Statistics for Social Sciences

There are two groups of problems and a paper to read and comment on. You need to chose one problem from each group, and you need to choose one of the papers below, read it and write a short opinion of at most two pages on it. If a problem requires calculations or simulations you can use the statistical package of your choice. Include only the essential graphs and/or tables in your solutions. Do not spend time editing the solutions. Use the time to read good statistical books instead. There is no deadline but you get the grade once the seminar assignment is completed.

Group 1:

• Cluster sampling estimation 1
• Cluster sampling estimation 2
• Estimating net asset value
• Randomized response in sampling
• Parameter estimation for Weibull distribution
• Parameter estimation for variance-gamma distribution
• Parameter estimation for the generalized gamma distribution
• Parameter estimation for autoregressive time series

Group 2:

• Analysis of variance
• Testing for subspace for multivariate normal distribution
• Correlation test for multivariate normal distribution
• Regression with dependent errors
• Logistic regression for kyphosis (data)
• Probit model (data)
• Logit model (data)
• Generalized regression

Papers:

• C. Chatfield, Avoiding Statistical Pitfalls, Statistical Science, 6(3), 1991, 240-268.
• R. L. Smith, A Statistical Assessment of Buchanan's Vote in Palm Beach County, Statistical Science, 17(4(, 2001, 441-457.
• D. A. Freedman, From Mouse-to-Man: The Quantitative Assessment of Cancer Risks, Statistical Science, 3(1), 1988, 3-56.
• S. T Ziliak, D. N. McCloskey, The Cult of Statistical Significance, Joint Statistical Meetings 2009, Section of Education.
• D. A. Freedman, On types of scienfific inquiry: The role of qualitative reasoning, The Oxford Handbook on Political Methodology, 2008, Edited by Janet M. Box-Steffensmeier, Henry E. Brady, and David Collier, Oxford University Press.