As I've already saw, many people here are related to admission process. And I want to ask their opinion on what is priority of all stuff from application: what is the most important (letters of recommendation or personal statement, GPA or maybe test scores), what is less important for admission and what is the least important? Would admission committee prefer applicant with good scores, high GPA, but with not outstanding personal statement and without letters from famous or well-known professors (for example, my scientific advisor in undergraduate school get his PhD only a couple years ago), or student with less GPA, test scores, but with letter from university's prof?

Some graduate schools require General GRE (Graduate Record Examinations). Would it better to send GRE scores (general or subject or both) even if it's not required (in case of relatively good result), or them wouldn't be considered at all?

2 Answers 2


As aeismail says, what we are looking for is concrete evidence of research potential. So, in decreasing order of importance:

  • Decent grades and (if required) test scores; otherwise, no one will read your application at all. Here, "decent grades" means three-point-something from a good undergraduate program.

  • Research publications, if you have any.

  • Recommendation letters. You must have at least one (and preferably three) strong recommendation letter from a faculty member who praises your research potential in specific and credible detail. Letters that draw specific comparisons to other successful PhD students are best. Letters from junior faculty are perfectly fine; they can draw comparisons to their recent graduate school peers. Letters that say only "He got an A+ in my class" are useless; we can read your transcript.

  • Research statement. Your statement must discuss your research experience and interests in specific and credible detail. A statement that only describes your sources of inspiration ("Ever since man walked on the moon...") and/or brags about coursework is useless.

  • Other concrete evidence of independent research/scholarship/creativity.

  • No red flags. Potential red flags include low grades in classes central to your proposed research area, missing key classes entirely, abysmal test scores, negative (or overly delicate) recommendation letters, recommendation letters obviously written by the applicant, spelling and grammar mistakes in the research statement, any evidence of immaturity or personality issues, and any evidence that the applicant is not prepared/informed/serious about research.

Notice what's not listed.

  • 1
    I wonder, what are the symptoms of a recommendation letter written by the applicant?
    – Ran G.
    Mar 30, 2012 at 17:21
  • 2
    In some cases, the statement and the letter(s) had several descriptive phrases in common, or even entire sentences. In other cases, letters for different applicants, ostensibly from the same person, had significantly different levels of English fluency.
    – JeffE
    Mar 30, 2012 at 21:15
  • Thanks a lot for your answer. And once again about tests - are they used only for pass-fail decision (red flag)? Do better test scores increase chances to be admitted? And what about sending test scores when they are not required?
    – MVB
    Mar 31, 2012 at 20:18
  • That varies a lot by department, and even from one evaluator to the next. There's no way to tell in advance whether your target department will care about your test scores, so it's best to assume that yes, higher scores help. On the other hand, I don't see any advantage of sending non-required scores, unless they're truly outstanding and/or you need to overcome some other weakness in your record.
    – JeffE
    Apr 2, 2012 at 20:29
  • 1
    @DavideChicco.it The list is not that different for postdoc hiring (not "admission"), at least in my field (computer science). The big differences are (1) grades do not matter at all and (2) the focus changes from "potential for research" to potential for intellectual leadership backed by actual published research.
    – JeffE
    Apr 23, 2012 at 20:49

If we're talking about a PhD candidate, the most relevant factors, for me, are those that demonstrate the applicant's capability to be a productive researcher. That means a compelling statement of purpose (why the applicant wants to be a PhD student, with ideas for potential areas of exploration, if not necessarily an entirely thought-out project) counts for quite a bit.

Equally valuable are letters of recommendation that actually talk about the candidate. I don't care for a letter that regurgitates the student's academic performance, or that reads like a template in which the name of the applicant was swapped in for somebody else's. That does me no good in evaluating a candidate, and to some extent can work as a mark against a candidate, since it indicates a lapse in decision-making (why get a letter from someone who doesn't know you well?). A really good letter of recommendation, however, can sometimes make the difference between somebody who's on the bubble and somebody who gets an offer of admission.

Transcripts and test scores do matter, but for me are less valuable, because they don't give me a lot to go on—it's not easy to tell what's a "good" performance and what isn't, particularly when schools use "nonstandard" scales (such as reporting scores out of 100 without telling me what the average score is!).

The other factor that we do here that not all programs do is an interview (in person or via Skype). This is perhaps the most important part of the process for us, because we can see if the paper record matches the "actual" applicant. Sometimes candidates look good in writing, but can't really talk knowledgeably about what they have studied. That's usually a clear sign that someone really won't be a good fit as a graduate student.


You must log in to answer this question.

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