The paper examines UK PhD completion and withdrawal rates, in a competing risks framework, using the 1986 National Survey of 1980 Graduates. The statistical problem of thresholding of completion data is also addressed. We argue that our results suggest that there are problems with the use of PhD...