Through this course, doctoral students and advanced masters students will build their skills and intuition to use statistical methods to conduct public health and health services research. Rather than providing a menu of options for statistical analysis, the course will emphasize key concepts that unify different approaches, using linear regression as a case study in statistical and causal inference. The course will cover critical sources of bias in estimating point parameters and standard errors, including: confounding, measurement error, missing data, and correlated and heteroskedastic errors. The instructional model for the course will be the integration of lectures, in-class and at-home simulation exercises in R, critique of existing studies, and analysis of real data. Students completing this course will have built the intuition and gained the hands-on-experience needed to implement regression-based analyses in their own future work; to take higher level courses in statistical analysis and study design successfully; and to engage with the quantitative literature in public health and health services research with a critical eye.