I am a PhD student, and was recently approached by a (very bright) student in an exercise class I teach, requesting a reference letter for his PhD application. This makes me wonder: Is it OK for a PhD student to write a reference for a PhD applicant?

There are obvious reasons why I might be able to say something informative: I saw a lot of his solutions to assigned problems, and I saw him work through an extended period of time (as opposed to say, a lecturer, who only sees his grades, and possibly final exam). However, I am concerned that I am not senior enough for my opinion to be taken seriously. Is it a legitimate concern? Would the student be better of asking someone more senior?

  • 1
    Did the student check whether the school he is applying to allows PhD students to act as reference? Even if the school allows it, the recommendation would have more weight with you providing the text and substance of the recommendation and the professor of the course providing the signature based on his trust in your observation and judgement. Commented Dec 21, 2014 at 12:23
  • Very similar (but from applicant side): academia.stackexchange.com/q/32829/2692
    – earthling
    Commented Dec 21, 2014 at 13:35
  • 7
    Apart from the other points raised, it is very possible that the undergrad has no idea how recommendation letters work, and you should prompt the student to think about possible (good-) letters from faculty, in case the students has not thought in such terms. They might be more comfortable talking to you... not realizing the realities of the letter-writing game. Commented Dec 21, 2014 at 15:21
  • academia.stackexchange.com/questions/34631/…
    – user131131
    Commented Oct 30, 2020 at 3:21

2 Answers 2


I think this is in part a cultural question. For example, in the US, I think letters from current graduate students are not given nearly as much weight as from more "senior" professionals. However, in German schools, it is quite common for senior doctoral students to write letters of recommendation for undergraduates, particularly in disciplines such as engineering, where the group sizes tend to be quite large, and in many cases the professor in charge of the group may never meet the undergraduate in question.


In the United States, there is nothing against it, but it is better to get the letter from a professor if possible. Usually, especially with a junior graduate student, the graduate student's supervisor will be involved in the work as well, in which case the recomender really should be the professor.

That said, there are situations where the graduate student is the right person to recommend, because there has not been a professor significantly involved. If the only options are:

  1. a letter from a graduate student explaining in detail how they have worked with this student and praising the excellent research that the student has done, versus

  2. a letter from a professor saying "this undergraduate exists and other people have told me good things"

then the letter from the graduate student is the right way to go. As a graduate student, I ended up writing some letters like this for undergraduates who worked for me. While it's impossible for me to know how their admissions committee rated my letter or what other letters they had, the people that I recommended did end up in Ph.D. programs, so it at least must not have been a problem to have me recommend.

You must log in to answer this question.

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