PhD Students:

At the end of a PhD the graduate should have achieved the following skills:

• The ability to identify and address problems of interest to the research community.
• The ability to articulate these problems and solutions orally and in writing for audiences of all levels.
• The ability to read and understand advanced technical material in your area with ease.
• World class expertise in your chosen research topic.

 

General principles for getting there:

• Listen to instructions: everyone hears what they want to hear, unfortunately when your supervisor tells you something you don't want to hear, you still have to respond in a meaningful and timely way.
• Keep asking questions: even if they seem trivial. Do spend a little time trying to figure out the problem on your own, especially if your supervisor has explained it more than once. Just don't let it stop your progress.
• Experiments and understanding: If I ask you to try something out using existing code and data, I generally want a) a quick and dirty experiment so we know how to proceed, b) a time critical task completed for a funded project. In this case hack it together, don't take time out to read about the Philosophy of Mind before you get to the experiment. Often getting your hands dirty with some real data will greatly improve your understanding.
• Articulate your ideas: talk to your supervisor, labmates, other professors etc about your ideas. If they have trouble understanding you, try to refine your explanations so they "get it". Take every opportunity to give formal and informal presentations. If you attend a conference try to meet and talk to as many people as you can, and tell them your ideas. All the "hot" researchers are distinguished by their ability to communicate their ideas: in print, one on one and in formal presentations.
• Join the lab community: students who "prefer to work at home" generally have more difficulty developing good research skills than those who spend more time in the lab. Your fellow students can be great teachers.
• Read, Read, Read
• Get involved in projects: working on small problems can help you understand the real issues in the field better, give you new skills and in many cases can turn into a result that will contribute to publications and your dissertation.
• Most dissertations are not based on a single monolithic result: neither is one big result better than several incremental results that move the field forward. Don't just wait for "the big one", don't despise a small result, work toward interesting, publishable results in the projects you are involved in.

 

In order to complete your degree:

• You should have achieved the 4 skills listed above.
• You should have several medium sized results (typically published as tier 1 conference papers), or one large result, complete and approved by your supervisor.
• You should have at least one journal article published or under review.
• Give yourself 3 months to write up your results. Half the people in grad school think they are soooo articulate they can prepare their thesis in 2 weeks, but they can't.
• You should hand your supervisor a complete (but not final) draft of your thesis including results, at least 2 months before your defense date.
• Your defense date should be at least 3 months before the absolute final date you are willing to hand in your thesis to the library.
• Just because you've been offered a job in the kingdom of Far, Far Away, doesn't mean you get to skip the necessary steps to make your final submission acceptable.