WORKSHOPS / SHORT COURSES
Applied Multilevel Modeling with STATA Multilevel models, also called hierarchical models, represent a powerful class of linear and non-linear statistical models with wide applications in social and behavioral sciences research. They are designed primarily for nested data structures (for example, students nested within classrooms or schools, patients nested within doctors who are nested again within hospitals) but also offer flexible modeling of repeated measures / longitudinal data. This two-part workshop focuses on the conceptual foundation, application and interpretation of multilevel statistical models.The goal of this workshop is to provide participants with balanced material on the underlying concepts as well as techniques for practical implementation of multilevel models on empirical data. |
![]() |
Advanced Data Visualization with R and GGPLOT R provides comprehensive data visualization capabilities with its package ‘ggplot’. ggplot is the complete implementation of the ‘grammar of graphics’, a book by Leland Wilkinson on statistical graphics that detailed a system of constructing graphics using superimposed layers of visual elements. ggplot allows a researcher to move away from cookie-cutter graphics and create custom visualizations based on any specialized analytical need. This workshop provides essential material to get started on ggplot and create custom visualizations. |
![]() |
Statistical Power Analysis for Social and Health Scientists This workshop provides an introduction to the conceptual elements of power analysis with a view of designing statistical studies and determining sample sizes. The free software G*Power will be introduced with examples, guidelines and considerations for practicing power analysis in social sciences research. Reviewing topics in probability, sampling, and hypothesis testing would be useful before participating in this workshop. Participants must bring their own laptop to download G*Power for the hands-on exercises.
|
![]() |
Missing Data Analysis and Multiple Imputation Much of the data that social scientists analyze features some type of missing data. Students may choose to skip a survey question, government records may be lost, or patients may drop out of treatment. Missing data analysis, in particular multiple imputation techniques, allow researchers to account for this partial loss of information and reduce biases. This workshop is designed to help participants conduct multiple imputation analyses using STATA. By the end of this workshop, you should be able to:
|
![]() |
Structural Equation Modeling with STATA Structural equation modeling (SEM) is a comprehensive and versatile analytical framework for estimating relationships between variables and testing complex theories in social sciences. SEM can be effectively used to identify and estimate direct and indirect effects between variables while handling both observed and latent variables in a single framework for both cross-sectional and longitudinal data. Learning SEM opens possibilities to fit a large gamut of statistical models. This workshop aims to be a nifty introduction to SEM using STATA while providing necessary tools and techniques to get you started modeling complex relationships. |
![]() |
Ever dreamt of doing statistical analysis in R? Or had an inkling to change your current data analysis software and enter the unlimited world of R? Then this bootcamp is for you. Designed exclusively for researchers and faculty of UNC Charlotte, this workshop gets you started on analyzing and modeling data with R. We have organized the content in a way that answers the typical questions and addresses challenges in data management and modeling workflow of social scientists. No previous R experience is necessary but other statistical software and modeling experience is needed to attend this workshop.
|
![]() |
Quantile regression has gained significant popularity in economics and other social sciences in recent years. A typical multiple linear regression framework focuses on modeling the mean of the outcome variable Y conditioned on explanatory factors X. But these regression estimates don’t tell you the story of relationship between X and Y over the entire distribution of Y. Quantile regression is a powerful and flexible method to model the distribution of Y therefore giving a richer, detailed picture of the relationships. This workshop will give you a concise introduction and mechanics of quantile regression, and a hands-on tutorial to implement quantile regression models in STATA. |
![]() |
Latent Growth Curve Modeling with STATA Latent growth curve models flexibly extends structural equation models to analyze changes in relationships between variables over time. This class of models have gained popularity in several social science disciplines in the recent years. This workshop focuses on analyzing longitudinal data with growth curve models using STATA. Topics include exploring and visualizing longitudinal data, concepts in latent growth curve model, hands-on examples of longitudinal modeling scenarios and estimation techniques |
![]() |
Non-parametric Methods in Social Sciences with R A large majority of empirical research in social and health sciences follow parametric approaches to statistical analysis. Notwithstanding their popularity, parametric methods also limits the researcher in significant ways. The need to assume distributional structures can be limiting and may even lead to incorrect estimation. This workshop provides useful tools and analytical methods to conduct non-parametric analyses. |
![]() |
Spatial Regression Modeling with STATA Geographic data and the ability to manipulate spatially referenced data is fast-becoming an invaluable skillset for researchers and analysts of various academic disciplines. As a system designed to capture, store, manipulate, analyze, manage, and present all types of spatial or geographical data, geographic information systems (GIS) plays the key role of applying spatial knowledge & techniques to all kinds of applications. The workshop aims to develop capacity in handling spatial data and using appropriate techniques for analyses and modeling. |
![]() |