4/8 Graphics in R | 4/9 Multidimensional scaling
Full announcement by Professor Harold Clarke:
I am very pleased to announce that Professor William Jacoby, Director of the ICPSR Research Methods Summer Program at the University of Michigan will be doing the spring short course on April 8th and April 9th. All EPPS and UTD faculty, graduate students and staff are welcome. The course has two topics: day 1 will focus on doing simple and advanced graphics with R; day 2 will focus on contemporary multidimensional scaling methods for social science data. Details to follow.
Asbel Smith Professor
STATISTICAL GRAPHICS FOR VISUALIZING DATA WITH R
This short course will cover methods for obtaining visual displays of quantitative information. We will discuss ways to, quite literally, look at your data. This is important because graphical representations avoid some of the restrictive assumptions and simplistic models that are often encountered in empirical analyses. The material presented in the workshop should be useful for people at varying levels of technical sophistication. Virtually everybody who is exposed to these ideas and methods seems to agree with a variant of the old cliché: A picture is worth a thousand numbers.
The course will begin with a discussion of some basic ideas, concepts, and examples. We will examine the advantages of graphical displays relative to tabular presentations, the situations in which graphical displays are most useful, the principles underlying effective graphical displays, and a tour of some specific graphical methods for univariate, bivariate, and multivariate data.
Next, we will learn how to produce visual displays of quantitative information, using the lattice package within the R statistical computing environment. Lattice displays are very easy to produce and they can be customized in many ways. We will examine the R functions for 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.
Workshop participants will get hands-on experience in constructing and modifying graphs using the Lattice system. Some prior experience with the R statistical computing environment would be helpful, but it is definitely not required. In fact, the workshop material will assume that participants have no prior experience with, or exposure to, R.
MULTIDIMENSIONAL SCALING: AN INTRODUCTION
Would you like to draw pictures of your data, in ways that reveal structures which are not obvious from inspection of the data values, alone? Multidimensional scaling (MDS) tries to accomplish exactly that objective. To be more precise, MDS produces “map” of stimuli, based upon information about the “proximities” among those stimuli.
Multidimensional scaling methods have many potential applications in empirical research. They can be used to: simplify the contents of large, complex datasets; model similarities among sets of objects; estimate the cognitive structures underlying survey responses; and optimize the measurement characteristics of qualitative observations. MDS can be generalized to show individual differences across distinct data sources (e..g, subsets of survey respondents or data collected at different time points). It also can be adapted to represent respondent preferences among a set of stimuli (so-called “ideal points” models).
This short course will provide a basic introduction to multidimensional scaling. Specific topics to be covered include: The basic idea of MDS; types of data that might be input to MDS; the general estimation procedure; interpretation of results; different varieties of MDS; and software options for performing MDS analyses. The course will consist of two parts. The first (and longer) part will occur in the classroom, to introduce the basic concepts and ideas. The second part will occur in the computer laboratory, to give course participants hands-on experience with the MDS routines in one or more of the major statistical software packages.
This short course is intended for a general audience. It does not assume any prior experience with MDS or familiarity with advanced statistical methods (i.e., beyond basic regression analysis). Participants should be able to perform basic data processing tasks with a statistical software package (e.g., Stata, SAS, SPSS), but no special knowledge of MDS software is assumed or necessary.
Professor Jacoby’s biographical sketch:
William G. Jacoby is a Professor in the Department of Political Science at Michigan State University. He is also a Research Scientist at the University of Michigan, where he serves as Director of the Inter-University Consortium for Political and Social Research (ICPSR) Summer Training Program in Quantitative Methods of Social Research.
Professor Jacoby received his Ph.D. from the University of North Carolina, Chapel Hill in 1983 and his main professional interests are mass political behavior (public opinion, political attitudes, voting
behavior) and quantitative methodology (measurement theory, scaling methods, statistical graphics, modern regression). Professor Jacoby’s current research focuses on citizen ideology and belief system organization, value choices and their implications for subsequent political orientations, measuring policy priorities in the American states, the consequences of measurement assumptions for statistical models, and graphical strategies for data analysis. His most recent book (with Michael Lewis-Beck, Helmut Norpoth and Herbert Weisberg) is The American Voter Revisited (Ann Arbor: University of Michigan Press, 2008).