CARMA Live Online Short Courses

The Americas Region – January 2023

Live Online Short Courses – January 3-6, 2023

Sponsored by University of South Carolina

Short Course Sessions and Groupings

All courses in are taught concurrently, so a participant can take only one course.

Complete Course Listing

CARMA Workshop: Basics of R

This four-hour Workshop provides information on the package R to prepare attendees for follow-up training in CARMA Short Courses that use R. By attending this online workshop, participants will learn basic skills for using the R Studio interface to: load and activate R packages, import and manage data, and create and execute syntax. Having these basic skills will allow Short Course participants to more easily learn about use of R for data analysis and will enable Short Course instructors to better plan and deliver their content. This Workshop is only available to those who will be attending one of the CARMA Short Courses. It will be available on-line.

During this Basics of R Workshop, attendees will learn:
1. Using R through the R Studio interface
2. Importing data into R
3. R data sets (a.k.a data frames and tibbles)
4. Data types
5. Subsetting columns of data and selecting cases
6. Recoding data and dealing with missing data
7. Merging data (columns and rows)
8. Output objects
9. User defined functions
10. Getting help

Option 1: “Introduction to R and Data Analysis” – Dr. Scott Tonidandel, University of North Carolina-Charlotte

Introduction Video by Dr. Tonidandel (Most Recent Version)

–

This course will provide a gentle introduction to the R computing platform and the R-Studio interface. We will cover the basics of R such as importing and exporting data, understanding R data structures, and R packages. You will also learn strategies for data manipulation within R (compute, recode, selecting cases, etc.) and best practices for data management. We will work through examples of how to conduct basic statistical analyses in R (descriptive, correlation, regression, T-test, ANOVA) and graph those results. Finally, we will explore user-defined functions in R and lay the groundwork for understanding how to perform more complex analyses presented in other CARMA short courses.

Required Software: R (download here), R Studio (download here)

Option 2: “Introduction to Multilevel Analysis with R” – Dr. Paul Bliese, University of South Carolina

The CARMA Introduction to Multilevel Analysis short course provides both (1) the theoretical foundation, and (2) the resources and skills necessary to conduct basic multilevel analyses. Emphasis will be placed on techniques for traditional, hierarchically nested data (e.g., children in classrooms; employees in teams). The first part of the course introduces issues related to multilevel theory (e.g., multilevel constructs; principles of multilevel theory building; cross-level inferences and cross-level biases). The second part of the course discusses issues related to multilevel measurement (e.g., aggregation; aggregation bias; composition and compilation models of emergence; estimating within-group agreement). The last part of the course focuses on the specification of basic 2-level models (e.g., children nested in classrooms; soldiers nested in platoons; employees nested within work teams) analyzed via multilevel regression (i.e., random coefficient regression; hierarchical linear model; mixed effects model). The R software package will be introduced, explained, and emphasized during this short course in preparation for the advanced short course offered in Session II. Participants who prefer HLM, SAS, SPSS, or MPlus (and have expertise with these programs) have the option of completing some assignments with these programs. Participants are encouraged to also bring datasets to the course and apply the principles to their specific areas of research. The course is best suited for faculty and graduate students who are familiar with traditional (i.e., single-level) multiple regression analysis, but have little (if any) expertise related to conducting multilevel analyses.

• Module 1: Multilevel Theory: Constructs, Inferences, and Composition Models
• Module 2: Multilevel Measurement: Aggregation, Aggregation Bias, & Cross-Level Inference
• Module 3: Multilevel Measurement: Estimating Interrater Agreement & Reliability
  • Examples using R
  • Examples using SPSS Software (time permitting)
• Module 4: Multilevel Measurement and Multilevel Modeling: A Simple Illustration of Analyzing Composite Variables in Hierarchical Linear Models
  • Examples using R
  • Examples using SPSS Software (time permitting)
• Module 5: Review of the 2-Level Model and Final Q & A

• Other topics (only if time permits) might include:
  • Extension of the 2-level model to the study of growth and change (i.e., growth model)
  • Different centering/scaling stragies (e.g., group-mean centering vs. grand-mean centering)

Required Software: R (download here), RStudio (download here)

Option 3: “Introduction to SEM with LAVAAN” – Dr. Robert Vandenberg, University of Georgia

Introduction Video by Dr. Vandenberg (Most Recent Version, June 2021)

This introductory course requires no previous knowledge of structural equation modeling (SEM), but participants should possess a strong understanding of regression AND have understanding about the basic data handling functions using R. All illustrations and in-class exercises will make use of the R LAVAAN package, and participants will be expected to have LAVAAN installed on their laptop computers prior to beginning of the course. No course time will be spent going over basic R data handling and installing the LAVAAN package. The course will start with an overview of the principals underlying SEM. Subsequently, we move into measurement model evaluation including confirmatory factor analysis (CFA). Time will be spent on interpreting the parameter estimates and comparing competing measurement models for correlated constructs. We will then move onto path model evaluation where paths representing “causal” relations are placed between the latent variables. Again, time will be spent on interpreting the various parameter estimates and determining whether the path models add anything above their underlying measurement models. If time permits, longitudinal models will be introduced.

Required Software: R installed with LAVAAN package (download here), RStudio (download here)

Option 4: “Statistical Analysis of Text with R” – Dr. Jeff Stanton, Syracuse University

The use of natural language text in research has proliferated in recent years. Many statistical tools are available to aid in the discovery of patterns in natural language text – particularly in large data sets of short text segments, such as one might find in social media, web pages, or news articles. Text mining techniques can support theory development by uncovering patterns that would be challenging to find with traditional analysis techniques. Text mining can also be used alongside standard confirmatory statistics such as regression and classification. In this CARMA short course, we will use R-Studio and Jupyter Notebooks to learn text mining and various related methods for statistical analysis of text.

