Multilevel studies and multilevel analysis have been increasingly used in the public health field. This course will discuss the rationale for multilevel studies and multilevel analysis in public health as well as differences with other study designs and other analytical approaches. Although the course will not be heavily mathematical, the basics of fitting multilevel models for different types of outcomes as well as the interpretation of estimates obtained from multilevel models will be reviewed and practiced. Emphasis will be on conceptual understanding, application and interpretation of multilevel analysis in the context of urban health research. The course will also review and critique empirical applications in urban health research and discuss conceptual and methodological challenges in using multilevel analysis.

After completing this course, participants will be able to:

• To understand the fundamentals of multilevel studies and multilevel analysis and their differences with other study designs and analytical approaches.
• To be able to fit multilevel models and interpret estimates derived from them. 
• To be familiar with applications of multilevel analysis in urban health research.
• To understand the strengths and limitations of multilevel analysis for urban health research

Prerequisite knowledge: Knowledge of regression analysis (linear, logistic, Poisson) is required for this course.

As data is collected from more and more sources and across life areas, the applications for using big data for public health purposes are growing. This class will focus on the use of large data sets, specifically claims data and other national survey sources, and provide an overview of structuring data, selecting optimal analytic approaches, and display results in compelling data visualization formats.

The class will be taught in SAS, though other statistical software is possible. Tableau will also be used (free trials are available).

After completing this course, participants will be able to:

• Identify large data set sources
• Describe methods for cleaning and structuring data for use
• Outline methods for identifying data analysis opportunities
• Review data visualization strategies to maximize reach and impact

Prerequisite knowledge: Prior experience in SAS, SPSS, STATA, R, or other statistical coding software is required for this course.

The course is an introduction to the basic concepts and techniques of Geographic Information Systems (GIS). This course contains lectures and hand-on exercises in ArcGIS. The lectures introduce the basic concepts of GIS, data sources, data structure, data management, map projections and coordinate system, making maps, and spatial analysis. Students will also learn how to use the fundamental knowledge of GIS to solve real world problems. Through the hand-on exercises, students will get familiar with the interfaces and analysis tools in the ESRI software package ArcGIS. Students will also practice applying GIS as a tool and a methodological approach to spatially analyze environment and public health data.

After completing this course, participants will be able to:

• Understand basic concepts and terms of GIS, and know how to spatially analyze data using ArcGIS
• Acquire spatial data that is relevant to urban health, e.g., census data, land use data, greenspace, road networks, satellite images, etc.
• Build knowledge of data structure, data management, and coordination and projection systems
• Practice featuring symbols and making elegant maps in ArcGIS
• Apply the knowledge of GIS to solve research questions in urban health research

No prerequisite knowledge or skills are required for this course.

This course is designed to provide students with a general understanding of programming and analytic concepts in R to address challenges that a public health data analyst with a master’s degree might encounter.  The focus of the course is not comprehensive knowledge of R, but rather a basic foundation of knowledge that can be expanded on after the course concludes. The interactive classes will feature a lecture component and a laboratory component. The RStudio environment will be the interface used for all classroom discussion.

After completing this course, participants will be able to:

• Be comfortable operating in the R environment
• Perform basic data management in R
• Understand generic programming concepts in R
• Analyze data in R

Prerequisite knowledge: Prior experience with a statistical programming language (e.g., SAS, Stata) is recommended but not required for this course.

Improving the quality of data visualizations can help in communicating results more clearly and transparently to other scientists, stakeholders and the general public. This course will use a workshop format to help students build their own data visualization projects using R, with Urban Health examples. This course will include a general introduction to basic data visualization concepts and best practices to generate static figures. This course will also cover basic concepts regarding interactive data visualization and spatial data. This course will provide all the required code, and some of the code may be useful to apply to your own visualizations with some adjustments. Thus, participants with a stronger R background will be able to create more complex visualizations. 

Last, each student will bring their own plotting project concept to receive feedback on it and work on the code needed to get a production quality figure.

After completing this course, participants will be able to:

• Understand the basic principles of data visualization
• Perform basic data management to get data ready for visualization using R
• Use the ggplot2 R package to plot high-quality figures
• Create animated plots, interactive web-plots using R, and map spatial data using R
• Convert research/advocacy ideas into effective plots, from data to code to plot

Prerequisite knowledge: Prior basic knowledge of R or concurrent enrollment in the “Intro to R” Summer Institute course taught by Brian Lee, PhD is required for this course. Advanced knowledge of R is not required for this course.

