Dependable Systems Lab Dependable Systems Lab

Writing a Research Statement

A research statement is also known as a statement of purpose (SoP), and it’s a standard part of a PhD application folder. 99% of the SoPs I read are quite bad—not because the candidate is not qualified but because (s)he does not understand the purpose of a SoP and therefore does not optimize accordingly. Here are some thoughts on what I think makes a good SoP for graduate school applications.

Purpose

The purpose of an application is to get admitted, and the SoP is part of the application, so the goal of your SoP should be to get you admitted. Always keep this in mind, and make sure every sentence in the SoP serves this goal. The SoP should be coherent and conceptually focused.

Content

The SoP is read by scientists, and the admission decision is made by scientists, therefore target your SoP to an audience of scientists. Some of these are among the foremost experts in their fields. Focus on concrete facts. You can’t imagine how many SoPs make the mistake of starting with a highly emotional description of the candidate’s passion for computers since childhood.

What the SoP reader wants to know is:

  • Why do you want to do a PhD (= apprenticeship in how to do scientific research)? Do you want to pursue an academic career? Is there a particular challenge (describe it!) that you would like to tackle? Do you want to explore some ideas (describe them!) and follow an entrepreneurial route?
  • Which area of research do you want to specialize in and why? Make sure you know the field well enough to be able to talk about it in the language of the field; avoid coming across as an amateur.
  • Why are you a good fit for pursuing research in this area? Are there specific academic experiences or extracurricular activities that provide evidence for this fit? Describe them, describe your role in them, and describe the outcomes. Please keep in mind that your grade transcripts and CV are already in the application folder, so your SoP should complement them.
  • If there is any weirdness in your record, explain it. For example, if you want to do research in operating systems and you got a B in your undergraduate OS class, use this opportunity to clarify what happened. Don’t make hollow excuses, though—if that semester you were madly in love and couldn’t care less about school, don’t make up stories, just skip it altogether.
  • Explain unusual circumstances (e.g., you held a full-time job while pursuing your university studies).
  • Make sure the SoP conveys a clear picture of you as a future scientist and you as a human being. Explain what drives you, how you make your difficult decisions, how do you resolve ethical dilemmas, what aspects of yourself you are trying to improve, etc.

When describing things you’ve done, make sure you clearly articulate the value of those experiences to you, what you learned, and how that experience improved you. Use evidence to show that these transformations have indeed had effect. Do not just list the things that can be found in your CV.

Language

Make sure your grammar is correct. Spell-check.

Keep in mind that your SoP is one of hundreds. Get to the point quickly, avoid unnecessary introductions. Be objective and direct, use straightforward prose, avoid convoluted phrases (which are typically believed to denote confused thinking). Be specific whenever possible, provide clear examples and use specific instances to illustrate your statements. This will truly make your SoP stand out.

There are significant cultural differences in how candidates present themselves, and these differences are sometimes quite hard to overcome. Keep in mind that science favors objectivity, so statements like “EPFL is a highly esteemed institution of superior learning in the wonderful domains of science” will not improve your chances of admission, quite the contrary. Aim for your SoP to have a high density of information—it’s better to have a 1-page SoP that tells us a lot about you than a 2-page SoP stuffed with empty phrases.

Avoid subjective assessments and hyperbolae. If something is “significant” or “invaluable” then surely you can state this quantitatively. If something is “important” or “appealing” or “incredible” then surely it had some concrete effect on the world that you can use to support such a statement.

Did I mention spell-checking?