Measuring time to the doctorate: reinterpretation of the evidence.

There has been increasing concern that the length of time it takes to earn a doctorate in this country has increased dramatically over the last 20 years. The regularly cited evidence--organized by the year in which recipients of doctorates were awarded their degrees--is seriously misleading, however. The application of stable population theory to the problem suggests that the steady fall in the sizes of entering cohorts to graduate school has inflated both the measure of the absolute level of median time to degree and the increase in time to degree. When the same underlying data are reorganized by the year in which recipients of doctorates received their baccalaureate, the statistical bias is eliminated, and the median total time to degree in the humanities is shown to have risen 15-20% rather than the reported 40%.