Director: Ian McKeague, PhD
The Doctor of Philosophy (PhD) in Biostatistics prepares candidates for leadership roles in the development and application of statistical methods to biomedical research for the advancement of public health. The program requirements for the PhD degree differ from those for the Doctor of Public Health (DrPH) in that the curriculum, examinations, and dissertation involve more emphasis on statistical theory in the context of public health applications.
The PhD in Biostatistics is awarded by the Graduate School of Arts and Sciences (GSAS) of Columbia University as governed by the Doctoral Program Subcommittee on Biostatistics. The program is administered by the faculty and staff of the Mailman School of Public Health.
Building on the foundation of the MS in Theory and Methods (MS/TM), the PhD includes the completion of any MS/TM requirements not previously met in the student’s prior master’s program, the equivalent of four semesters of additional full-time study, written and oral comprehensive examinations, and the completion and oral defense of an independent, original dissertation on a problem in statistical theory or method with relevance to a biomedical or public health application. The typical time for completion of the PhD degree is four to five years- two years for course work and written qualifiers and two to three years for the dissertation.
While many of the applicants admitted to Columbia’s PhD program in biostatistics have already completed (or are completing) master’s degrees in biostatistics, statistics, or a related field, admission is open to well qualified students holding (or completing) bachelor’s degrees. Those admitted with a bachelor’s degree are typically strong students from programs that emphasize a rigorous background in mathematics and/or statistics.
Depending on prior training and background, students may be required to take additional master’s level course work in the Mailman School of Public Health as part of their PhD training.
In addition to the requirements listed below, all applicants must submit an official transcript from each prior institution, a statement of academic purpose, and three letters of evaluation from academic sources. All international students whose native language is not English or whose undergraduate degree is from an institution in a country whose official language is not English must submit Test of English as a Foreign Language (TOEFL) or IELTS scores.
|Deadline for Fall Admission||December 1|
|Deadline for Spring Admission||No spring admission|
The PhD program builds on the foundation of an MS in Biostatistics (or its equivalent). Any course work or other requirements of the MS in Biostatistics (or its equivalent) not included in a doctoral student’s previous master’s training must be completed before (and in addition to) the requirements for each doctoral program. The specific course requirements of the PhD program are designed to prepare the student to take the doctoral qualifying examinations.
Students are encouraged to take full advantage of graduate course offerings in other departments of the Mailman School of Public Health, other departments and schools of the Columbia Medical Center (CUMC), and Columbia’s Morningside Heights Campus including, but not limited to the Department of Statistics, to extend their knowledge of theory and methods, develop an area of expertise, and familiarize themselves with the content and issues specific to the biomedical or public health problem or application of their research.
A grade of B or better is necessary in all required courses for both doctoral programs, but up to two elective courses may be taken on a pass/fail basis, especially to encourage students to take courses outside their field of expertise.
PhD students gain all competencies of the MS Theory and Methods Track (MS/TM), and achieve additional competencies in the areas of data analysis and computing, public health and collaborative research, consulting, data management, a cognate field requirement, teaching, and biostatistical research.
Upon satisfactory completion of the PhD in Biostatistics, graduates will be able to:
Data Analysis and Computing
Identify and implement advanced statistical models for the purposes of estimation, comparison, prediction, and adjustment in non-standard settings.
Public Health and Collaborative Research [Sub-header]
Develop and execute calculations for power and sample size when planning research studies with complex sampling schemes;
Formulate and prepare a written statistical plan for analysis of public health research data that clearly reflects the research hypotheses of the proposal in a manner that resonates with both co-investigators and peer reviewers;
Evaluate research reports and proposals for research funding on the basis of their scientific integrity, validity, and the strength of the quantitative analysis; and
Prepare written summaries of quantitative analyses for journal publication, presentations at scientific meetings, grant applications, and review by regulatory agencies.
Identify the uses to which data management can be put in practical statistical analysis, including the establishment of standards for documentation, archiving, auditing, and confidentiality; guidelines for accessibility; security; structural issues; and data cleaning;
Differentiate between analytical and data management functions through knowledge of the role and functions of databases, different types of data storage, and the advantages and limitations of rigorous database systems in conjunction with statistical tools;
Describe the different types of database management systems, the ways these systems can provide data for analysis and interact with statistical software, and methods for evaluating technologies pertinent to both; and
Assess database tools and the database functions of statistical software, with a view to explaining the impact of data management processes and procedures on their own research.
Cognate Field Requirement
Identify important scientific problems in an area of biomedical or public health research outside of biostatistics/statistics (i.e., cognate field) that require the development of innovative biostatistical methodology for their solution;
Review and evaluate the use of biostatistical methods in the cognate field of study;
Demonstrate proficiency in the language of the cognate field; and
Build productive collaborations across fields and disciplines.
Review and illustrate selected principles of study design, probability theory, estimation, hypothesis testing, and data analytic techniques to public health students enrolled in all levels of graduate public health courses; and
Explain advanced concepts in the theory of statistical inference to beginning and advanced graduate students in biostatistics and mathematical statistics.
Identify and integrate new developments in the statistical literature for challenging research problems in public health;
Generate original computer code for new statistical techniques;
Recognize gaps in current inferential methods that limit further public health research and propose solutions based on rigorous theoretical justification; and
Develop guidelines for practical implementation and evaluation of public health research and programs.
Course selection, done in consultation with academic advisors, is based on background, previous education, and research interests.
Students who already have a master’s degree in biostatistics or statistics enter the program with introductory course work already completed in the theories of probability and statistical inference. Students may then immediately register for the required three-semester sequence of advanced courses in the theory of statistical inference and asymptotic statistics. Students will also have had master’s level courses in regression analysis, experimental design, multivariate methods, and the analysis of categorical data and be able to register for advanced courses on these topics as well.
