A dissertation committee is the faculty panel that approves each step of a doctoral degree. Its signatures move a student from coursework to candidacy, then to the final defense. The Council of Graduate Schools ran one national tracking project. It covered 29 large research universities. In that sample, 57% of doctoral candidates finished a PhD within ten years. Rates varied by field.
That number is old. The cohorts date to the mid-2000s, before pandemic disruption and newer online programs. Its 2008 monograph reported a wide spread: 64% finishing in engineering against 49% in the humanities. Committee rules differ from campus to campus as well. Campus life at these schools has also been reshaped by the mechanics of graduate student unionization.
This explainer covers who serves, what gets approved, and how far the research holds up. Related explainers live in the site’s academia section.
Who Sits on a Dissertation Committee?
Rosters are small and fixed by rule. The University of Oregon’s Division of Graduate Studies sets a four-member minimum for PhD committees. At least two must come from the student’s own program. The policy names four roles: advisor, chair, core members, and an institutional representative.
Some fields split the leadership. In five Oregon programs, including biology and chemistry, the advisor and chair must be two different people. A late change has a cost. At Oregon, a new advisor resets a six-month clock before the defense. This connects to our earlier piece, One Advisor, Five Years, No Recourse: Academia's Most Concentrated Job.
What Does the Committee Approve Before the Defense?
The panel’s formal work runs through gates rather than one vote at the end. The roster itself sits under review. Oregon requires the program and the graduate division to confirm it at least six months before the defense. Petitions can override that rule when circumstances are unusual. Later gates include proposal review, written exams, and the oral defense of the finished manuscript.
The checkpoints exist because finishing, not admission, is the documented bottleneck. The council treated procedures as one lever among six. The others were selection, mentoring, funding, program climate, and the research norms of each field. That framing places the committee near the hinge of the process.
What Did the Completion Project Find?
The council ran the project in two phases with funds from Pfizer and the Ford Foundation. Twenty-nine major universities took part. It tested changes and reported results in a 2008 monograph. In the sample, 57% of doctoral candidates finished within ten years. Engineering sat at 64%, humanities at 49%.
A 2009 volume in the same series added exit surveys. They covered 1,406 new PhDs at 18 universities, gathered from March 2006 to August 2008. Graduates credited financial support first, with mentoring and advising close behind. Family support came next. A 2010 monograph then listed campus practices that schools linked to better odds of finishing. The list ran from admissions screening to program climate.
How Strong Is This Evidence?
- Tracking study: 29 campuses, ten-year windows, baseline published in 2008.
- Exit surveys: 1,406 completers at 18 campuses, self-reported, 2006 to 2008.
- Committee rules: published campus policies, such as Oregon’s four-member minimum and six-month clock.
The limits are clear. Cohorts from the mid-2000s predate newer funding models and pandemic-era disruption. Exit surveys capture only graduates, so the views of students who left are missing. These designs track and survey; they do not test cause. They can describe what accompanied success, not prove that committee practices caused it.
Conclusion: A Gatekeeper With a Paper Trail
The committee works as a governance body with votes, rosters, and deadlines. It is not a private advisory club. Its powers are formal and documented, from proposal sign-off to the final defense. The research on how panels shape outcomes is real but dated and self-reported. The local graduate school handbook remains the most current authority on what a committee can do.
