Data analysis and applied regression

Course: MGMT 680A; Data analysis and applied regression

Instructor: Seth Spain

Time: Weds, 12 - 3pm, AA-370

Office hours: W, 3 - 4pm, and by appt., AA-312

Course website: sites.google.com/site/sethmspain/data-analysis

Course Description:

An important aspect of being a competent social scientist is understanding statistics and data analysis. We must both consume others' research (and therefore must understand the techniques that they have used) and conduct our own (be able to explore, model, and interpret our data).

The goal of this course is to develop a good understanding of applied data analysis for empirical research. Students should be able to identify appropriate statistical tools for a given research problem and be able to implement those tools in appropriate software.

We will begin by reviewing the general linear model and progress through various complications thereof -- including generalized linear

models and multilevel/hierarchical models. Topics we may cover if time (and interest) permits:

Bayesian modeling and Bayesian Computing
Causal inference (including but not limited to the Rubin Causal Model/Judea Pearl's graphical approach/Propensity Scores)
Structural equations and graphical models

Why I encourage use of a text editor:

Word Processors: Stupid and Inefficient

Choosing your workflow applications

Cosma Shalizi's minimal advice on programming (less applicable, but informative):

http://cscs.umich.edu/~crshalizi/weblog/593.html

News:

Added cars-fixed.dat (data for lecture on PCA & Factor analysis) -- 2 November 2011
Added slides from 26 October 2011
The document "Spain_Seth.pdf" is my dissertation. Please refer to it for some details on multivariate multilevel modeling.
31 August 2011 -- Uploaded Bliese intro to multilevel modeling in R. The first chunk is a decent introduction to R.