and the anonymous reviewers for candid and substantive feedback, support and encouragement over too many years.At least 40% of the students who begin a doctoral program fail to complete it. This remarkable fact is one of the hidden flaws of a system usually touted as the crown jewel of the American higher education system. American graduate education is generally regarded as the best in the world, and yet, at the turn of the century, many researchers, administrators, government agencies, foundations, professional associations, and other interested parties are casting critical eyes on graduate education in an effort to understand the ways in which the inherited system does and does not continue to function effectively (e.g.,

Little is known about the process of doctoral student attrition. Relying on theories of socialization and integration, this paper examines the cases of three students who left their doctoral programs. Their experiences show that academic integration pla ys a greater role in attrition than social integration. Furthermore, the academic integration process can be ruptured, particularly when a student-advisor relationship ends. The paper also questions the idea that degree completion equals success and tha t attrition equals failure.

A graduate student's decision not to continue in an initial program is that student's answer to one or more questions.

No abstract

A satisfactory relationship between doctoral students and their advisors is an essential component of successful doctoral training. Using responses to a national survey of doctoral students in the US from 27 universities and 11 disciplines, this paper explores factors affecting students' satisfaction with the advising relationship. We find that both the criteria used in selecting an advisor and reported advisor behaviours influence satisfaction. Moreover, there are pronounced disciplinary differences in both choice criteria and advisor behaviour, and these are more robust predictors of satisfaction than individual characteristics.