a study of ph.d. completion at duke university

Jump to first page

A STUDY OF Ph.D. COMPLETION AT DUKE UNIVERSITY

Lewis M. SiegelVice Provost and Dean of the Graduate School, Duke University

DUKE UNIVERSITY GRADUATE SCHOOL

Jump to first page

THE DATA TO BE PRESENTED

Ph.D. cohorts from Fall 1991 through Fall 1995 were examined for each Duke degree granting program for % Completion (as of Fall 2004) Patterns of withdrawal from Ph.D. program Median Time to Degree Effect of Variables such as GRE scores, GPA, Race,

Gender, and Merit Fellowship selection Following Some Interventions, Attrition Rates Through

5th Year Compared for 1996-99 and 1991-95 Cohorts


Jump to first page

Ph.D. COMPLETION RATES MIXED IN HUMANITIES AND SOCIAL SCIENCES; TIME TO DEGREE GENERALLY HIGH, AND LATE ATTRITION A PROBLEM IN SEVERAL LARGE DEPARTMENTS

No. Percent Median MedianProgram Matrics Completion TTD TTW

Art History 10 90% 7.7 2.7Religion 61 74% 6.3 3.3Literature 48 63% 7.3 4.8Music 32 63% 5.8 1.5English 73 60% 6.7 4.0Romance Studies 45 60% 6.7 1.9Philosophy 25 52% 6.7 1.3German 14 50% 7.7 2.3Classics 23 43% 5.7 1.7Slavics 13 38% 6.3 2.8

HUMANITIES 344 61% 6.7 2.7

Psychology* 66 85% 4.7 1.1Business Administration 45 69% 5.3 1.2History 91 64% 7.7 4.3Cultural Anthropology 30 60% 6.3 1.9Political Science 65 57% 7.3 4.9Sociology 43 53% 6.3 2.9Economics 110 49% 5.0 2.1

SOCIAL SCIENCES 450 62% 6.0 2.7

TOTAL GRAD SCHOOL 1843 64% 5.7 2.3


Jump to first page

Ph.D. COMPLETION RATES HIGH AND ATTRITION EARLY IN MOST BIOLOGICAL SCIENCES; BIOLOGICAL ANTHROPOLOGY IS AN EXCEPTION


Molecular Cancer Biology 17 88% 5,3 2.2Genetics*** 14 86% 5.3 2.6Neurobiology 33 85% 6.2 1.0Pathology 21 81% 4.3 2.4Microbiology 28 79% 6.3 3.3Cell Biology**** 72 76% 5.0 1.0Pharmacology 41 76% 5.7 2.7Environment 48 71% 5.0 2.5Biology** 103 70% 5.7 2.5Biochemistry 56 66% 5.3 2.3Immunology 19 63% 5.0 3.0Biological Anthropology 19 53% 7.5 4.3

BIOLOGICAL SCIENCES 471 73% 5.5 2.3



Jump to first page

Ph.D. COMPLETION RATES IN PHYSICAL SCIENCES AND ENGINEERING MIXED, BUT TIME TO DEGREE IS LOW AND WITHDRAWAL GENERALLY EARLY


Chemistry 103 65% 4.7 2.0Statistics 28 64% 4.0 2.0Physics 63 60% 5.7 1.9Mathematics 43 58% 5.0 1.8Computer Science 52 54% 5.7 2.7Earth & Ocean Sciences 30 47% 5.3 2.3

PHYSICAL SCIENCES 319 60% 5.0 2.1

Biomedical Engineering 67 72% 5.3 2.0Mechanical Engineering 70 70% 4.3 1.7Electrical & Computer Eng'g 70 51% 4.7 2.2Civil & Environmental Eng'g 52 42% 5.3 2.3

ENGINEERING 259 60% 4.9 2.1



Jump to first page

GENERAL FINDINGS Completion Rates are highest in the Biological Sciences

(74%) and similar in the Humanities, Social Sciences, Physical Sciences, and Engineering (60-62%).

Both the Median Time to Degree and Time to Withdrawal are greatest in the Humanities (6.7 and 2.7 years, respectively) and Social Sciences (6.0 and 2.7 years) and shortest in the Physical Sciences and Engineering (about 5.0 and 2.1 years).

