Creating Powerful and Effective Graphic Displays, William Jacoby
This short course will cover methods for producing visual displays of quantitative information, using the Lattice package within the R statistical computing environment. The session will begin with a brief discussion of the principles underlying effective graphical displays and a tour of some specific graphical methods for univariate, bivariate, and multivariate data. But, most of the course will be devoted to the R functions for reading data and constructing various kinds of graphs (including histograms, smoothed histograms, dot plots, scatterplots, and trellis displays). We also will cover many “tips and tricks” for customizing graphs to fit specific data analysis situations.
Applied Political Network Analysis, Paul Thurner
This course introduces to the practice of political network analysis. Contrary to most introductions to social network analysis, all empirical examples will be drawn from political science contexts: bureaucratic networks, networks of political communication, co-membership in committees and organizations, contract networks, trade flows, etc. Special attention will be given to the analysis of within-governmental and transgovernmental networks of ministerial bureaucracies.
Using Qualitative Information to Improve Causal Inference, Adam Glynn & Nahomi Ichino
This course introduces a new approach to mixed methods based on the Rosenbaum (2002, 2009) approach to observational studies. We demonstrate how to use this approach to address causal questions with small to medium sample sizes while retaining the ability to produce p-values and confidence intervals. Furthermore, we show how to incorporate qualitative information into the analysis to ameliorate the effects of difficult-to-measure outcomes, to construct qualitative confidence intervals, and to consider the robustness of the results in a sensitivity analysis.
Experiments, Causality, and the Study of Politics
This working group will consider new ways that experiments can be used to address causal questions of interest to traditional political scientists. We will attend a selection of panels and poster sessions of two types: ones where experiments will be discussed and ones on topics suggested by the group members that have not been addressed experimentally but potentially might. We will meet to discuss both design and validity issues relevant to both existing experimental work and possible new ways of thinking of old topics using experiments
- 2007: Rick Wilson (coordinator)