Workshops

Six scheduled pre-conference workshops, whose main focus is on software skill development, will be held on-site before the formal start of the conference.

• In order to secure your spot in a workshop you must register in advance. Workshop registration information will be available in February 2008. Please check our website then to reserve your spot in the workshop(s) of your choice. Space is limited!
• Workshops will be held in instructional computer labs therefore there is no need to bring a laptop.
• Fees: $75 half-day, $100 full-day. You may register for any combination of workshops.
• Session I workshops begin at 9 a.m. and end at 1 p.m. Session II workshops begin at 2 p.m. and end at 6 p.m. All workshops will take place on Wednesday, June 25, 2008.

Bayesian Analysis with WinBugs

Matthew Johnson, Ph.D.
Assistant Professor of Statistics, Baruch College

Session I: Introduction to Bayesian Methodology, MCMC and WinBugs

The first portion of this workshop will introduce participants to the basic concepts required for Bayesian data analysis. Topics will include the statistical model, prior and posterior distributions, Directed Acyclic-Graphs, and Bayesian inference. These topics will be illustrated using the normal and/or binomial distributions. After this general introduction, the workshop will introduce participants to the basics of Markov chain Monte Carlo (MCMC) and WinBugs. Specifically, topics will include an overview of MCMC, the most common algorithms, using WinBugs (syntax and the Doodle utility), and output analysis (monitoring convergence, point estimates, posterior variance and standard deviation, approximating credible intervals). These topics will be illustrated using the same simple examples from the first portion of the workshop as well as a simple normal measurement error model.

Session II: Psychometric examples using WinBugs.

In this workshop participants will develop Bayesian models for a number of popular models from psychometrics, and then use WinBugs to approximate the posterior distributions of the model parameters. Models for discussion will include models for paired comparisons, IRT models, unfolding models, latent class and mixture models and factor analysis models. After the discussion and implementation of Bayesian psychometric models in WinBugs, some advanced MCMC topics will be discussed. Possible topics for discussion will include approximating Bayes factors with MCMC output and goodness of fit and model checking with posterior predictive checks.

R

Keith Markus, Ph.D.
Associate Professor of Psychology
John Jay College of Criminal Justice

Session I: An Introduction to R

This workshop will provide a general introduction to R that does not presuppose any prior familiarity with R. Topics will include installing R, the R environment, R preferences, R help and documentation, reading and writing data, creating and manipulating R objects, indexing complex R objects, working with data frames, statistical functions, descriptive statistics, simple graphs, classical statistics, and fitting general linear models, as time permits.

Session II: Advanced Statistical Analyses in R

This workshop will have two parts. The first part will explore R graphics in greater detail, building on the material covered in Workshop I. This will include using external syntax files to create complex graphs, and writing ad hoc R functions to create graphs. The second part will survey some of the packages available in R for estimating modern statistical models such as item response theory models, structural equation models, latent class models, and linear random coefficient models, as time permits. This workshop will presuppose familiarity with R basics covered in the first workshop and some prior familiarity with the models discussed in the second half although prior facility with each of these models is not required.

SAS

Wendi Wright, B.S.
Senior Statistical Analyst, CTB/McGraw Hill

Session I: Loop Do Loop Around Arrays

Have you ever looked at your newly created SAS program and thought … there must be a better way. Many SAS programmers avoid arrays because they think that they are difficult. The truth is that they are not only easy to use, but they make your work easier. Arrays are SAS Data step statements that allow iterative processing of variables and text. This workshop will present the basics of using array statements and will cover syntax of arrays and many tips and tricks to make them useful for you. Some examples include making repetitive comparisons, setting zeros to blanks (or vice versa) in a series of variables, input and output of files, use of multidimensional arrays, and finding a certain value in a series of variables. 

Session II: %Include a SAS Macro

This workshop will present general information and specific techniques with examples for the SAS programmer to begin or expand their use of the SAS macro language. Introductory topics covered will include the differences between regular SAS code and Macro code, the difference between a macro variable and a macro, creating and invoking both macro variables and macros, defining and passing macro parameters, interfacing with the data step using call symput and symget, and trouble shooting and common errors. Intermediate topics will include conditionally executing macro code (%IF, %THEN), repeatedly executing macro code (%DO, %TO), the difference between global and local macro variables, macro functions, indirectly referencing macro variables, automatic vs. user-defined macro variables and efficiency tips.

Note: Prerequisites for SAS Workshops

Participants should be familiar with SAS using the PC in the Microsoft Windows environment.  He/she should be able to open SAS, code a simple program, submit it and then clear the log and output windows. Knowledge equivalent to an introductory SAS course is sufficient.