How to find a research problem by Lindsay Nicholson

http://www.bris.ac.uk/cellmolmed/air/qanda/archive/how_to_find_a_problem.html

Finding a good research problem is tough thing to do. From a distance, it appears easy. There are so many things that we need to know: how to prevent childhood leukaemia; how to reverse global warming; how to maintain a stable economy; how to treat autoimmune disease. Then as you approach these questions more closely, they reveal themselves to be too large to tackle in one go: They have to be broken down.

When you do this, you suddenly find yourself with the choice of thousands of small problems. Now, you may be able to figure out ways to investigate these fragments, but you have to choose which is the most important.