Community-based participatory research (CBPR) is an orientation to research that begins with community research interests and seeks to use the knowledge gained to inform action for community health improvements and social change. Although this approach to research is widely adopted in urban health and generally familiar to both community-based organizations and researchers, many still struggle with the lack of knowledge or experience with applying practical CBPR approaches to a pressing community health issue or potential research partnership.

This course will review the history and principles of CBPR, and introduce participants to practical approaches and tools for equitable and authentic community-researcher engagement in all phases of research: from the creation of research questions, to study design, data collection, data analysis, and dissemination. Students will also learn strategies for participatory grant-writing, budgeting, and opportunities for workforce development and capacity building for community residents and leaders in research. This course is open to researchers and community-based organizations.

After completing this course, participants will be able to:

• Describe the history and principles of CBPR, and what distinguishes it from community-placed research
• Understand the importance of and practical approaches for community-researcher engagement in all phases of a research process
• Apply CBPR principles to research partnership development and grant-seeking processes

No prerequisite knowledge or skills are required for this course.

This course will provide the landscape of policy analysis in real-world settings using a continuum of data tool approaches. Since policy development and implementation occurs in diverse and fast-paced environments, approaches to studying research and policy questions that keep pace with policymakers will be described. Data approaches from individual case studies to large, publicly available data sets will be reviewed and used. Policy products (e.g., policy briefs, infographics) that resonate with policymakers and constituents will be the key deliverables generated and provide scaffolding for future policy analysis and dissemination.

After completing this course, participants will be able to:

• Identify, describe, and execute policy analysis methods
• Utilize data to form and answer policy questions
• Create engaging policy products that contribute to policy solutions

Prerequisite knowledge: Prior working knowledge of the three branches of the US government is recommended for this course.

Bayesian methods combine information from various sources and are increasingly used in biomedical and public health settings to accommodate complex data and produce readily interpretable output. This course will introduce students to Bayesian methods, emphasizing the basic methodological framework, real-world applications, and practical computing. Special consideration will be given to methods for spatial data analysis.

After completing this course, participants will be able to:

• Understand the fundamentals of Bayesian inference and the differences between Bayesian and frequentist (classical) methods
• Formulate research questions and develop Bayesian approaches to address these questions
• Be familiar with the available software for implementing Bayesian methods
• Understand advanced Bayesian methods used in the scientific literature

Prerequisite knowledge: Basic understanding of linear regression and "generalized linear models" (e.g., logistic and Poisson regression) is required for this course. Statistical programming experience is recommended but not required.

This course will provide students with the foundational knowledge and skills necessary to complete their own analyses of data from the National Health and Nutrition Examination Survey (NHANES). NHANES is a continuous annual survey designed to assess the health and nutritional status of adults and children in the United States. The course is comprised of didactic presentations and a hands-on lab. The course will review the survey design and sampling; survey components (i.e., questionnaires, dietary data, laboratory data, and examination data); measurement; and navigation of data documentation.  During the laboratory component, students will conduct an analysis of NHANES data that covers the following: 1) data acquisition and management (e.g., downloading the data; merging data from multiple data files), 2) combining data from multiple years, 3) use of weights (i.e., which weights to apply, creating weights for multiple survey cycles), 4) use of data from multiple survey components (e.g., combining questionnaire data with lab data), and 5) descriptive and multivariable analyses. Note that the purpose of the class is to provide hands-on, practical experience in conducting analyses of NHANES data. While the course will include a series of scripted labs, there will also be an opportunity for you to conduct an analysis of your own. The course will be taught in Stata, but students with proficiency in other statistical packages can work in other packages.

After completing this course, participants will be able to:

• Develop a working knowledge of NHANES design, sampling, data documentation, and data file structure
• Understand how to select and apply appropriate survey design variables (i.e., weights, strata, PSU)
• Download, merge, recode, and conduct basic analyses of NHANES data
• Identify resources available to support further analyses

Prerequisite knowledge: Basic working knowledge of Stata or intermediate experience with another statistical software is required for this course.

This course is designed for researchers who are interested in learning what it means to approach health from a complex systems perspective – exploring why this is needed, what it looks like, and how to leverage complex systems science methods in their own work. Participants will be introduced to many complex systems science methods without focusing in detail on a specific method. The course will help researchers appreciate which tools to use, when, why and how, including whom to partner with and where to learn more.

After completing this course, participants will be able to:

• Understand why many public health and urban health challenges are complex
• Have familiarity with a variety of complex systems science methods
• Plan how to integrate these methods in their own research agendas
• Structure and gain feedback on written and oral communication motivating and explaining the use of complex systems science methods in their work

No prerequisite knowledge or skills are required for this course.