In individual programs, there is little correlation between Completion rate, Time to Degree, or Time to Withdrawal.

It is generally felt that late withdrawal from a Ph.D. program (e.g., after 3 or 4 years) is more serious than early withdrawal (e.g., in year 1 or 2).


Jump to first page

COMPLETION RATES FOR DIFFERENT GROUPS

High GRE (V and Q) and high GPAs correlate well with completion only in the Humanities.

Duke’s African-American Ph.D. students complete the degree at higher rates than the general popula-tion in all fields except the Physical Sciences and Engineering (where the numbers are very low).

There are striking differences in gender completion rates in the Humanities and Biological Sciences (more male than female), but the effect is opposite in the Social Sciences (more female than male). Overall, there is little net effect of gender.

Duke faculty can pick students who are more likely to complete than the general population based on reading applications. In nearly all fields, J.B.Duke fellowship awardees complete more than others with similar high GRE scores and GPA’s.

Jump to first page

EFFECT OF GPA/GRE SCORES, J.B. DUKE FELLOWSHIP SELECTION, RACE, AND GENDER ON Ph.D. COMPLETION RATES (1991-95 Cohorts)

Type of Matriculant Human Soc Sci Biol Sci Phys Sci Eng'g Grad Sch

All matriculants 61% 62% 73% 60% 60% 64%

Low GPA (3.2-) 54% 65% 75% 64% 54% 63%High GPA (3.7+) 65% 56% 71% 63% 67% 64%

Low GRE-V (590-) 54% 64% 75% 60% 60% 64%High GRE-V (710+) 62% 54% 65% 58% 60% 60%

Low GRE-Q (590-) 59% 60% 69% --- --- 63%High GRE-Q (710+) 69% 60% 72% 59% 60% 64%

J.B.Duke Fellows 76% 61% 78% 72% 66% 73%

African-American 70% 75% 83% 22% 25% 64%

Males 67% 58% 76% 60% 60% 65%Females 56% 64% 67% 59% 59% 62%


Jump to first page

AFRICAN-AMERICAN Ph.D. COMPLETION AND TIMES TO DEGREE OR WITHDRAWAL (1991-95 Cohorts)

Program No. % Median Median % Withdrawn in YearMatrics. Compl. TTD TTW 1 to 3 4 to 5 6+

HUMANITIESTotal matrics 344 61% 6.7 2.7 22% 3% 9%African-American 20 70% 6.7 4.5 10% 5% 15%

SOCIAL SCIENCESTotal matrics 450 60% 6.0 2.7 22% 5% 9%African-American 32 75% 6.3 2.8 16% 0% 9%

BIOLOGICAL SCIENCESTotal matrics 471 73% 5.5 2.3 18% 6% 2%African-American 12 83% 5.7 5.7 0% 8% 8%

PHYSICAL SCIENCESTotal matrics 319 60% 5.0 2.1 30% 8% 3%African-American 9 22% 4.7 2.0 78% 0% 0%

ENGINEERINGTotal matrics 259 60% 4.9 2.1 29% 7% 4%African-American 8 25% 4.3 2.0 50% 13% 13%

TOTAL GRADUATE SCHOOLTotal matrics 1830 64% 5.7 2.3 24% 6% 6%African-American 81 64% 6.3 2.7 22% 2% 9%


Jump to first page

CUMULATIVE PERCENT OF WITHDRAWALS FROM Ph.D. PROGRAM BY YEAR

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

1 2 3 4 5 6 7 8 9 10

Year of Ph.D. program

% o

f Tot

al W

ithdr

awal

s fro

m P

h.D

. Pro

gram

Total Graduate School EngineeringPhysical Sciences Biological SciencesSocial Sciences Humanities


Jump to first page

PERCENT OF WITHDRAWALS FROM Ph.D. PROGRAM BY YEAR

0%

5%

10%

15%

20%

25%

30%

35%

1 2 3 4 5 6 7 8 9 10


% o

f Tot

al W

ithdr

awal

s in

Giv

en Y

ear

Total Graduate School EngineeringPhysical Sciences Biological SciencesSocial Sciences Humanities


Jump to first page

Ph.D. COMPLETION RATES AND WITHDRAWAL PATTERNS: Attrition High and Often Late in Many Humanities and Social Science Departments