No prior experience with R is required, but I recommend that you take CARMA’s basic R introductory R course if you have no previous exposure. We will discuss the conceptual steps involved in text mining, and then use R to put those concepts to work on a variety of open data sets I will provide. Students are welcome to bring their own data sets as well, but this is not required. We will examine the document feature matrices, dictionary based sentiment analysis, latent semantic analysis, exploratory topic modeling, structural topic modeling, and word embedding. We will test some predictive techniques, using features of text documents as predictors. Time permitting we will briefly examine a few aspects of natural language processing including deep learning techniques such as BERT.

Students who participate successfully in this short course can expect to learn enough about text mining to begin experimenting with these tools in research. The ideal participant will have an interest in improving their skills with R, knowledge of basic descriptive and inferential statistics, and curiosity about exploring empirically driven strategies for analysis of data sets containing text. No prior knowledge of text mining or natural language processing is needed.

Required Software: R (download here), R Studio (download here)

Option 5: “Advanced Data Analysis with R” – Dr. Justin DeSimone, The University of Alabama

Introduction Video by Dr. DeSimone (Most Recent Version, June 2021)

–

This short course will begin with an introduction to linear regression analysis with R, including models for single/multiple predictors and model comparison techniques.  Particular attention will be paid to using regression to test models involving mediation and moderation, followed by consideration of advanced topics including multivariate regression, use of polynomial regression, logistic regression, and the general linear model. Exploratory factor analysis and MANOVA will also be covered. For all topics, examples will be discussed and assignments completed using either data provided by the instructor or by the short course participants.

Required Software: R (download here), R Studio (download here)

Option 6: “Systematic Reviews and Meta-Analysis in R” – Dr. Ernest O’Boyle, Indiana University

Introduction Video by Dr. O’Boyle (Most Recent Version, June 2021)

Meta-analyses have now become a staple of research in the organizational sciences. Their purpose is to summarize and clarify the extant literature through systematic and transparent means. Meta-analyses help answer long-standing questions, address existing debates, and highlight opportunities for future research. Despite their prominence, knowledge and expertise in meta-analysis is still restricted to a relatively small group of scholars. This short course is intended to expand that group by familiarizing individuals with the key concepts and procedures of meta-analysis with a practical focus. Specifically, the goal is to provide the necessary tools to conduct and publish a meta-analysis/systematic review using best practices. We will cover how to; (a) develop research questions that can be addressed with meta-analysis, (b) conduct a thorough search of the literature, (c) provide accurate and reliable coding, (d) correct for various statistical artifacts, and (e) analyze bivariate relationships (e.g., correlations, mean differences) as well as multivariate ones using meta-regression and meta-SEM. The course is introductory, so no formal training in meta-analysis is needed. Familiarity with some basic statistical concepts such as sampling error, correlation, and variation is sufficient.Required Software: R (download here), RStudio (download here)

Option 7: “Introduction to Python for Research”– Dr. Jason T. Kiley, Clemson University

Introduction Video by Dr. Kiley (Most Recent Version, June 2021)

–

Python is a general purpose programming language that includes a robust ecosystem of data science tools. These tools allow for fast, flexible, reusable, and reproducible data processes that make researchers more efficient and rigorous with existing study designs, while transparently scaling up to big data designs. This short course focuses on the foundational skills of identifying, collecting, and preparing data using Python. We will begin with an overview, emphasizing the specific skills that have a high return on investment for researchers. Then, we will walk through foundational Python skills for working with data. Using those skills, we will cover collecting data at scale using several techniques, including programmatic interfaces for obtaining data from WRDS, application programming interfaces (APIs) for a wide range of academic and popular data (e.g., The New York Times), web scraping for quantitative and text data, and computer-assisted manual data collections. From there, we will assemble and transform data to produce a ready-for-analysis dataset that is authoritatively documented in both code and comments, and which maintains those qualities through the variable additions, alternative measure construction, and robustness checks common to real projects.

By the end of the course, you will have the skills—and many hands–on code examples—to conduct a rigorous and efficient pilot study, and to understand the work needed to scale it up. The course design does not assume any prior training, though reasonable spreadsheet skills and some familiarity with one of the commonly–used commercial statistical systems is helpful. In particular, no prior knowledge of Python is required, and we will cover a general introduction to Python in the beginning of the course content.

Option 8: “Web Scraping: Data Collection and Analysis” – Dr. Richard Landers, University of Minnesota

Introduction Video by Dr. Landers (Most Recent Version, June 2021)

In this course, you will learn how to create novel datasets from information found for free on the internet using only R and your own computer. First, after a brief introduction to data source theory, web architecture, and web design, we will explore the collection of unstructured data by scraping web pages directly through several small hands-on projects. Second, we will explore the collection of structured data by learning how to send queries directly to service providers like Google, Facebook and Twitter via their APIs. Third, we will briefly explore analytic techniques for analyzing scraped data. Finally, we will walk through the various ethical and legal issues to be navigated whenever launching a web scraping project.

Option 9: (POSTPONED) “Doing Grounded Theory Research” – Dr. Elaine Hollensbe, University of Cincinnati

Introduction Video by Dr. Hollensbe (Most Recent Version, July 2021)

–

This course will explore the process of conducting a grounded theory study. Through readings, discussion (exemplar and how-to articles) and hands-on exercises, the session begins with generating research questions and interview protocols; collecting data (e.g., participatory, interview, secondary); the coding process; other data analytic processes beyond coding; generating a grounded model; and navigating the review process. This seminar will examine how to ensure trustworthiness and rigor in grounded theory research, and consider challenges of conducting such research when you’ve been trained primarily in quantitative research.