’re clearly being asked for your own (even if you think it’s a silly exercise).

  8. NUM*

    What is it you want to have happen in the interview and what kind of employee do you want?

    When you give someone (an employee) a short coding task with a little internet application, do you want that person to spend hours and hours developing their unique solution OR do you want that person to draw on their own or others experience to give you something that works? Your interview candidates clearly possess enough technical knowledge to clearly understand the problem, identify the relevant solutions, search for alternative solutions, and give you the results. What’s so wrong about that?

    I only ask to raise two points. A) The behavior these two are exhibiting may in fact be exactly what you want your employees to do in their real world job. B) Your initial statement doesn’t explain how this is lying or cheating. Based on the information you give, it is not possible to know whether this is unethical behavior on the part of your candidates or whether you are applying a sort of “purity test” based on your own beliefs and prejudices.

    2. KellyK*

      These are good points to consider. I think there’s a difference between looking up a way to solve a problem to avoid reinventing the wheel and copying someone else’s work without understanding how it works and being able to customize and tweak it as needed.

      It’s also never ethical to pass off something someone else has done as your own. So someone who’s looking up answers or copying code should be forthright about the fact that that’s what they’re doing.

    3. Elizabeth*

      If the interview questions are well-designed (not saying they always are), then it shouldn’t take a qualified candidate “hours and hours” to come up with a solution without copying and pasting. Drawing on your own experience and using things you’ve done before is fine; looking things up may or may not be fine depending on what the interviewer wants (and I think it is valid, sometimes, for an interviewer to say, “We want to see how you do on this without looking things up”). If you’re not sure, I’d ask: “Is it okay for me to look some things up on these questions?” Or say, “I referred back to some code I wrote for ABC, and there I did X. Since this is different in way Y, I’d make change Z.”

      1. NUM*

        I agree the interview question shouldn’t take that long. I only meant that once the person was hired and was working on these tasks, you’d want them to be efficient as possible.

  9. Steve G (from NYC)*

    Maybe the tests should just be done on computers without internet access and/or monitored?

    1. Mike*

      When I did my in-house interview I used the interviewer’s laptop and he was sitting right over my shoulder. Was an interesting experience as I was using an unfamiliar editor and have never coded with a person looking at my screen while coding.

  10. Mike B.*

    Raising the issue just alerts the person that they shouldn’t try to get away with this particular type of deception in the future. Why does someone with such poor character deserve that assistance? Let them get rejected endlessly until they figure it out or encounter an interviewer who doesn’t do their homework.

    1. KellyK*

      I think that actually depends. There are two possible messages a blatant cheater can take from it. Either “You will get caught and you won’t get the job, so you should knock that off” or “That was too blatant; you need to be more subtle.” Sending the first message is a good thing.

      Also, raising the issue is more for your benefit than theirs. Following up lets you know whether they’re actually cheating or there was some misunderstanding.

  11. Lily*

    Alison, thanks for explaining in a way I understand!

    Regarding #2, my experience says that people are likely to react defensively and/or offensively and I’m guessing that you will break off the conversation before you get upset yourself (and start sounding punitive). Or give me hope that some people actually appreciate the feedback?

    If this is getting too off topic, please tell me, but I would like to ask when it would be appropriate to confront someone about lying or cheating AFTER you hired them. Would you always confront all of the following types of workers: employees, temporary workers, contractors (who work full time for some months), free-lancers (who work only a couple hours a week)…?

    How would you handle people who are still officially learners, like student workers, interns, apprentices, differently?

    1. Ask a Manager* Post author

      With #2, I think I’ve always done it via email (because it’s been in response to written work they emailed to me), and that certainly helps keep it from becoming contentious. I had two people deny it, which wasn’t credible given the fact that what they’d produced was word-for-word identical to someone else’s work, and I just said, “I’m sure you understand that this raises a red flag and we can’t forward you on in our interview process” and that was that. If someone did get defensive, I’d just ignore it. Again, easier in email than on the phone.

      Once on the job… It’s a firing offense. It has to be. You have to be able to trust the people you work with not to deliberately deceive you. I would address it at once, with any type of worker.

      1. Elizabeth*

        “with any type of worker”

        As a teacher, I second this. By the time students are legally old enough to work, they have definitely been told that passing someone else’s work off as their own is Not Okay. There’s no excuse for someone over the age of 16 to copy-paste others’ work and claim it is theirs.