Students who enter with a bachelor's degree will take introductory coursework in probability, statistical inference, regression analysis and multivariate methods during the first year of the program.
Cognate Field Requirement
In addition to mastering course work in statistics, PhD students must satisfy a “cognate field” requirement to gain knowledge and expertise in a biomedical research field other than statistics/biostatistics. Examples of cognate fields of study include biomedical informatics, computational biology, environmental sciences, epidemiology, genomics, health policy research, human biology, physiology, and imaging. Students must take a minimum of two courses at the graduate level in the selected cognate field, chosen in consultation with their advisor and approved in writing by the Department chair or director of academic programs. The requirement is fulfilled when a grade of B or better is obtained in two cognate field courses, which should be completed by the end of the second year of study.
The consulting experience is designed to enable students to demonstrate their ability to integrate their academic studies with the role of biostatistical consultant/collaborator, which will comprise a major portion of their future professional practice. P9185 Doctoral Consulting Seminar is a course where students gain exposure to real world design, analysis, and report writing by helping CUMC investigators who come through the Biostatistical Consulting Service for design, data management, and statistical assistance. Students are required to enroll in the Doctoral Consulting Seminar prior to taking the Statistical Applications Exam.
PhD students are encouraged to take full advantage of graduate course offerings in other departments of the Mailman School of Public Health, other departments and schools of the Medical Center, and Columbia’s Morningside Heights Campus including, but not limited to, the Department of Statistics, to extend their knowledge of theory and methods, develop an area of expertise, and familiarize themselves with the content and issues specific to the biomedical or public health problem or application of their research.
Students are required to take the following biostatistics courses, or to have taken the equivalent courses elsewhere with grades of B+ or better.
|P6400||Principles of Epidemiology||3|
|P8105||Data Science I||3|
|P8116||Design of Medical Experiments||3|
|P8121||Generalized Linear Models||3|
|P8130||Biostatistical Methods I||3|
|P8131||Biostatistical Methods II||3|
|P8157||Analysis of Longitudinal Data||3|
|P8160||Topics in Advanced Statistical Computing||3|
|P9109||Theory of Statistical Inference I||4.5|
|P9110||Theory of Statistical Inference II||4.5|
|P9120||Topics in Statistical Learning and Data Mining||3|
|P9185||Doctoral Consulting Seminar||3|
|GR6301||Probability Theory (taken at Department of Statistics)||4|
Elective Courses in Biostatistics & Statistics
|P8139||Theoretical Genetic Modeling||3|
|G6101||Statistical Modeling for Data Analysis I||4|
|G6102||Statistical Modeling for Data Analysis II||4|
|G6103||Statistical Modeling for Data Analysis III||3|
|G6105||Analysis and Probability I||3|
|G6106||Analysis and Probability II||3|
Residence and Registration Requirements
All PhD students are required to accumulate six Residence Units (RUs). A Residence Unit is the equivalent of a semester of full-time study. After one year of study, students who entered with a master's degree may apply for advanced standing of two RUs representing work completed in their master’s program. All PhD students are expected to attend full-time, especially during the research and dissertation phases of their program. A whole RU, Extended Registration (ER), or Matriculation and Facilities (M&F) are all considered full-time registration statuses. In instances of extreme financial hardship, students may be permitted to work part time during course work when tuition costs are higher.
The Department of Biostatistics offers a limited number of fellowships for students in both the DrPH and PhD doctoral programs. Admission to both doctoral programs is highly competitive, and departmental fellowships are awarded to the most outstanding applicants as funding allows. Full fellowship support consists of tuition and a stipend. International students are also eligible for departmental doctoral fellowships. As part of their fellowship training and duties, all doctoral fellows are expected to serve as Teaching Assistants for one or two courses each year.
To ensure full consideration for admission and funding, submit your completed online application by December 1st. Applicants interested in financial aid, including fellowship support, should indicate their interest on their online applications.
In addition to offering doctoral fellowships, the Department of Biostatistics participates in several other training programs listed below which provide funding for doctoral students. Training grant funding is limited to American citizens and permanent residents. Please refer to each of the following programs for their specific guidelines. Doctoral applicants who meet the additional eligibility criteria for any of the programs below should indicate their interest in that program on their application to the Department of Biostatistics as well.
Cancer Training Program
The multidisciplinary Cancer Training Program, directed by Dr. Alfred I. Neugut of the Department of Epidemiology and funded by the National Cancer Institute, supports both predoctoral and postdoctoral trainees involved in cancer-related studies and research, including students from the Department of Biostatistics. Fellows receive a stipend, partial tuition support, and travel funds. Predoctoral fellows are typically students in the Department’s PhD or DrPH programs who are interested in cancer biostatistics. Post-doctoral fellows may be MDs or PhDs from other areas of science who wish to acquire more training in cancer biostatistics, or holders of a doctoral degree in statistics or a related area who wish to develop a research specialization in cancer. For more information about the Cancer Training Program visit their website or contact: bls85 [at] columbia.edu (Brenda Scariff), Cancer Training Program Coordinator.
The Initiative for Maximizing Student Development (IMSD)
The purpose of the National Institutes of Health-funded IMSD program is to increase the number of historically underrepresented students who receive doctoral training in public health. For more information about the IMSD program for full-time DrPH or PhD students in Biostatistics contact: aabraido [at] columbia.edu (Ana Abraido-Lanza, PhD), IMSD Program Director.
Justine Herrera, MA
Director of Academic Programs
Department of Biostatistics
jh2477 [at] columbia.edu