Statistical Theory for Social and Policy Research
Description: Since social and policy analysis is rooted in empirical studies of complex phenomena, researchers make extensive use of statistical tools in designing data capture, obtaining useful information from data, and presenting results convincingly to various audiences. Probability and statistical theory provide the basis for these effective statistical tools. At the doctoral research level, you gain confidence in the application of statistical tools as you better understand probability and statistical theory. Further, probability and statistical theory provides background for more advanced methodology courses such as econometrics, probabilistic operations research, decision analysis and multivariate statistical analysis. Additionally, and importantly at the PhD level, having skills in probability and statistical theory permits the researcher to tailor statistical techniques and probabilistic methods to the problem at hand, rather than forcing the problem into a set of cookbook procedures. Student Audience: Within the Heinz School, 90-905 is appropriate for first-year PhD students plus first and second year Masters students who desire and are prepared for a rigorous theory course in probability and statistics as a base for more advanced research methodology. Outside the Heinz school, 90-905 may be of interest to graduate students in Applied History, Architecture, Engineering and Public Policy, GSIA, Philosophy, Psychology, or Social and Decision Science who need background in probability and statistical theory targeted toward social and policy research. Coverage: This is a fast moving one semester course that focuses on basic concepts of mathematical probability and statistical theory. Topics covered include: joint, marginal, and conditional probability, Bayes rule, probability distributions, principles of statistical inference, sampling distributions, maximum likelihood, non-parametric estimation, Bayesian methods, the bootstrap, and regression.