        The only time I might possibly conceive of giving someone a second chance is if they were from a very different culture and had only recently moved to this country. In that case, I’d try to find out if plagiarism is less of a big deal in the person’s native culture, and if so, I might give one VERY firm warning followed up with watching that person’s work very closely. Second offense would be firing.

    2. Joey*

      On thr job you confront them as soon as you find out. Give them a chance to explain. Personally I look at how big of an impact the lie had, the intent, and whether theyre truthful with me before I decide what to do.

    3. Not So NewReader*

      As an aside, I remember seeing an article quite a while ago, that told people (especially those new to their field) to Google search their own ideas BEFORE presenting the idea as their own.

      A good idea is a good idea, period. Some one comes up with a good idea it is not unheard of for others to arrive at the same conclusion. (Ask the US Patent office- they see a lot of this. And it is not all theft, some of it is actual coincidence.) The article suggested that people check on the internet to see if their idea is already in place some where else.
      Of course, this is of no help to people who are in interviews and need to produce something on the spot. However, for interviewing purposes if everything else about the candidate is great- except for one example of questionable work- it might not be a bad idea to ask the candidate about it. I would think a sincere (honest) person would be stunned then grateful that the interviewer pointed it out- that would give the applicant an opportunity to explain. (I would be complimented if an interviewer thought enough of me to ask me.)
      Again- this would probably be only in rare cases and only when the applicant checks out okay on most of the interview.

  12. EngineerGirl*

    Please ask clarifying questions. I actually was accused of cheating because my work looked exactly like someone else. It turns out that someone had plagerised my work. The hiring manager decided to reject both of us instead of allowing me to prove it was mine. Later it became very very apparent who had performed the work. The hiring manager then tried to recruit me to the team – to clean up the mess created by a third person he hired. I cheerfully told him I wasn’t available (a true statement). I realize that the hiring manager was allowed to hire anyone but the unwillingness to look deeper was a red flag to me. That program had a lot of problems because management wouldn’t look one level down.

    1. AgilePhalanges*

      This is less likely to happen in a hiring situation, but I was accused of plagiarism by a professor once, and thankfully he allowed me to explain and ultimately believed me. However, it wasn’t an accusation of cutting and pasting. He viewed the properties of the Word document, saw that the document had originally been created prior to the start of the course (over a year prior, if I remember correctly, but that’s not relevant), and accused me of using someone else’s paper. However, especially for college classes and super-especially when different professors have different formatting requirements (MLA vs. APA vs. their own formatting rules), I like to open up a previous document with the correct formating (everything from margins to font to spacing, as well as MLA/APA/whatever) and update the original text with new text, but keeping the formatting. I explained this to the instructor, and I think I sent the original document as well, and a screen print showing that the “last saved” date on the document in question was prior to the accusation (so I hadn’t JUST done something in reaction to the accusation) but after the assignment was given. I also asked the instructor to run the text through the searches I know they can use.

      Granted, document creation date isn’t a very good measure of plagiarism, but I’m glad that professor gave me a chance to explain rather than giving me a zero with no questions asked. So while I’m sure there are circumstances where #1 is appropriate, as there is zero question that they cheated in some way, if there’s any possibility for a misunderstanding by the applicant OR by the interviewer of the applicant’s technique, or any possibility for coincidence (or for another party to have plagiarized FROM the person in question as above, though I’d imagine that’s rare in interviewing situations), please please PLEASE give the accused a chance to explain.

      1. Judy*

        If you’re still in school and doing this, Save As -> Template, and open the template to start the document.

        1. AgilePhalanges*

          I’m done with school now (and finished with that particular school a few years ago), but yes, that might have saved me some heartache. However, I found it easier to just browse through my existing documents for the one that fit best for that particular paper (first, same professor, if that didn’t work, different professor with same formatting requirements, etc., and finally, choosing one with similar TYPES of reference materials (web site vs. scholarly article vs. book) so I could use THAT formatting rather than having to re-look-up MLA/APA formatting online all over again). I would have had a million templates if I’d use templates for everything. :-)

      2. Anonymous*

        I can’t believe the professor – there are probably lots of people that use a ‘template’ to start their papers!

        1. AgilePhalanges*

          I like to think the professor learned from that experience, and only questioned students on plagiarism based on the CONTENT of their papers from then on, not the creation date, but who knows.

