Ten tips for the successful research novice

DISCLAIMER. Research is tough. Doing research for the first time is very tough; research is hardly ever exercised during undergraduate studies. It takes most graduate students by surprise.

Following are ten tips devised for graduate students who seek my supervision; others may also find them of interest. They should help you getting started. These advice do not guarantee your research will be successful but hopefully they will help you to sidestep some of the most obvious traps. So be warned: this list is neither complete nor consistent!

TIPS

1. Do it for yourself
2. Trust yourself!
3. Find a supervisor who is right for you
4. Don't underestimate your task
5. Set specific goals/milestones
6. ... and change them if necessary
7. Take risks
8. Write well
9. Stand on the shoulders of giants
10. Balance work with play

BIBLIOGRAPHY

Links:

• Teaching page, AH Eden
• Notes for prospective graduate students
• PhD in software design: research ideas

Good luck!

1. Do it for yourself

If you've enrolled for graduate studies only to improve your employability or purely to satisfy your mother's expectations—if you're not enjoying your research—you're wasting my very precious time (and yours). I do not wish to supervise a student unless s/he is primarily driven by genuine thirst for knowledge.

2. Trust yourself!

You can do it! You were admitted as a research student for a reason.

Don't get disheartened if things don't turn out the way you expected them. This is the nature of research.

3. Find a supervisor who is right for you

Choosing the right supervisor is by far the most important parameter in the success of a research student. There is no "ideal" supervisor, except the one who is right for you.

4. Don't underestimate the difficulty of your task

Don't expect miracles. Every research project involves dreary and unglamorous labour, including studying, programming, debugging, fishing the library for literature and reading some difficult, no matter how exciting, papers.

Be patient. The fable of Einstein flunking school is a lie and a poorly veiled excuse for laziness. Scholarship is gained by work, not by divine forces.

If English is not your native tongue, don't underestimate the complexity of writing your dissertation in English. Plan ahead your report, possibly by subdividing it into sections with appropriate headings and by setting milestones in writing each section (see commandments #5, #6, and #7), leaving wide error margins.

5. Set specific goals and milestones

Plan ahead, part I. If you don't know where you're going, you will never get 'there'. Set yourself specific objectives. Consult your supervisor to ensure they are realistic.

Plan ahead, part II. Schedule specific milestones in achieving your objectives, in particular as constrained by the schedule imposed by your degree scheme, and make a genuine attempt to follow them through. Remember Murphy's law ("anything that can go wrong, will") and Hofstadter's law ("It always takes longer than you expect, even when you take Hofstadter's Law into account.") So plan realistically, leaving wide margins for error.

6. ... and change them if necessary

If you change the direction of your research during its course, you're taking a risk. Taking a risk may be the only way to succeed. Taking unnecessary or too many risks will get in the way of completing your work. Consult your advisor before you take such a leap.

7. Take risks

Let your intuition lead you. Follow your instincts, particularly if doing so requires a stretch of the imagination. If you sense an insight, you may be unto something that is worth your while. The bolder is your idea, the greater is the risk (and possible benefits).

Doctoral education in the sciences must emphasize the personality, character, habits of heart and mind, and general scholarly dispositions of the steward of the discipline. Doctoral programs must ask how they can encourage risk taking and intellectual adventurousness while fostering the importance of precision and rigor. For rigor must not be permitted to dominate the personality of a future investigator so that the speculative and conjectural courage needed to do good science is destroyed. [Doctoral] Programs must model, practice, and reward risk taking.

We wish to educate a generation of scholars who are considerate, egalitarian, democratic, modest, and so on. Yet in my opinion, the greatest deficiency a young scholar can have is being so cautious that scientific endeavor is paralyzed. [Elkana 2006]

8. Write well

Write with care, coherently, concisely. Prove every claim you make or cite the appropriate reference (see APA Citation Format).

Avoid common fallacies, most commonly the argument of authority (see: Guide to logical Fallacies). It is not enough to convince yourself in your arguments, you must also convince your reader.

Set out to clarify the main theme of your paper/report/dissertation: What is your thesis? What evidence corroborate it? Identify context, including terminology and underlying assumptions (such as object-oriented-programming, software science) and let your reader know where have you borrowed your vocabulary from (using citations if possible).

