Teaching

My goal in teaching is to demonstrate the impact that statistics have on a wide variety of fields, especially in public health and public policy, and prepare students to engage with those statistics. To use evidence to improve society, the public needs an understanding of how to use data and statistical results. And we need statisticians and scientists to fully understand the statistical methods they are using and how to communicate their methods and results to one another and to the public at large. My work with students aims to prepare them to be discerning and critical consumers and producers of statistics and data.

Undergraduate statistics education is a perfect place to improve these skills. Because of this, I encourage my students to learn how to write results for a non-statistician audience and learn how to apply statistics and probability to many different settings. I also encourage them to question statistical and modeling assumptions, and to consider how faulty assumptions and dangerous goals contributed to the misuse of statistics in the past, such as its role in eugenics movements, and contributes to potential misuse today, as in issues of algorithmic bias.

Below you’ll find some of the courses I’ve taught, as the primary instructor or teaching assistant. If you’re interested in learning more about the courses or think that the materials from these courses may be useful to you, especially if you are a contingent faculty member, postdoc, or student teaching a course, please let me know and I’d be happy to share. You can reach me at lkennedyshaffer (at) vassar (dot) edu.

Courses Taught at Vassar College

Spring 2022

• MATH 242: Applied Statistical Modeling. Covering linear and logistic regression and methods for handling time-to-event (aka survival or failure time) data, this second course in statistics teaches students the central role of statistical modeling. Examining the assumptions of common methods helps students understand the power and the dangers of statistics. Real-world applications are emphasized, as well as hands-on work with data and the principles of reproducible research.
• MATH 301: Clustered and Correlated Data. In this senior intensive, I worked with a group of mathematics and statistics majors to explore methods beyond linear and logistic regression for handling clustered and correlated data. Students learned new methods such as mixed effects models and generalized estimating equations and applied them in projects on topics like energy mix, COVID-19 vaccination rates, and educational outcomes.
• MATH 399: Quasi-Experiment Statistics. In this independent study, I supervised a small group of students to go beyond their statistics and econometrics coursework and investigate some modern methods of causal inference. Focusing on difference-in-differences and synthetic control, we read the latest research on these methods and implemented them to answer questions about voter ID laws, financial policy, and abortion restrictions.

Fall 2021

• MATH 241: Probability. In this intermediate-level course, students learn the mathematical basis of the study of probability. This includes the central role calculus plays in defining data-generating distributions and identifying asymptotic results. While the frequentist perspective is emphasized, key Bayesian results are shown and students are encouraged to develop their own understanding that links mathematical axioms of probability to the real world.
• MATH 348: Statistical Principles for Research Study Design. I designed this course based to allow students to explore the role of the statistician in the scientific process. Covering survey studies, randomized experiments, and observational studies, we touch on a variety of study designs, considering both their scientific and statistical advantages and disadvantages. Beyond discussing how statistics can improve study design, we also discuss the key role for ethical considerations and communication and interpretability of study results. Statistics cannot be done in a vacuum.

Spring 2021

• MATH 242: Applied Statistical Modeling
• MATH 348: Statistical Principles for Research Study Design

Fall 2020

• MATH 241: Probability

Teaching Assistant Roles at Harvard College, the Harvard T. H. Chan School of Public Health (HSPH), and the Harvard Graduate School of Arts and Sciences (GSAS)

• Fall 2018 and Fall 2019. BIOSTAT 232 (HSPH/GSAS): Methods I. In addition to working as a TA for this first-year PhD course in the biostatistics department, I helped redesign it to fit the new department curriculum. The theory and practice of a wide variety of statistical methods are covered with their applications to public health and medicine. Primary instructor: Professor Brent Coull.
• Fall 2019. GENED 1129 (College): Infectious Diseases and Social Injustice. In this eerily prescient general education course, we took a historical view at the role of infectious diseases in society and, in particular, how they reflect and exacerbate social injustices. Infectious diseases exploit and perpetuate discrimination along race, gender, sexual orientation, class, and nationalist lines. Fiction and non-fiction that addresses, either explicitly or implicitly, past epidemics can shine a light on how science and society interact. Primary insructors: Professors Donald Goldmann and Kenneth McIntosh.
• Spring 2019. BST 223: Applied Survival Analysis. Primary Instructor: Andrea Bellavia.
• Spring 2018. BST 216: Introduction to Quantitative Methods for Monitoring and Evaluation. Primary Instructors: Professors Bethany Hedt-Gauthier and Marcello Pagano.