          1. Anonymous*

            Who knows, said professor might even have learned that simply handing out the appropriate LaTeX style file is going to be the easiest way to ensure they get the appropriate format. Then let their students worry about the content.

  13. Tina*

    The thing is, even when you’re creating “original” code, you’re still looking stuff up. No software developer keeps it all in their head. You’re looking up specific libraries or commands and maybe looking up parameters or attributes or something. I very much doubt that the interviewer expecting a coder to work solely from memory is holding appropriate expectations.

    Furthermore, in the real world, of COURSE coders reuse other people’s code. I use widgets and other things made by other people for bits and bobs all the time. In my experience, reasonable employers care more about the thing working and care less about whether or not you could create the code if locked in a room using only a pen and paper. The technology moves fast, and there’s so much of it, it’s unrealistic to expect it all to be just in someone’s brain. And why would you want that anyway? Don’t you want someone who is willing to learn new stuff to build something?

    Rote knowledge is a poor test for software developers, IMO. As AAM often says, real-world type tasks are better.

  14. Dan*

    While I don’t write code full time at my job, I do dabble in it. A good interview should be looking at at least two things: What the candidate knows, and how the candidate thinks. What you’re really getting at is why the candidate made the design choices that he did — it demonstrates both technical knowledge and his ability to think. While “fizz buzz” is a bit elementary, an interviewer can always ask someone to code something a bit more difficult. Then, you can go back over the code, and ask the candidate which parts he likes, and which parts could use some work.

  15. Dan*

    Asking *good* technical questions is a must, and companies need to learn to do it better. At one job, I was giving 30 minutes to do some technical work. They gave me a bunch of messy data and wanted me to make some excel charts and answer some questions.

    Sounds fine, right? The kicker is that the data was massively garbled, and using Excel to straighten it out was akin to making a square peg fit into a round hole. TBH, I’d use some other free ware/open source stuff to clean up my data before analyzing it in Excel.

    All the interviewer really learned about me was that I’m not very good at using Excel to clean up crappy data.

    The regular interview went rather poorly, so I opted to let the interviewer know that the technical portion was pretty messed up and they could come up with better methods of candidate evaluation. The response I got was “we realize that our technical interview is a work in progress.”

  16. Brad*

    I’m the OP. I’ll try to answer most of your questions here. Let me know if I miss anything.

    I’ve definitely seen it two times during my career so far. In the first, I asked a technical question and heard the guy typing over the phone while saying “hrmmm…I kind of remember…hrmmm” before spitting out a highly technical and detailed answer. My question was about what he knew, not what he could look up. There, I was trying to figure out where he was in his career from a technical perspective — it wasn’t a make-or-break question. I feel like I got was some info about his character instead. I didn’t bring it up and the rest of his interview was underwhelming, so we just dropped him.

    In the most recent case, it was verbatim copy of a StackOverflow solution and it was clear that I was asking him to produce a small, simple code fragment. Moreover, he told me that he was writing the code. What he produced was some fairly complicated C-code (bitwise logic and pointer arithmetic) for a guy that told me he’d rather do it in Java because he isn’t comfortable with C++. And he did it in 5-10 minutes. And the author of that StackOverflow solution wasn’t him. I was going to question him about it in a fairly non-confrontational way (have him explain the code, ask why he used bitwise logic, etc.), but his phone died, so I just emailed him a link to the StackOverflow solution (per a coworkers suggestion) and I don’t expect to hear back from him.

    The question isn’t really tricky and we usually help candidates out if they get stuck on something. It’s meant to establish a few things. The first set are skills-based: Can you code? How do you go about it? The last set are more character based. Things like: When you make a mistake or get stuck, how do you react? He could have said, “I don’t know, but I could look it up,” and that would have been a fine answer. Not a great answer, but definitely better than what happened.

    Usually, we don’t get into the second part until we start asking more difficult questions during an in-person interview. We specifically ask them difficult questions because it takes some thought to work through them and lets us see what they’ll do when there isn’t an obvious answer to a problem. We’re not asking questions to see if they have technology-specific answers (minus a minimum-level of developer ability). And we do ask other questions as well that dip more into their character and less into their technical prowess.

    As to whether or not we want candidates that can search for solutions, of course we do. But I also want them to think about what and why they’re using the code in the first place. And I don’t want them to present me with some open-source code and tell me that it’s what they wrote it last week (or who take credit for other people’s work). “Cheating” probably wasn’t the best word to use (it was in the context of a skills test, which is really only part of what this is about). “Plagiarizing” seems more inline. The question is mostly about catching someone misrepresenting their capabilities, I guess (in this case, by passing off someone else’s work as your own).

