How Do You Support Completion of Graduate Degrees and Engender Commitment to a Research Career?


Until they see the data, many departments don’t realize their rate of graduate attrition. Recent data released by the Council for Graduate Studies PhD Completion project shows that mathematics and physical sciences, the category in which they place computer science, had the highest attrition rate of all doctoral programs for three cohorts beginning in the 1990s. In Computer and Information Sciences, only about 28% of PhD students had completed their degrees by the sixth year (see Graph); only about 13% more complete by the end of the tenth year. The later the loss of a student, the larger the loss to the department, the advisor, and to the field.

Admission of a graduate student is an investment in a knowledge producer and a junior colleague. Graduate students who simply don’t have the ability are most likely to leave in the first year (accounting for one-third of all attrition). But when graduate students leave after the first year, a department loses not only the funding paid to that graduate student, but also the opportunity cost of space in classes and labs, a member of a research community, the resources put into mentoring and teaching, and possibly, the loss to other institutions. Likewise, graduate students lose self-esteem, as well as time and resources invested.



Advisors interested in building and maintaining the intellectual tradition of their field know that training future experts — graduate students — is the means of accomplishing this goal. Building the legacy requires viewing graduate students as junior colleagues and ushering them into the field, ensuring that they build the research, teaching, and professional skills of a successful future knowledge producer. Graduate development is decentralized, varying by department, discipline, and advisor. But departments can ensure that expectations for performance and milestone completion are clear; implement and act on substantive monitoring systems for both student and advisor; enforce written policies for both students and faculty; engender graduate integration into an intellectual community; and provide credible resources for supporting advisors in improving their advising efforts.


Studies show that some students have mismatched expectations of graduate school. Without regard to expectations, however, Lovitts and Nelson (2000) found that women and underrepresented minority graduate students are more likely to leave their programs than men, and they leave with high grade point averages. Ulku-Steiner, Kurtz-Costes, and Kinlaw (2000) found that both male and female doctoral students in male-dominated programs had lower academic self-concept and commitment to career than did students in gender-balanced programs, but women’s academic self-concept and commitment to career was lower than men’s in either male-dominated or gender-balanced programs. Students most likely to complete their graduate studies are those who are viewed as junior colleagues in a positive relationship with their advisors and who are well integrated into their department’s or lab’s intellectual community.


  • Council of Graduate Schools PhD Completion Project:
  • Golde, C. G. & Walker, G. E. (Eds.) (2006) Envisioning the Future of Doctoral Education: Preparing Stewards of the Discipline - Carnegie Essays on the Doctorate. Jossey-Bass Publishers.
  • Lovitts, B. E., & Nelson, C. (2000). The hidden crisis in graduate education: Attrition from Ph.D. programs. Academe.
  • Nettles, M., & Millett, C. (2006). Three magic letters: Getting to PhD. Baltimore: The Johns Hopkins University Press.
  • Ulku-Steiner, B., Kurtz-Costes, B., & Kinlaw, C. R. (2000). Doctoral student experiences in gender-balanced and male-dominated graduate programs. Journal of Educational Psychology.

View related case study:


Authors: Lecia Barker