Thursday, October 1, 2015

Wrapping up an undergraduate research thesis

How do you do it while maintaining sanity and quality?

In Australia a Bachelor of Science is a three year degree. Students have the option of doing an additional year, and being awarded an Honours degree. This is necessary to do a Ph.D. and may help to get into some government jobs.
The honours year is roughly half course work and half a research project leading to a thesis.
The thesis is a bit like a mini Masters degree in other countries.
The thesis is meant to involve original research. For the best students their results may be part of a publication. The thesis is typically 40-80 pages.

I believe that Princeton is unusual among US universities in requiring ALL their students to complete a senior thesis.

Assigning, supervising, and completing these projects is particularly challenging for both advisors and students. Previously I posted advice for students giving seminar talks based on these projects.
Now I turn to the thesis.

First, students don’t actually know much science or have much real research experience. Although there are now a host of summer scholarship schemes and other  opportunities for working in research groups.
Second, the greatest challenge is that there is a fixed deadline for submission of the thesis.
This is quite unlike a Masters or Ph.D where people tend to submit when they “have enough results”. I realise that funding running out or starting a new job sometimes plays a role in when students submit.
Unfortunately, it often seems that just when a student starts to get some useful results, or worst they just appear on the horizon, the thesis deadline is looming. Further stress is created by impending final exams!
Students can also put a lot of pressure on themselves, because getting a good grade for the thesis may influence their future options, particularly if they want to do a Ph.D.

So what should students (and advisors) do to maintain sanity and produce a quality thesis?

Don't wait until the last minute to start writing.
Too many students do this in the hope they are going to finally get their experiment or calculation to work and then slap together a thesis at the last minute, sometimes pulling “all nighters”.

Realistic expectations.
Face your perfectionism.
You are not going to write the best thesis ever.
Look at copies of theses from previous years.
You may have higher expectations than your examiners.

Know what is expected or hoped for.
The presentation of the thesis is as important as any new scientific content.
Examiners enjoy learning something new.

Again, start writing early. 
Don’t wait until you get your results.
Introduction. Literature review. Clear description of the project and its goals.
Methods. Polish what you have. This can all be done sooner than later.

Watch your mental health.

7 comments:

  1. A bit off-topic, but I recommend the blog by Anshul Kogar, and specifically the latest post:
    https://thiscondensedlife.wordpress.com/2015/10/01/some-questions-that-arose-during-my-graduate-years-2/

    That blog, similarly to this one, presents a nice mix of basic physics discussions and broader topics.

    ReplyDelete
    Replies
    1. I agree this is a nice blog. I have already put a link on my sidebar.

      The questions posed are worth thinking about. I think that it is a bit of sad statement of current affairs that grad students are being forced to asked such questions. I don't recall during my Ph.D 30 years ago any discussion of these topics!
      Some of the questions I have addressed in various posts.

      Delete
  2. Good advice: I think I followed these principles.

    The other point I'd mention is that if you think you have the basis of a publication and aren't planning to continue in the group after Honours, start writing it straight away! Writing gets so much harder after you leave the group.

    ReplyDelete
    Replies
    1. Well done, if you did these! It is not easy.

      I agree with you additional suggestion. I think there can be a significant emotional obstacle to doing what you recommend. Students are understandably so exhausted and sick of the thesis they just don't want to look at it again. But, after a deserved holiday they do need to force themselves back to it. A supportive and gently nagging supervisor can help too!

      Delete
  3. What are your thoughts of having a publication before beginning a PhD? I was discussing with a friend wanting to go into the PhD program and she seemed to think she needed one from her Honours. Personally I didn't think it was necessary although I had a publication from Honours when I went into the PhD program. Isn't the project outline and supervisor statement more important in Australia for scholarship competitiveness?

    ReplyDelete
  4. I've definitely found it challenging to start writing early. In particular, I'm always getting preoccupied with assignments and more short-term coursework, and sometimes this crowds out thesis work (I'm in the process of writing one up, so this is good timing!).

    I find it encouraging to hear your thoughts on having realistic expectations - I think that's one big factor which just makes the whole thing loom in front of me all the more.

    Also, I've definitely felt the pressure of not knowing enough science. What would be your recommendation in terms of the depth to which I should go, i.e. balancing the basics and boring technicalities? As an experimentalist, at least in honours, I tend to know the intuitive physics, but need to cover a lot more ground on the theory, and I find it hard to balance going into too much detail on basic things and knowing when I'm getting too technical.

    ReplyDelete
  5. Thanks for this Ross! They are all extremely insightful suggestions!

    Tihan, I'd suggest giving your thesis to someone who is in a different field if you're worried you are being too technical.

    ReplyDelete

A very effective Hamiltonian in nuclear physics

Atomic nuclei are complex quantum many-body systems. Effective theories have helped provide a better understanding of them. The best-known a...