    I tend to lean towards asking them about it, if only because it dispels any doubt I might have about them. That wasn’t what I used to do and it wasn’t what I was originally instructed to do. So, I was curious to see what Allison and others thought.

    1. CassJ*

      I’m semi-curious how those two would do in an in-person interview when asked to whiteboard. (Wonder if they’re the type of people who say, “I don’t whiteboard” during interviews that I’ve heard about).

      1. Brad*

        “I don’t whiteboard” would be a deal killer around here. Whiteboarding is an important aspect of what we do. You may as well have said, “I don’t do work.”

  17. Lee*

    I’m sorry, but if one can simply look up the answers to your questions on the internet, and be correct and they choose to exercise that resource rather than spew out tech crap they’ve memorized, whats the difference? If I can do every tech spec, even if I have to look it up every way, how am I less qualified than a candidate whose be able to afford years of school, who just memorizes and parrots back the same thing?

    “Never memorize what you can look up in books”- Einstein

    1. Anonymous*

      Tech crap eh? If you don’t see the value in having a candidate that actually understands the fundamentals rather than simply parrotting something off the internet, then I don’t know what to say…

      1. Laura L*

        Exactly. This is why teacher’s often ask students to explain something in their own words. It shows the teacher the student actually understands the concept and isn’t copying the definition out of the textbook or lecture notes.

    2. Elizabeth*

      There are some things you should have memorized, though, in every field, even if you can look it up. For example, a HR person should know the gist of what constitutes sexual harassment, a teacher should know CPR, a truck driver should know when it’s legal to turn right on red, and a fisherman should know if the thing he just caught is a salmon or a carp.

      It’s a bad interview if all it tests is “have you memorized every technical detail,” but it’s also a bad interview if it doesn’t make sure you can think on your feet and not rely on Google for all the basics.

      It also doesn’t sound like the interviewers who have experienced this thought that the interviewees who were looking things up actually *did* get them correct. Rather, they were able to say something with some of the right words, but not understand it.

  18. Anonymous*

    You could at least look to see if they plagiarised a good solution to the problem. A friend of mine once broke up a “CS homework copying ring” (he made it sound suitably dramatic) because they’d all solved the same simple problem in a really stupid way. There were only a couple of sensible ways to solve it, so this really stood out. If they’d managed to copy a good solution, he’d never have caught them.

  19. NUM*

    RE: Brad’s post @ 3:37pm

    Good answer. I am still confused about how this situation presents itself to the candidate and how the candidate might perceive it very differently than you do. Here’s why:

    You are combining two different tests – skills and character. But, you are using a test that only works as a skills test to do it. You say you are not asking these questions to determine whether the candidate has the correct technology-specific answer. However, the fact remains that the question DOES have a correct technology-specific answer. How is the candidate supposed to know that this is not a knowledge test when you assign a C coding task? The only answer that makes sense to the candidate is that, of course, you are testing C expertise.

    Where this gets really confusing is that now you are interviewing a candidate who has little or no C++ expertise. He knows he doesn’t have the expertise. You know he doesn’t have the expertise (since he told you that himself). Yet, here you are talking to him about the job anyway andall of a sudden here comes this (!!) C coding task. Obviously, a technical hurdle he has to overcome. So, using the skills he does have, he finds an answer. Perhaps not the best solution and clearly not the approach you were looking for – but, yes, an answer to the question he was asked.

    Would it have been better for him to repeat that his expertise is in Java coding? Maybe, but perhaps he would think it rude or argumentative to have to tell you that again. And, yes, it would have been better had he told you, “Since I don’t have big C coding skills, I did a search and found an example of something that should work in this situation ….”

    All this takes me back up to the very top of this thread, where Anon-2 referred to an old Chinese proverb and asked how does it help your decision if you ask twenty questions and the candidate slips up on one or two. Clearly you are trying to give your candidates something with which they will struggle and demonstrate their problem solving approach/abilities. In most cases, this probably works fine (since you’ve only had trouble with two candidates in 5 years of interviewing). I suspect, however, that every now and then you are coming across a candidate or two who (very reasonably) approaches this particular question differently. Not lying or plagiarizing or cheating, just with a different approach.

    If I had enough really good candidates, I cannot see how eliminating these few would cause problems. But, if these candidates excelled in every other way, I don’t think I would write them off because of this.