No. Median % Matrics Withdrawn in YearProgram Matrics TTW 1 to 3 4 to 5 6+

Philosophy 25 1.3 24% 0% 12%Music 32 1.5 28% 0% 9%Classics 23 1.7 48% 0% 9%Romance Studies 45 1.9 33% 2% 4%German 14 2.3 36% 7% 0%Art History 10 2.7 10% 0% 0%^Slavics 13 2.8 38% 0% 15%Religion 61 3.3 10% 3% 10%English 73 4.0 16% 7% 15%Literature 48 4.8 8% 4% 13%

HUMANITIES 344 2.7 22% 3% 9%

Psychology* 66 1.1 12% 2% 2%Business Administration 45 1.2 31% 0% 0%Cultural Anthropology 30 1.9 23% 10% 3%Economics 110 2.1 34% 11% 5%Sociology 43 2.9 23% 7% 12%History 91 4.3 15% 4% 15%Political Science 65 4.9 18% 3% 20%

SOCIAL SCIENCES 450 2.7 22% 5% 9%

TOTAL GRAD SCHOOL 1843 2.3 23% 6% 6%


Jump to first page

ROMANCE STUDIES: PERCENT OF WITHDRAWALS FROM Ph.D. PROGRAM BY

YEAR

0%

5%

10%

15%

20%

25%

30%

35%

40%

45%

50%

1 2 3 4 5 6 7 8 9 10


% o

f Tot

al W

ithdr

awal

s in

Giv

en Y

ear

Romance StudiesHumanites

MTW = 1.88


Jump to first page

ENGLISH: PERCENT OF WITHDRAWALS FROM Ph.D. PROGRAM

BY YEAR

0%

5%

10%

15%

20%

25%

30%

1 2 3 4 5 6 7 8 9 10


% o

f Tot

al W

ithdr

awal

s in

Giv

en Y

ear

English Humanites

MTW = 4.00


Jump to first page

ECONOMICS: PERCENT OF WITHDRAWALS FROM Ph.D. PROGRAM

BY YEAR

0%

5%

10%

15%

20%

25%

30%

35%

1 2 3 4 5 6 7 8 9 10


% o

f Tot

al W

ithdr

awal

s in

Giv

en Y

ear

Economics Social Sciences

MTW = 2.08


Jump to first page

HISTORY: PERCENT OF WITHDRAWALS FROM Ph.D. PROGRAM

BY YEAR

0%

5%

10%

15%

20%

25%

30%

1 2 3 4 5 6 7 8 9 10


% o

f Tot

al W

ithdr

awal

s in

Giv

en Y

ear

History Social Sciences

MTW = 4.33


Jump to first page

Ph.D. COMPLETION RATES AND WITHDRAWAL PATTERNS: Biological Science Attrition Low and Generally Early; Biological Anthropology an Exception


Neurobiology 33 1.0 15% 0% 0%Cell Biology**** 72 1.0 19% 1% 1%Molecular Cancer Biology 17 2.2 6% 6% 0%Biochemistry 56 2.3 25% 9% 0%Pathology 21 2.4 10% 10% 0%Environment 48 2.5 19% 8% 2%Biology** 103 2.5 23% 3% 4%Genetics*** 14 2.6 7% 7% 0%Pharmacology 41 2.7 15% 7% 0%Immunology 19 3.0 21% 11% 5%Microbiology 28 3.3 11% 7% 4%Biological Anthropology 19 4.3 16% 11% 16%

BIOLOGICAL SCIENCES 471 2.3 18% 6% 2%



Jump to first page

BIOLOGY: PERCENT OF WITHDRAWALS FROM Ph.D. PROGRAM

BY YEAR

0%

5%

10%

15%

20%

25%

30%

35%

40%

1 2 3 4 5 6 7 8 9 10


% o

f Tot

al W

ithdr

awal

s in

Giv

en Y

ear

BiologyBiological Sciences

MTTW = 2.5 yr


Jump to first page

BIOLOGICAL ANTHROPOLOGY: PERCENT OF WITHDRAWALS FROM Ph.D. PROGRAM

BY YEAR

0%

5%

10%

15%

20%

25%

30%

35%

40%

1 2 3 4 5 6 7 8 9 10


% o

f Tot

al W

ithdr

awal

s in

Giv

en Y

ear

Biological AnthropologyBiological Sciences

MTW = 4.33


Jump to first page

Ph.D. COMPLETION RATES AND WITHDRAWAL PATTERNS: Physical Sciences and EngineeringAttrition Generally by Year 3; Computer Sci an Exception


