Successful PhD Defense!

I recently successfully defended my PhD. Yay! It was almost 3 hours, as there were many questions. There have since been many congratulations and questions from others.  The most common question is:

How did you complete a part-time PhD in 5 years?

This is a really good question. I had previously completed a part-time masters degree in the 1990’s which unfortunately took my 6 years to do. Doing any kind of independent research it’s easy to fall into a hole where you get side-tracked on something not important, over-work some code more than necessary, design a poor experiment, complete a task without being aware of prior work, and so on. Back when I started the PhD, I specifically made a list of things to avoid/improve so that I wouldn’t fall into the same trap as before.

  1. Meet with your supervisor frequently. It’s easy to have scheduling conflicts, but in the days of Skype, web meetings, Slack, etc., it’s pretty easy to reschedule and do live meetings. My supervisor and I both agreed on meeting at least once a month and we’d reschedule as needed so the meetings didn’t get missed. This is really important to to avoid the above pitfalls.
  2. Lots of small tasks instead of really big tasks. Decomposing a big research project into small tasks is a good idea regardless of the circumstances. However, when part-time, this is really important. Small tasks can be chunked into a weekend or two.
  3. Know your limitations. You’re not on campus, you don’t have the same access to resources, you don’t have the same access to big blocks of time. I would have liked to do a evaluation study, but it had more overhead (e.g. experiment design, ethics committee), less access to students, and it would have been a big task. Instead, I did a number of small surveys.
  4. Submit, submit, submit. Submit posters, talks, papers, journals and so on. The submission process means that you have to organize your ideas, perform some focused research, analyse results — all of which are good. Then, you get reviews. Sometimes these are disappointing rejections (I got a -3 on a 1-5 score range on one paper), but there are lots of good nuggets of useful information in each rejection.
  5. Workshops. Instead of really big conferences, workshops and side-conferences are a great venue to get feedback on work in progress. Workshop papers are smaller scale making it easier to do the work and write the paper rather than the big conference. The workshops also provide for a more collaborative environment to get feedback from your direct peers specifically interested in your topic, as opposed to the mega-conference where questions can be somewhat random. If you do a really good job on a workshop paper, you might get invited to submit to a journal too. Overall, I had 12 peer-reviewed publications during my PhD vs. 2 for my masters (which was longer duration).
  6. Solicit cross-disciplinary feedback. Likely whatever you’re working on has applications across domains or at least there are different constituents of stakeholders. Directly approach those different stakeholders and get their input. They have different viewpoints. In my case, I reached out to typographers and cartographers a couple years into my thesis; and both these groups helped identify significant gaps in my work. I might have been able to get away without their feedback since my thesis reviewers were not typographers nor cartographers, but it made for a much stronger, much more defensible thesis because I’d incorporated their feedback.
  7. Background. Too many papers that I review seem to be missing related relevant research. Google Scholar has made search through a lot of current peer review research relatively easy. But don’t stop there: there is likely older relevant research that can also be found: many of the world’s largest libaries and museums are online, old websites and old texts can be found on archive.org, and so on.
  8. Blog! I used the blog as means of forcing me to always write about something related to my research (Ahem, I did not always achieve one post per month). It’s great to get feedback from the Internet at-large and see what resonates across the Internet. I thought my posting about Pokemon would have more reposts than it did. I got more reposts on my discussion regarding 500 years of separation than expected.
  9. Time-outs. There are unplanned events that always occur and need to be accommodated. My wife’s step father passed away. My mom sold her house and downsized. I had pneumonia for a couple months. You have to take some time out, but then you need mechanisms to get started again so that you don’t lose momentum. Always having a paper submitted  somewhere means you’ll get a response. Having commitments such as supervisor meetings or blog posts to do gets you back on track.

In case it’s not obvious yet, rapid iteration with frequent feedback is at the core of almost all the above tasks. Essentially, it’s about putting in place mechanisms to keep you on track, guided and focused.  It worked fairly well for me so far – now I just need to do the “minor revisions” and keep focused on getting those done.

Advertisements

About richardbrath

Richard is a long time visualization designer and researcher. Professionally, I am one of the partners of Uncharted Software Inc. I am also pursuing a part-time PhD in data visualization at LSBU. The opinions on this blog are related to my personal interests in data visualization, particularly around research interests related to my PhD work- this blog is about exploratory aspects of data visualization not proven principles.
This entry was posted in Data Visualization and tagged . Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s