EPSY 887 Computational Statistics: Institute (Spring 2013)

Instructors: J. Bryer (jason@bryer.org) and R. Pruzek (rpruzek@albany.edu)
Meeting Time: Monday 6:00 p.m. - 9:00 p.m.
Grading: 3 credits, S/U grading
Course Website: https://github.com/jbryer/CompStats

This seminar will provide an introduction to statistical programming for data analysis with an emphasis on the analysis of large datasets. With the increased availability of large national and international datasets (e.g. PISA, TIMMS, NAEP, ECLS) there is a great opportunity and potential for researchers to focus on important substantive research questions that are difficult to address by other means. However, the analysis of large datasets requires special analytical procedures not found in commercial statistics software. Utilizing the open source statistical software R, students will be introduced to tools and procedures for analyzing large datasets with an emphasis on conducting transparent and reproducible research.

Topics:

• Introduction to R (e.g. data input, recoding, etc.)
• Reshaping data (reshape package)
• Data visualization vis-à-vis a grammar of graphics (ggplot2)
• Introduction to programming for data analysis (e.g. loops, conditional statements, functions, etc.)
• Missing data (MICE, Amelia II)
• Analysis of complex surveys (e.g. use of replicate weights and multiple plausible values) (survey package)
• Document preparation and typesetting with LaTeX and Sweave
• R package development
• Software project management principles as applied to data analysis (e.g. source countrol, progress tracking, versioning, Github, R-Forge, etc.).
• Other data analytic topics as identified by students and appropriate for analysis of large datasets. Topics may include propensity score analysis, multilevel modeling, IRT, random forests, regression trees, etc.

Students are encouraged to bring their own data and/or research questions as this seminar will emphasize applied statistics and analysis. Class examples however, will utilize the Programme for International Student Assessment (PISA; OECD 2009).

Software

This course will make substantial use of the R software language. The following software is required and freely available. See the installation page for details.

• R
• Rtudio
• LaTeX (MiKTeX on Windows or MacTeX on Mac)

Textbooks

These are the recommended books for learning R. They provide two different perspectives on R. Kabacoff presents R from a data analyst point-of-view whereas Matloff provides more of a software programming perspective. The complement each other nicely.

• Kabacoff, R.J. (2011). R in Action: Data Analysis and Graphics with R. Shelter Island, NY: Manning.

• Matloff, N. (2011). The Art of R Programming. San Francisco, CA: No Stratch Press