Mathematics 43 1.8 33% 7% 2%Physics 63 1.9 32% 6% 3%Chemistry 103 2.0 27% 7% 1%Statistics 28 2.0 29% 0% 7%Earth & Ocean Sciences 30 2.3 43% 10% 0%Computer Science 52 2.7 23% 13% 10%

PHYSICAL SCIENCES 319 2.1 30% 8% 3%

Mechanical Engineering 70 1.7 24% 3% 3%Biomedical Engineering 67 2.0 16% 10% 1%Electrical & Computer Eng'g 70 2.2 34% 9% 6%Civil & Environmental Eng'g 52 2.3 42% 8% 6%

ENGINEERING 259 2.1 29% 7% 4%



Jump to first page

CHEMISTRY: PERCENT OF WITHDRAWALS FROM Ph.D. PROGRAM

BY YEAR

0%

5%

10%

15%

20%

25%

30%

35%

1 2 3 4 5 6 7 8 9 10


% o

f Tot

al W

ithdr

awal

s in

Giv

en Y

ear

ChemistryPhysical Sciences

MTW = 2.00


Jump to first page

COMPUTER SCIENCE: PERCENT OF WITHDRAWALS FROM Ph.D. PROGRAM BY

YEAR

0%

5%

10%

15%

20%

25%

30%

1 2 3 4 5 6 7 8 9 10


% o

f Tot

al W

ithdr

awal

s in

Giv

en Y

ear

Computer SciencePhysical Sciences

MTW = 2.67


Jump to first page

ELECTRICAL & COMPUTER ENGINEERING: PERCENT OF WITHDRAWALS FROM Ph.D.

PROGRAM BY YEAR

0%

5%

10%

15%

20%

25%

30%

35%

40%

1 2 3 4 5 6 7 8 9 10


% o

f Tot

al W

ithdr

awal

s in

Giv

en Y

ear

Electrical & Computer Engineering

Engineering

MTW = 2.17


Jump to first page

BIOMEDICAL ENGINEERING: PERCENT OF WITHDRAWALS FROM Ph.D. PROGRAM BY

YEAR

0%

5%

10%

15%

20%

25%

30%

35%

40%

1 2 3 4 5 6 7 8 9 10


% o

f Tot

al W

ithdr

awal

s in

Giv

en Y

ear

Biomedical EngineeringEngineering

MTW = 2.00


Jump to first page

INTERVENTIONS TO IMPROVE Ph.D. COMPLETION RATE AT DUKE AFTER 1995: STUDENT FUNDING

Institute 5-6 year funding guarantee in all A&S units; generally support to degree in sciences

Departments given budgets based on graduate education parameters, not service needs

3-year budgets to deal with variations in yield Reduction in student teaching loads; more fellowship

years Competitive summer research support introduced Funded capstone teaching fellowships at dissertation

stage


Jump to first page

INTERVENTIONS TO IMPROVE Ph.D. COMPLETION RATE AT DUKE AFTER 1995: BETTER INFORMED SELECTION

Reduce emphasis on GRE scores and GPA’s – data show poor correlation with completion as Duke uses them

Emphasize demonstrated research experience Transparency: Put Duke data on placement, time to

degree, completion rates in each program – student should know what (s)he is getting into

Interview students before admitting them – campus visits and/or by telephone with international students

Exchange information to learn “fit” rather than just sell program


Jump to first page

INTERVENTIONS TO IMPROVE Ph.D. COMPLETION RATE AT DUKE AFTER 1995: IMPROVED PROGRAMS

Require structured teaching experiences for graduate TA’s where career-appropriate

Graduate school workshops in pedagogy and uses of instructional technology; expanded PFF program; Pathways to the Professoriate

Responsible conduct of research training extended to all Ph.D. students in all disciplines

Field-specific symposia to introduce career options to Ph.D. students

Enhanced student services; strong investment in subsidizing child care for graduate students

“Best Practices” charter for faculty, students, Grad School


Jump to first page

EFFECT OF INTERVENTIONS ON COMPLETION – AN EXPERIMENT THAT HAS NO CONTROLS

Compare attrition in first 5 years for Fall 1996-99 vs. Fall 1991-95 cohorts

Significant reduction in attrition in disciplines where funding change is greatest: Humanities 5-year attrition reduced from 25% to 18%. Effect strongest in Language instruction departments where reduction in required service greatest.

