The Society for Social Work and Research

2014 Annual Conference

January 15-19, 2014 I Grand Hyatt San Antonio I San Antonio, TX

127
Using R to Analyze Local and Global Issues: Implications for Collaboration, Replication, and Dissemination

Saturday, January 18, 2014: 10:00 AM-11:45 AM
Marriott Riverwalk, Alamo Ballroom Salon C, 2nd Floor Elevator Level BR (San Antonio, TX)
Cluster: Research Design and Measurement
Speakers/Presenters:
Joseph A. Mienko, MSW, University of Washington, Richard Smith, PhD, Wayne State University and Gregor Passolt, MS, University of Washington
Background and Purpose: This workshop is an introduction to the free and open source statistical programming language, R. This language is considered essential in disciplines that address local and global issues such as biostatistics, political science, and demography. Because R is free and open source, it has several advantages: 1) the algorithms used to calculate models are available to the user; 2) users may write their own commands and publish them to facilitate methodological dissemination; 3) R has no license or upgrade fees. While other proprietary statistical packages have a scripting language, some do not allow user modification and others are cost prohibitive for many social work practitioners and researchers.

This workshop will demonstrate the use of R and how to document quantitative statistical methods in R for collaboration, replication, and dissemination in peer-reviewed journals. The process of peer review allows for other scientists to inspect and replicate a given project before accepting scientific conclusions. Like any social science, social work researchers often deal with questions that involve complex statistical and data handling procedures that are not easily communicated in a typical methods section. R allows the user to record every step of their statistical analysis in a script file that can then be shared with reviewers, colleagues, and ultimately with the consumers of a given research project that can be easily replicated. There are now several Graphical User Interfaces (GUI), such as RStudio, that facilitate analysis and documentation with an integrated script window, output window and various workspace browsers.

Pedagogical Techniques: This workshop will be presented by two social work researchers who utilize R as their primary tool for statistical analysis. We will explain the limitations of R and situations where other software is a better fit. This workshop will provide individuals with a basic overview of the R environment that includes the following topics: 1) instructions on installing R and R-Studio, 2) the overall structure of the R/S language, 3) reading and writing data in R to generate descriptive statistics, 4) generating a simple linear regression model in R, 5) graphing in R, and 6) how R can improve scientific transparency and replicability in social work research. We will have time to take questions from the audience about running other models in R.

The workshop pedagogy will be a demonstration and no prior knowledge of R or programming is required. Individuals are encouraged to bring their own laptops running any operating system to the workshop. The instructors will provide attendees with an opportunity to install R and workshop data on their laptops which will allow people to follow along with the demonstration.

Implications: Social work researchers who wish to be competitive in a contemporary interdisciplinary research environment need to learn R because it saves time, money and increases the credibility of research on local and global issues.

See more of: Workshops