Once you have found, or have been asked to supervise, a research student, it is as well to think about the nature of supervision and the supervisor's role and also about the factors that are associated with a student's likely success.
A few years ago Dr Di Bills undertook a piece of research on UniSA, the full report of which can be found here: Characteristics associated with research degree student satisfaction, completion and attrition at the University of South Australia (PDF 893kb - opens in new window. Download Adobe Acrobat) which suggested that the following tended to impede a student's progress: studying part-time; a student working full-time; a student having difficulties in the use of English; and, a student not adhering to the agreed milestones and the individual schedule established with the supervisor for making progress towards completion.
There are now quite a number of books available on the craft of research degree supervision and details of some of them can be found here: Books on research degree supervision. Many of these are in the library and, unless you're studying the area of doctoral education in its own right, you probably won't need to do more than dip in and out of this literature. You will, however, find lots of useful hints and tips in it.
One of the key things to remember when starting supervision is that the expectations of doctoral students and their supervisors can be very different. If you're going to get one thing right at the start, that should be to clarify what your and their expectations are.