Results clouded by “housecleaning” in certain departments with new leadership (History, Sociology)

Virtually no overall effect on Biological Sciences, where attrition already lowest and funding with choice already in place

Reduced attrition in the most TA-dependent Physical Science departments (Chemistry, Math)

Attrition worsens in “tub on own bottom” units (Engineering, Environment) that resist following Graduate School’s new budgeting system


Jump to first page

INTERVENTION: EFFECT ON “EARLY” ATTRITIONIN DUKE GRADUATE SCHOOL Ph.D. PROGRAMS

For the Graduate School as a whole, there is no effect

1991-95 Cohorts 1996-99 Cohorts %Matrics Withdrawn in Year %Matrics Withdrawn in Year

Program No. Matrics 1 to 3 4 to 5 Tot Yrs 1-5 No. Matrics 1 to 3 4 to 5 Tot Yrs 1-5

GRAD SCHOOL 1830 24% 6% 29% 1378 23% 6% 29%


Jump to first page

YEAR 1-5 ATTRITION AFTER “INTERVENTION”: Significant improvement in Humanities, particularly in highly T.A.-dependent Language & Literature departments



Slavics 13 38% 0% 38% 2 0% 0% 0%Art History 10 10% 0% 10% 10 10% 0% 10%Religion 61 10% 3% 13% 38 8% 5% 13%Literature 48 8% 4% 13% 30 10% 3% 13%Classics 23 48% 0% 48% 14 14% 0% 14%English 73 16% 7% 24% 42 14% 5% 19%Philosophy 25 24% 0% 24% 14 21% 0% 21%Romance Studies 45 33% 2% 36% 30 17% 7% 23%Music 32 28% 0% 28% 19 21% 5% 26%German 14 36% 7% 43% 11 0% 27% 27%

HUMANITIES 344 22% 3% 25% 210 13% 5% 18%

GRAD SCHOOL 1843 23% 6% 29% 1380 23% 6% 29%


Jump to first page

YEAR 1-5 ATTRITION AFTER “INTERVENTION”: Mixed results in Social Sciences: Stable units generally improve, but some problem units with large populations start to “clean house”.



Business Administration 45 31% 0% 31% 36 3% 0% 3%Cultural Anthropology 30 23% 10% 33% 14 7% 0% 7%Psychology* 66 12% 2% 14% 48 13% 2% 15%Political Science 65 18% 3% 22% 52 12% 6% 17%History 91 15% 4% 20% 39 23% 8% 31%Sociology 43 23% 7% 30% 31 23% 23% 45%Economics 110 34% 11% 45% 64 44% 2% 45%

SOCIAL SCIENCES 450 22% 5% 28% 284 20% 5% 26%

GRAD SCHOOL 1843 23% 6% 29% 1380 23% 6% 29%


Jump to first page

YEAR 1-5 ATTRITION AFTER “INTERVENTION”: Little overall change in Biological Sciences, where full funding model and many Best Practices already were in placeSome improvement in highly T.A.-dependent units



Genetics*** 14 7% 7% 14% 26 8% 4% 12%Biochemistry 56 25% 9% 34% 48 19% 0% 19%Cell Biology**** 72 19% 1% 21% 31 16% 3% 19%Pathology 21 10% 5% 15% 16 6% 13% 19%Biological Anthropology 19 16% 11% 26% 14 7% 14% 21%Biology** 103 23% 3% 27% 82 15% 10% 24%Pharmacology 41 15% 7% 22% 41 24% 0% 24%Molecular Cancer Biology 17 6% 6% 12% 29 21% 7% 28%Neurobiology 33 15% 0% 15% 19 26% 5% 32%Microbiology 28 11% 7% 18% 12 25% 8% 33%Environment^ 48 19% 8% 27% 60 25% 8% 33%Immunology 19 21% 11% 32% 16 31% 6% 38%

