How do I write a good or great statement of purpose?

Faculty who are reviewing PhD student applications need to fit this task in between teaching, advising students, meetings, writing letters, end of term preparation, grading, submitting proposals, drafting papers, reviewing proposals and papers, and all the other mini-jobs that professors do. They also need to review hundreds of applications in a very short time. If someone were reviewing just 150 applications and spent just 2 minutes on each, that is still 5 hours straight through with no breaks! Clear, succinct, and mission-driven writing will carry the day. The points below are my personal opinions about what makes an application stand out and will be most applicable to students applying to programs in Electrical and/or Computer Engineering. Following these is not a substitute for a strong previous preparation or a good knowledge of problems in the field, but they will give you an advantage amongst two otherwise similarly prepared students.

You are welcome to take as much or as little of the suggestions below as you’d like. If you are a student who is applying to my group, I strongly recommend that you review and incorporate these principles into your statements. Good luck!

Tips for statement of purpose writing that may help you stand out:

  • Make an “executive summary” at the beginning that describes:
    • a very brief outline of your background
    • your preparation
    • your research interests
    • faculty whose interests your work aligns with and who you’d be interested in working with
    • very few reviewers will read to the end, at least on the first pass
    • you do not need to make “executive summary” a heading. If your writing doesn’t flow nicely from the first to the second paragraph, you may wish to make some horizontal line, or to otherwise use headings, to distinguish the summary from the other sections
  • Make it non-chronological
    • Focus on the most important details first: what you want to do and how your studies have prepared you to start doing it
    • A good paragraph-by-paragraph structure might be:
      • executive summary
      • more specific paragraph of research interests and problems
      • description of overall preparation for PhD study
      • more specific preparation describing projects, papers, etc.
      • summary that ties things together and describes next career goals
  • Make it forward looking. The statement of purpose is the only place in the application that you have to directly convince potential advisors that you are ready to embark on PhD study and have thought about the following questions:
    • What are the important problems in your area of interest?
    • What types of problems would you like to work on?
    • What are the applications of these problems, and who cares beyond our small circle of researchers?
    • How are you prepared to start addressing these problems?
    • How do you know you’ll like working on this problem for 3-6 years?
  • Make it specific
    • A professor should be able to read the statement and say “this student wants to work on X, and I should mention their application to my departmental colleague that does X+Y.”
    • If you don’t write with a level of detail where it’s clear what you want to work on (not just area/sub-field!) you might as well not apply. If you don’t know the answer to this level of detail– read more at the end of the document!
  • Make it about your qualifications as a scientist or engineer NOW
    • Emotional appeals or setups about “when I was a child” or “my grandfather taught me” have their place in creative writing, but not here.
    • Focus on what you’ve done in the last ~3 years and what you want to do in your PhD
    • Unless they place you in the top ~1-2% of students worldwide, don’t include things from high school or the first year of your undergraduate program
  • Make it fewer than two pages. Seriously— the ability to write well and concisely will impress a lot of potential advisors.
  • Make it active voice
    • “I was fortunate to attend…” -> “I attended…”
    • “I was awarded” -> “I won”
    • “I was accepted into Prof. Toad’s lab” -> “In Prof. Toad’s Lab, I….”
    • “I was thrilled to [x]” -> “I [x]ed”
  • Make it well-written
    • Yes, grammar, spelling, and format count!
    • Visit your university’s writing center or ask others to read it
    • Generative AI can help with writing mechanics and clarity, but from scratch it will only write a bland statement that won’t impress anyone. Always at least make your own first draft.
  • Make it legible
    • The more esteemed the potential advisor, the likelier that their eyesight is not stellar 😉
    • Absolute minimum is 11 pt font (Helvetica, Times, Arial, etc.), 1/2” margins. I can hear some of you saying “but you can just increase the PDF size! Not all application portals permit easy resizing of the document.
    • Bold, underline, or italics text is good but should be used once a paragraph or less.
    • Use bullet points very sparingly. One of your jobs is to demonstrate that you can write like a (developing) academic, and academic papers have few to no bullet points

How do I email a potential advisor for the first time?

These instructions are for students who are interested in applying to my group, and they may also be useful for others. If you’re emailing anyone other than me, please check the potential advisor’s website and make sure there are not more specific instructions. Some faculty have specific protocols to streamline their email.

