Biostatistics

Biostatistics involves the theory and application of statistical science to analyze public health problems and to further biomedical research. Biostatistics faculty include leaders in the development of statistical methods for clinical trials and observational studies, studies on the environment, genomics/genetics, and the decision sciences. The department’s research in statistical methods and interdisciplinary collaborations provide many opportunities for student participation.

Current departmental research on statistical and computing methods for observational studies and clinical trials includes survival analysis, missing-data problems, and causal inference. Other areas of investigation are environmental research (methods for longitudinal studies, analyses with incomplete data, and meta-analysis); statistical aspects of the study of AIDS and cancer; quantitative problems in health-risk analysis, technology assessment, and clinical decision making; statistical methodology in psychiatric research and in genetic studies; Bayesian statistics; statistical computing; statistical genetics; and computational biology.

Collaborative research activities include coordination of national and international clinical trials, participation in studies of potential environmental hazards, design of health surveys, evaluation of health interventions and medical technologies, and consultation with federal, state, and local agencies. Many of these collaborations involve biomedical scientists in other Harvard-affiliated institutions.

The programs offered by the Department of Biostatistics provide rigorous training in the development of methodology, collaboration, teaching, and consultation on a broad spectrum of health-related problems. The department prepares students for academic and private-sector research careers in the fields of biostatistics and health decision sciences. Recent graduates have assumed faculty posts at universities, as well as positions in research laboratories, federal government centers, pharmaceutical companies, and research -institutes.

Doctor of Philosophy (PhD)

The doctoral program in biostatistics is designed for those who have demonstrated both interest and ability in scholarly research. Qualified applicants may apply to the doc-toral program without a prior advanced degree.

of courses in probability theory and applications, statistical inference, and statistical methods. In addition, students must complete a selection of advanced coursework in biostatistics. These courses are chosen in consultation with the faculty advisor. Given the increasing reliance of statistical practice on computing technology, students are recommended to take one or more courses in statistical computing as part of their program. Courses in statistical genetics and computational biology can be included in the program. Detailed information about specific requirements and specialized tracks for the PhD degree is outlined in the Biostatistics Graduate Student Handbook.

Qualifying Examinations. In addition to coursework and residency requirements, other
formal requirements for the degree include passing of both a written and oral qualifying examination and the completion of a PhD dissertation. The written qualifying examination assesses the student’s background in probability and statistical theory and in applications. The oral qualifying examination assesses the student’s potential to perform research in a chosen field, and examines the student’s knowledge of his or her fields of study.

Dissertation. Each student is expected to complete a dissertation. The dissertation should be an original contribution to scientific knowledge in biostatistics. It can contribute to a subject matter field through innovative application of existing methodology, can produce an original methodologic contribution, or be a combination of the two. When the dissertation is complete, the student defends it to the Research Committee at a public presentation. The defense must be scheduled at least three weeks in advance. Copies of the dissertation should be given to members of the Research Committee and the department chair at least two weeks before the defense.