BIOLOGICAL SCI 471 18% 6% 24% 394 19% 6% 25%

GRAD SCHOOL 1843 23% 6% 29% 1380 23% 6% 29%


Jump to first page

YEAR 1-5 ATTRITION AFTER “INTERVENTION”: Highly T.A.-dependent Physical Science depts generally improve;“Tub on own bottom” depts in Engineering, EOS get worse



Chemistry 103 27% 7% 34% 85 21% 4% 25%Statistics 28 29% 0% 29% 27 26% 0% 26%Mathematics 43 34% 7% 40% 28 29% 4% 32%Physics 63 32% 6% 38% 46 30% 4% 35%Computer Science 52 23% 13% 37% 45 27% 13% 40%Earth & Ocean Sciences 30 43% 10% 53% 18 50% 11% 61%

PHYSICAL SCIENCES 319 30% 8% 38% 249 27% 6% 33%

Biomedical Engineering 67 16% 10% 27% 72 13% 8% 21%Mechanical Engineering 70 24% 3% 27% 46 37% 4% 41%Electrical & Computer Eng 70 34% 9% 43% 87 46% 5% 51%Civil & Environmental Eng 52 42% 8% 50% 38 63% 5% 68%

ENGINEERING 259 29% 7% 36% 243 37% 6% 43%

GRAD SCHOOL 1843 23% 6% 29% 1380 23% 6% 29%


Jump to first page

ROUNDTABLE DISCUSSIONS HELD TO DESIGN AND SHARE BEST PRACTICES ACROSS DISCIPLINES

Graduate School distributes published material on issues concerning Ph.D. attrition; invites guest speakers; summarizes current thinking about factors that might increase attrition

Graduate School sends questions to departments: Explain and justfy your processes for selection, recruitment, orientation, integration, examination, mentoring, and teacher training of graduate students in the context of improving your Ph.D. completion rate

Dean links response to questions to the annual budget meetings with departments– gets their attention

Graduate School engages in dialogue with individual DGS’s and student representatives to refine best practice ideas, then distributes them as basis for discussions at dinners for DGS’s

Graduate School publishes report on best practices to promote sharing of ideas across disciplines----“Now I understand why we have a Graduate School!”


Jump to first page

INTERVENTIONS TO IMPROVE Ph.D. COMPLETION RATE AT DUKE AFTER 2002---ROUNDTABLE INPUTS

Comprehensive orientation programs by both school and department – introduce program faculty, history and issues in field, possible career options, opportunities to improve skills, introduction to field methodologies

Distribute and discuss clear program guidelines and expectations, with milestone achievement markers

“Best Practices” principles to be discussed in departments by faculty and students together

Make sure there is an interested first year advisor or multiple advisors, often graduate “buddy” assigned

Strongly encourage freedom of mentor choice; rotations if possible – first year funding free of mentor if possible


Jump to first page

INTERVENTIONS TO IMPROVE Ph.D. COMPLETION RATE AT DUKE AFTER 2002---ROUNDTABLE INPUTS Encourage first or second year research projects to be

completed as a group collaborative effort Regularly scheduled seminar and faculty/student

luncheon series which include faculty and advanced students

Clarify faculty expectations on qualifying and preliminary examinations; hold all students to same standard

Ensure that dissertation prospectus meeting/exam is held no later than six months after prelim. exam

Graduate School to publicize examples of excellence in mentoring; give annual faculty mentor awards

Require bi- or annual reports to DGS and committee on student’s progress; “no student gets lost”


Jump to first page

INCREASING Ph.D. COMPLETION---A WORK IN PROGRESS

It will probably never be possible to select students who are perfectly suited for graduate study in a particular unit or school – ETS trying to measure “non-cognitive” factors

Better GRE exams may help – Subject tests a largely unused resource for selection of students

Axiom: transparency can only help matters – Is the long road to a Ph.D. right for you? Is this the right “fit”?

What is an “acceptable” level of attrition, given the large financial investment the university usually makes in each new Ph.D. matriculant?

Given the long time it takes to obtain a Ph.D., how we will know that any intervention we undertake is actually having an effect?


a study of ph.d. completion at duke university

Documents