Be brief. Try this exercise: After writing your first draft, try writing the same in half as many words. Repeat as necessary.

Use plain English. Avoid ostentatious adjectives, trite phrases, public relations and the thesaurus syndrome.

For a good introduction to writing advice see: [Wilbers 2000].

9. Stand on the shoulders of giants

Don't reinvent the wheel. Do your homework and check the literature. Spend as long as it's necessary to read papers/books that are worth reading. Consult with your supervisor which ones are worth reading. Carry out a "fishing expedition" for bibliography. Use the library or the Internet for this purpose.

Use only reliable resources (such as Columbia Encyclopaedia and Wolfram Mathematical Dictionary). Anyone can publish a page on the Internet; and Wikipedia articles have the average lifespan of a fly. When in doubt, consult your supervisor before you pollute your mind with hogwash. Be selective in what you read, and reserve your reverence to works of real value.

10. Balance work with play

Balance your research with your personal life. Such balance is often a prerequisite for prolonged and continued progress. Certain "sacrifices" are necessary, others will merely hamper your progress.

In particular: Don't trade sleeping for studying. Take an afternoon naps whenever possible. You will only make things worse, because sleep deprivation has adverse effects on learning. See the following excerpt from New Scientist ["11 steps to a better brain", New Scientist, No. 2501, 28 May 2005, p. 28]:

SKIMPING on sleep does awful things to your brain. Planning, problem-solving, learning, concentration, working memory and alertness all take a hit. IQ scores tumble. "If you have been awake for 21 hours straight, your abilities are equivalent to someone who is legally drunk," says Sean Drummond from the University of California, San Diego. And you don't need to pull an all-nighter to suffer the effects: two or three late nights and early mornings on the trot have the same effect.

Luckily, it's reversible - and more. If you let someone who isn't sleep-deprived have an extra hour or two of shut-eye, they perform much better than normal on tasks requiring sustained attention, such taking an exam. And being able to concentrate harder has knock-on benefits for overall mental performance. "Attention is the base of a mental pyramid," says Drummond. "If you boost that, you can't help boosting everything above it."

These are not the only benefits of a decent night's sleep. Sleep is when your brain processes new memories, practises and hones new skills—and even solves problems. Say you're trying to master a new video game. Instead of grinding away into the small hours, you would be better off playing for a couple of hours, then going to bed. While you are asleep your brain will reactivate the circuits it was using as you learned the game, rehearse them, and then shunt the new memories into long-term storage. When you wake up, hey presto! You will be a better player. The same applies to other skills such as playing the piano, driving a car and, some researchers claim, memorising facts and figures. Even taking a nap after training can help, says Carlyle Smith of Trent University in Peterborough, Ontario.

There is also some evidence that sleep can help produce moments of problem-solving insight. The famous story about the Russian chemist Dmitri Mendeleev suddenly "getting" the periodic table in a dream after a day spent struggling with the problem is probably true. It seems that sleep somehow allows the brain to juggle new memories to produce flashes of creative insight. So if you want to have a eureka moment, stop racking your brains and get your head down.

Bibliography

Primary sources

• Elkana, Yehuda. “Unmasking Uncertainties and Embracing Contradictions: Graduate Education in the Sciences.” In: Envisioning the Future of Doctoral Education. San Francisco, CA: Jossey-Bass, 2006. 

  ... Most doctoral students in the sciences tend to internalize a sense of consensus in the sciences because the great majority of their supervisors, without much critical reflection, hold this notion. And this view leads supervisors to hold the notion that it is efficient, from the point of view of the economy of time, for the speed and depth of the training of the doctoral student, not to waste time on contradictions in the field. As a result, critical preparation for the possibility of new thinking at the very foundations is often absent from doctoral training. However, ... no scientific theory in any discipline has ever been complete.

• S. Wilbers. Keys to Great Writing. Cincinnati: Writer's Digest Books, 2000

Tools

• Zotero

Background reading

• "Truly great science is always ahead of its time" [E. Zangger, The Edge]