Modeling in Economic and Social Sciences
When is it the right time to bluff in poker and diplomacy? Why do stock markets crash? How should the delivery of global healthcare be managed? And why do people have the values and beliefs that they do? All of these phenomena depend on models for the way that humans make decisions. This cluster, with faculty from economics, anthropology, sociology, and statistics, will teach students how to build and test models in the social sciences.
There are two types of courses in this cluster. The introductions to game theory and statistical modeling provide students with training in mathematical methods so that they can quickly get to the point where they are able to engage in original research. And the courses on anthropology and sociology provide important domains for application of these methods.
As part of their applied courses, students will be asked to work in small groups and complete an original research project. Students will have access to computer labs and other pertinent research facilities. The goal is to insure that students in this cluster achieve the technical toolkit and analytic perspective that will enable success at Duke and in life.
Cluster Prereq: Math 21 or the appropriate AP score.
Statistics 110FS — An Introduction to Statistical Modeling (QS)
David Banks, Professor of the Practice, Department of Statistical Science; Director of the Statistical and Applied Mathematical Institute
In this course, students will learn about statistical modeling, with primary emphasis on developing critical thinking skills and performing analyses on real data sets. After completing this course, students will be able to design and analyze basic statistical studies, to understand and criticize statistical methods in journals and the media, and to appreciate the power and utility of statistical thinking. Examples and methods are drawn primarily from the behavioral, natural, and social sciences and public policy. As part of this course, students will also learn to use the statistical modeling package JMP.
Political Science 189FS: Machine Learning and Legislative Behavior (R)
Scott de Marchi, Professor of Political Science
This course is an introduction to using machine learning techniques to model behavior in legislatures. We will focus on two canonical activities in legislatures. First, we will build predictive models of voting in the US Congress using text as data to connect the content of bills to votes. Second, we will look outside the United States and model bargaining and coalition formation in proportional representation systems.
Psychology 190FS — Addiction: Brain, Individual and Society
Amir H. Rezvani, Professor in Psychiatry and Behavioral Sciences; Member of the Board of Directors of Triangle Residential Options for Substance Abusers (TROSA)
As the third leading cause of death in the U.S., substance abuse is ranked behind cardiovascular disease and cancer in total mortality. Between 25% and 40% of all general hospital beds are occupied by patients having complications related to alcoholism and substance abuse. When undetected, alcohol and substance abuse lead to major medical and social problems such as pancreatitis, cirrhosis, cardiovascular diseases, cancer and contribute to the occurrence of hypertension, diabetes, GI problems, psychiatric symptoms, violence and fetal anomalies. Alcohol and other drug abuse have far reaching and devastating effects on children, families, and society, including domestic violence, child abuse, sexual abuse, crime, and workplace injuries. The cost to the society is astronomical.
The major goal of the course is to develop scientific, clinical as well as social perspectives on the issue of addiction and substance abuse. Furthermore, students will be able to develop a scientific perspective on the etiology of alcoholism. This course will cover neuropharmacology of addiction particularly alcoholism. Students will be presented with the disease concept of addiction, several models of addiction and appropriate intervention and treatment strategies relevant to each model. Students will also visit Triangle Residential Options for Substance Abusers (TROSA) in Durham, a unique therapeutic community which is based on principles of self-help and will get to interact with recovering drug addicts and interview them. Furthermore, students will be given the opportunity to hear from several recovering drug addicts coming from different socio-economic and genetic backgrounds. Furthermore, a family with a child affected by fetal alcohol syndrome will be presented their daily challenges to students. Students will also get involved in self-designed individual and group projects to better understand the nature of drug addiction and the process of behavioral changes.
Overall, these activities will help students to better understand the problem of drug abuse and addiction particularly alcoholism in our society and to choose more efficient approaches for prevention, diagnosis, intervention and treatment as well as policy making.
- Professor of the Practice of Statistics
- Director of the Statistical and Applied Mathematical Sciences Institute (SAMSI)