  • First paragraph: Who you are and why you’re emailing
    • I’m writing to apply for [PhD, summer visiting student, etc.]
    • Info about BS/MS degree and GPA
    • Say whether you have funding or an application pending for funding (e.g., Fulbright, NSF GRF)
    • Make this paragraph *very* short
  • Second paragraph: Why my lab?
    • Why you’re interested
    • How you’re prepared
    • Concrete demonstration of your preparation
  • Let’s say you’re applying to a lab that works on Frog Engineering. Here are some not so good and better examples of how to write this paragraph.
    • Not great example: I read your group’s recent paper, “Jumping behavior of Rana draytonii in urban environments” and found it really interesting. I have worked on Rana tagoi jumping for the past two years, and I think your lab’s work would be great to keep doing this.
      • Problem: not specific, it seems like you have spammed every lab that does Frog Engineering research. Not clear what you did in your previous lab.
    • Not great example: I read your group’s recent paper, “Jumping behavior of Rana draytonii in urban environments” and found it really interesting. I have worked on designing cities in simulators, and so I think your lab is a great fit.
      • Problem: No clear alignment between your research and mine. I will assume it was a keyword search or that you’re spamming 300 professors and send a polite “our research interests are not aligned” auto-reject.
    • Not great example: I read your group’s recent paper, “Jumping behavior of Rana draytonii in urban environments” and found it really interesting. I want to get a PhD in Frog Engineering, and your lab seems like a good place to do it.
      • Problem: Not clear that you have thought at all about what specifically you want to do in your PhD or that you’re prepared in any way, just that you want to get a PhD. See below for more tips!
    • Much better example: I read the FROGGER Group’s recent paper, “Jumping behavior of Rana draytonii in urban environments.” In my PhD, I would like to optimize rana survival in cities, and particularly in high vehicle traffic areas. As an undergraduate student, I was a researcher in Professor Toad’s lab at the University of East Amphibia and contributed to the lab’s dataset of frog jumping strategies. We are working on a manuscript and anticipating submitting this by May 2025.
      • Why it’s better: Hits all the important points, succinct
    • Much better example: I read the FROGGER Group’s recent paper, “Jumping behavior of Rana draytonii in urban environments.” I am currently a junior frog finder with the Department of Transportation, and I led a project how to improve frog identification on highways that resulted in a 20% higher detection rate. In my PhD, I would like to study this problem and would hope to talk with you further about opportunities to join your group.
      • Why it’s better: This person shows their knowledge and preparation through industry/job experience without papers, which is also valuable.
    • Much better example: I read the FROGGER Group’s recent paper, “Jumping behavior of Rana draytonii in urban environments.” In my PhD, I would like to investigate approaches to improve frog catching using soft material approaches due to their gentler capabilities than metal traps. My previous background in Paper Engineering might seem like it’s a poor alignment with your group’s work, but I believe my skills in paper folding and fabrication will apply to the challenge of building containers that contain frogs without causing them stress, which will make it easier to study them.
      • Why it’s better: This applicant has no direct experience in the sub-field but has made a convincing argument about why their skills would make them successful and also has displayed a good understanding of the problems in the field.
  • Third paragraph:
    • State that you’re applying to their university and would like to work in their lab
    • Attach your CV or resume
    • Offer to share application materials, writing samples, or set up a time to have a phone/video call
    • Close by thanking them politely
  • You will have more success if you tailor 10 emails using the format above than if you spam 100 professors that have little to no connection to what you do.
  • Always address the potential advisor with their title (Professor or Dr.), make sure their name is spelled right, make sure your font and formatting is consistent, and make sure you have the right university name and spelling.

How do I figure out what I want to work on?

The pathway to a PhD is both a joy and a long slog (3 years? 5? 8?) of work on one or a few related problems and it usually doesn’t make sense from a economic standpoint. Advisors need to know before they recruit a student that the student has thought about what their specific interests are and are willing to stick around long enough to make progress on that problem. It takes a lot of time and effort to introduce students to the methods and approaches in any field, and an advisor will have more confidence in recruiting a student if that student can show their interest through 1) previous research, projects, papers, etc. and 2) their knowledge of the field’s problems. If you don’t have a clear idea of the types of problems you want to work on at the level of the sub-field, you should pause your application (including waiting until the next cycle to apply) and learn more.

  • Read reputable popular science/engineering magazines like IEEE Spectrum, Discover, Scientific American, or MIT Tech Review to get a high level understanding and to appreciate the longer view of research
  • Read conference and journal papers and see what interests you. This is more possible than ever before thanks to open access policies, but I know it can still be tough if you/your university doesn’t have access. To read these papers (legally) you can also:
    • Check arXiv.org for related papers
    • Check “all versions” of the paper on Google Scholar. Sometimes the authors put up pre-prints on their personal website or institutional database
    • Email the corresponding author! Sometimes researchers are willing to share the pre-print version of a paper
    • Use inter-library loan or Worldcat.org if your university supports it
  • Watch academic talks on YouTube. Virtual/Zoom talks that are recorded and shared have made this easier than ever.
  • If you have visiting researchers to your university, attend those talks! They don’t have to be in your specific area, but it will give you a good idea of how people frame and address the problems they work on.