Master of Science in Public Health Data Science
The Master of Science in Public Health Data Science integrates biostatistics, epidemiology, and computer science. Students will be prepared for careers where there is a growing need for individuals who can learn from data to address important questions in public health and biomedical sciences.
The program is designed to provide students with rigorous quantitative training in statistical and computational skills needed to manage, analyze, and learn from health data. Students will learn to wrangle, scrape, create, and manage large health-related datasets; summarize, visualize, and interpret data; apply statistical methods to draw conclusions from the data; use machine learning to reveal features of large, complex health-related datasets; learn the statistical theory behind common data science methods, and effectively communicate results and findings to a broad audience. This program is designed to be a terminal degree, but for students interested in pursuing further education, it can be used to lay the foundation for a PhD in Biostatistics, Statistics, Data Science or Computer Science.
Priority: December 1st
Final: April 1st
Program at a glance
This 32 unit degree program can be completed in 4 semesters. It consists of 6 core courses (23-24 units) in Population and Public Health Sciences, 1 Computing Requirement (4 units), 1 Population and Public Health Sciences requirement (3-4 units), and a Practicum (3 units).
Core Coursework (26 Units)
Students take 6 core courses in Population and Public Health Sciences and 1 core course in Computer Science.
PM 566 Introduction to Health Data Science (4 units)
An introduction to the toolsets needed to create workable and reproducible datasets, conduct exploratory analysis and visualizations, learn from data, summarize and communicate analytic results.
PM 592 Regression Analysis for Health Data Science (4 units)
Through this course, students will become familiar with data analysis and regression using R.
PM 512 Principles of Epidemiology (4 units)
Terminology/uses of epidemiology and demography; sources/uses of population data; types of epidemiologic studies; risk assessment; common sources of bias in population studies; principles of screening.
PM 520 Advanced Statistical Computing (3 units)
Techniques for the solution of statistical problems through intensive computing; iterative techniques, randomization tests, the bootstrap, Monte Carlo methods.
PM 522b Introduction to the Theory of Statistics (3 units)
Theory of estimation and testing, inference, analysis of variance, theory of regression.
PM 591 Machine Learning for the Health Sciences (4 units)
Introduces Masters and Ph.D. students in the Health Sciences to Machine Learning.
Elective (3 units)
For elective offerings, visit the USC Course Catalogue.
Practicum (3 units)
Students must complete a semester-long practicum, which is an instructor-guided course culminating with a health data science project that combines and applies the knowledge acquired through the program.
PM 606 (3 units)
Trevor Pickering, PhD
Director of the Master of Science in Public Health Data Science program
A Message from Trevor Pickering
Social media clearly knows something about me—I’m frequently bombarded with advertisements for data science programs when I log on. With so many choices out there, I frequently think: what makes our program unique? Our students are set up for success with an emphasis on real applications in public health, research opportunities at the Keck School of Medicine, an amazing Trojan alumni network, and hands-on experience interacting with industry professionals during the practicum.