I would like to do a Ph.D. in CS and I will start doing research as an undergraduate student soon, and I will be applying next year (December 2014) at good (say top 15) US universities. My problem is that I have interest in many areas and I don't know which one to choose. At first, I thought that would like to do research in Machine Learning, but then I realized that there are many undergraduate students doing research in that area so it will be more difficult for me to get in a good school by having research experience at that area, as I am not coming from a highly ranked program.

In addition I am an international student, currently studying Computer Engineering and not Computer Science and this may affect my application as well. So I am thinking about doing research in another area that may be a little less competitive and choose my field of interest (it may be the same) after I got into graduate school.

So my question is:

  • Will it be better to choose something that is less competitive (that still interests me) as a research area for my undergraduate thesis?
  • 1
    Someone has voted to close this question as off-topic. It is not. However, it may be a duplicate of this question, or this one, or almost all questions generated by this search
    – Ben Norris
    Commented Oct 13, 2013 at 18:37
  • Even though this question asks something more or less similar to those mentioned, it is not the same. I am not asking whether I can change my major for a Ph.D., nor whether I can research in another field after I get into graduate schools. It asks whether it will be better tο choose something with less competition to get an "easier" admission into graduate school. So, I think, it is not duplicate.
    – Grey
    Commented Oct 17, 2013 at 19:05
  • 1
    This question asks about undergraduate research and graduate admission. So, I think, that it is not off-topic.
    – Grey
    Commented Oct 17, 2013 at 19:06

3 Answers 3


Put simply: No.

Your undergraduate thesis work need not have any connection with your graduate research. Certainly, staying in the same area lets you have more of a "running start," but there's always the chance that you find a new research project as a graduate student that grabs your attention and doesn't let go.

So don't worry too much about doing research work in a less competitive area unless you want to do that research anyway. It's better to do work that excites and motivates you, because the end product will almost certainly be better than taking a project that doesn't.

  • I think that admission committees admit people by their research area interests, and thus some areas may have more people and this makes it harder for me to stand out if I choose this area for undergraduate research. Thank you for your answer, I think that if I find out a good project I will do it no matter what area is in. I just had this question in my head.
    – Grey
    Commented Aug 27, 2013 at 11:14
  • 7
    A piece of advice from other faculty I've spoken with: if you're interested in more than one area, be sure to mention that in your statement.
    – aeismail
    Commented Aug 27, 2013 at 14:09
  • Thank you for your advice, I will do so! I just wonder how professors that work in the other area will see my application. For example if I work on Operating Systems and I do a heavy programming project, and I also mention Theoretical CS as my other area of interest (not my case but I am trying to make it generic to be useful for other people too), then how will someone in Theoretical CS value my application?
    – Grey
    Commented Aug 29, 2013 at 9:15
  • All else being equal, any research is a net positive, but of course ceteris is never paribus. If the specific scenario you describe, I would also look for other evidence of excellence in theoretical CS — graduate-level classes, technical details in your statement, comments in recommendation letters, and the like. You'd be competing with applicants with direct TCS research experience. One of the theory students in my department did his undergrad research in compilers, and one of my former PhD students did his undergrad work in sensor networks, so moves are certainly possible.
    – JeffE
    Commented Oct 21, 2013 at 11:17
  • Having a broad skillset as a new graduate or postdoc is very valuable both in academia and in industry jobs. Being interested and skilled in multiple areas is a strength, not a weakness. Just don't spread yourself too thinly.
    – Moriarty
    Commented Oct 23, 2013 at 4:37

I would suggest you to have a look at this doc written by a professor of the admission committee of Carnegie Mellon University


  • 5
    Maybe you could draw some excerpts from the (nice) document that would be relevant for this particular question?
    – F'x
    Commented Aug 28, 2013 at 13:06

At this point I would choose to do research that pushes you or challenges you most. The admissions panel at any school are looking for skills and evidence of excellence, if you do research that challenges you, you are better able to show that off further along the line.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .