This course is designed to teach students the fundamentals of applied statistical data analysis. Students successfully completing this course will be able to: choose appropriate statistical analyses for a variety of data types; perform exploratory data analyses; implement commonly used one and two-sample hypothesis testing and confidence interval methods for continuous variables; perform tests of association for categorical variables; conduct correlation and simple linear regression analyses; produce meaningful reports of statistical analyses and provide sound interpretations of analysis results. Students will be able to implement the statistical methods learned using SAS and JMP software on personal computers.

The lab portion of BIOS 500 is designed with two purposes in mind: 1) to illustrate concepts and methods presented in the lectures using hands-on demonstrations and 2) to introduce SAS, a widely used statistical software package, as a data analysis tool. By the end of the semester, you should be able to produce and interpret statistical output for methods learned in BIOS 500 lecture.

Prerequisites: BIOS 500 or permission of instrcutor. This course is the follow-up to Biostatistical Methods I (BIOS 500). Students will apply many of the concepts learned in BIOS 500 in a broader field of statistical analysis: model construction. Topics covered include Linear Regression, Analysis of Variance, Logistic Regression and Survival Analysis. Students who successfully complete this course will have a deep understanding of many analytical methods used by public health researchers in daily life. BIOS 501 Lab is a component of this course.

Prerequisites: BIOS 500 & BIOS 501 or permission of instructor.We start with data analytic methods not covered in BIOS 500 & BIOS 501 (Statistical Methods I & II). We then focus on multilevel modeling of intra- and inter-individual change. Other hierarchical models will also be examined to analyze other types of clustered data. As in the prerequisite courses, we will learn how to specify an appropriate model so that specific research questions of interest can be addressed in a methodologically sound way. Students will use SAS to perform the statistical analyses.

Pre-Requisites: PRS 500D as prerequisite or special permission required to enroll. This course presents basic concepts and data analytic methods with an emphasis on interpretation of common statistical results. Topics covered include summary statistics; probability concepts; confidence intervals; hypothesis testing for means, proportions, and difference between means and proportions; contingency tables (including relative risk and odds ratio); and simple linear regression and correlation. Students will use Microsoft Excel for elementary statistical analyses. [Applied Epidemiology students and Applied Public Health Informatics students take BIOS 516D instead of BIOS 503D.]

Students outside of Biological and Biomedical Sciences must get permission from the instructor. Intended for PhD candidates in the biological and biomedical sciences. Introduces the most frequently used statistical methods in those fields, including linear regression, ANOVA, logistic regression, and nonparametric methods. Students learn the statistical skills necessary to read scientific articles in their fields, do simple analyses on their own, and be good consumers of expert statistical advice.

Prerequisite: Multivariate Calculus (Calculus III) or permission of instructor. This course is a mathematically sophisticated introduction to the concepts and methods of biostatistical data analysis. The topics include descriptive statistics; probability; detailed development of the binomial, Poisson and normal distributions; sampling distributions; point and confidence interval estimation; hypothesis testing; a variety of one- and two-sample parametric and non-parametric methods for analyzing continuous or discrete data and simple linear regression. The course will also equip students with computer skills for implementing these statistical methods using standard software SAS and R.

This is the first regression analysis course in applied statistics designed for BIOS MPH students. Both theoretical and applied aspects of linear regression and generalized linear regression modeling will be covered in this course. The emphasis will be on applications. The first part of the course covers the following topics: simple linear regression, multiple linear regression, confounding and interaction, residual and influence diagnostics, variable transformations, multicollinearity, model selection and validation. The second part of the course includes: generalized linear models, including logistic regression, nominal and ordinal logistic regression, and Poisson regression. Scientific interpretation of results will be emphasized throughout the course. Students are expected to use R (or SAS if preferred), when necessary, for homework assignments and projects. Prerequisites: Coursework in statistics up to and including an introduction to simple linear regression (BIOS 506 or equivalent). Familiarity with basic concepts of probability, statistical inference, and linear algebra (e.g., matrix calculation) is needed for successful completion of the course.

Prerequisites: Multivariate Calculus (Calculus III) and Linear Algebra. This course provides a mathematically sophisticated introduction to the concepts and methods of biostatistical data analysis. It aims to provide the students the skills to collaborate with investigators and statistical colleagues in the analysis of data from biomedical and public health studies and to communicate the results of statistical analyses to a broad audience. The topics include descriptive statistics; probability; detailed development of the binomial, Poisson and normal distributions and simulation of random variables from these distributions; sampling distributions; point and confidence interval estimation; simulation studies; hypothesis testing; power analysis and sample size calculations; a variety of one- and two-sample parametric and non-parametric methods for analyzing continuous or discrete data and resampling statistics. The course will also equip students with computer skills for implementing these statistical methods using standard statistical software SAS or R.

Prerequisites: BIOS 508 or permission of instructor. The course covers statistical methodology for the analysis of continuous outcome data, primarily from cross-sectional studies and designed experiments. We introduce the key matrix-based methods for estimation and inference based on the multiple linear regression model. Subsequently, topics include secondary hypothesis testing and restrictions, regression diagnostics, model selection, confounding and interaction, analysis of variance and covariance, and an introduction to random effects modeling. Students will also be introduced to logistic regression modeling for binary outcome data.