    1. Brad*

      “Good answer.”

      Thanks! :)

      “However, the fact remains that the question DOES have a correct technology-specific answer.”

      It has multiple answers that can be implemented in any technology. In fact, I told him to do it in Java instead of C++, since he was more comfortable in it. It would be equivalent to asking someone to write a function to sort a list of integers (without calling some frameworks List.Sort() method).

      “Where this gets really confusing is that now you are interviewing a candidate who has little or no C++ expertise.”

      He claimed to have C and C++ knowledge — something I’m interested in candidates having, but something I’m not going to drop anyone over not having. So, I specifically asked for it in C++ because that’s a skill we use here. And, again, when he countered he’d rather do it in Java, I told him it was fine. Had he said MIPS, on the other hand… ;)

      “since you’ve only had trouble with two candidates in 5 years of interviewing”

      Maybe I’m just not that good at catching them. ;) Most of my experience was with a different company with a completely different set of hiring practices and candidate pool. We did a lot of “beating heart” hiring and had an extensive training program (so, it was usually us trying to gauge where someone was and see how much latent talent they might have). Where I am now (past 6 months or so), we’re trying to only hire A+ candidates.

      “If I had enough really good candidates, I cannot see how eliminating these few would cause problems. But, if these candidates excelled in every other way, I don’t think I would write them off because of this.”

      We do have quite a few good candidates, so this one isn’t any skin off my back. But I *do* think that the way questions like these are answered should dictate whether or not you want the person working for you. I think this is particularly true on the other questions we ask during our in-person interviews (the next level after the phone screen). Some people immediately give up. Some ask questions. Some say they would search something or seek out advice from more experienced members. Some do some drawing and design and some jump straight into the code. Or maybe they storm out of the interview (which I haven’t heard about happening here). All of that tells you a bit about how they approach technical or personal problems that don’t have obvious answers.

      1. Chriama*

        A lot of people have commented on the nature of your interview questions and coding questions in general, but your follow up makes it clear that you’re talking about candidates who misrepresent themselves in situations with little or no ambiguity about what is required of them. In that light, I think Alison’s advice is spot-on. If I was the least bit unsure of their intent, I’d be inclined to go with #2 over #1 just because I like giving people the benefit of the doubt and calling them out on it gives them a chance to explain. Overall though, if you’re certain that a candidate has plagiarized on purpose with full knowledge of what they were doing, I might just quietly remove them from the running. It’s not worth your time to try and deal with any arguments or false justifications they might try to throw your way.

      2. Dan*

        Oh that’s funny. I guess my first question would be:

        1. You indicated a preference for the Java programming language. Why did you submit a solution in C?

        And second:

        2. Outline the much simpler solution, and then ask why he chose the more complicated route.

        If he gives you competent answers to 1) and 2), then start asking questions about pointer arithmetic. If he answers those competently and promptly, then well, he actually knows what he’s talking about.

        1. Brad*

          I was going to ask #1 and then say, “Walk me through your solution,” when his battery apparently died (assuming he was being honest there). I like #2 and the follow-ups, though.

          And, before anyone assumes I’m a terrible interviewer for making a phone screen so long it drained his *entire* phone battery, we were around the 30 minute mark and I told him it would take between 30-60 minutes at the beginning. He could’ve easily anticipated needing to charge his phone or to have it plugged in. He suspiciously disconnected with “phone problems” when I asked him to read some code and tell me what it did.

  20. A S*

    How is that lying, exactly? And how is that a negative reflection on the interviewee? S/he is exhibiting resourcefullness in how to get the job done! It isn’t about who is “smartest” (ugh, I hate that type of bravado!) , it’s about the ability to get the job done. NO ONE knows everything 100% of the time.

    This reeks to me of the OP just trying to prove s/he is better/smarter than everyone else in the world. I wonder if they ever have hired someone who did this but was more subtle? Probably dozens of times. Get over yourself.

  21. Coder*

    If a person can look up for an answer online easily and swiftly, does it mean the “test” is not cleverly crafted to test the coding skill of the candidate?

    The candidate knows where to find what they need and it shows they are “on trend” and know what you are asking and talking about. They delivered what you asked for. If you are looking for something unique, set a unique test question.

    Being an interviewer does not give you an upper hand to judge a person’s integrity when you have failed to ask the RIGHT question.

Comments are closed.