Professor: David Levin
• CRN 16590: Monday, Wednesday & Friday, 0800-0900 @ MCK 122
Note: HC 241H Science courses are equivalent to HC 207H and 209H Science courses in the pre-Fall 2020 curriculum requirements and will fulfill the HC207H / HC209H requirement. If you have already completed HC 207H or 209H, you do not need to take HC 241H in the Fall 2020 curriculum requirements.
Over the past twenty years, the explosion of computing power, coupled with the growth of networked communication and information systems, has allowed for the unprecedented collection of many types of information. Moreover, it is now possible to process this data in inexhaustibly many ways. Yet making sense of this data is difficult: what is a true signal, and what is just noise or chance fluctuations? Therefore, careful data analysis and statistical reasoning place among the most needed skills in the twenty-first century.
The goal of the course will be to develop critical thinking about the use of quantitative models in a broad range of disciplines. We will discuss both successful and fallacious use of statistical methods. This is particularly timely now, as daily coverage of the COVID-19 pandemic has brought our focus to many fundamental statistical problems of profound importance. Can we estimate the fatality rate and prevalence of COVID-19, and how reliable are these estimates? Can simulations make accurate projections? How can we know the efficacy of social restrictions? These are some of the questions we will look at from a critical perspective.
The mathematical background for the course will be minimal: college algebra (pre-calculus), or equivalent placement, as the course will emphasize conceptual understanding. The course will be centered around case studies and course readings, which will include diverse selections from textbooks, primary texts, and review articles.
For example, readings may include Excerpts from On the Mode of Communication of Cholera by J. Snow, an early and brilliant example of effective data analysis in epidemiology, Ioannidis JPA (2005) Why Most Published Research Findings Are False. PLoS Med 2(8), an essay on statistical fallacies in the medical research literature, Diaconis, P. and Mosteller, F. (1989). Methods for studying coincidences. Journal of the American Statistical Association 84, 853-861 The necessary statistical background will be covered as we move through the readings and consider different examples.