Welcome to the UConn Statistical Consulting Services (SCS). This page contains details about past and upcoming workshops. Please take a look below, and contact us if you would like to attend a workshop or have a wokrshop that you would like us to run for you!
The “Introduction to R” workshop offers a hands-on foundation in using R and RStudio for data analysis. Participants are introduced to the R environment, basic coding principles, and essential programming tools. They also learn how to set up their data and begin working with data manipulation techniques using the dplyr and tidyr packages. Attendees will then build on this by advancing skills in data visualization. Participants apply their coding and data manipulation knowledge to create informative graphics, reinforcing the practical utility of R for statistical analysis and research workflows.
Topics include
- R and RStudio
- Basic Coding
- Setting up your data
- Data Manipulation (dplyr & tidyr)
- Visualization
The “Intro to Statistics in R” workshop introduces participants to a modern approach to statistical analysis that emphasizes estimation over traditional null hypothesis significance testing (NHST). It begins by addressing common issues in research practices, such as bias, selective analysis, and poor replication, and advocates for a shift toward effect sizes, confidence intervals, and meta-analysis to improve research integrity and transparency. Participants learn the fundamentals of p-values, confidence intervals, and effect sizes, including practical tools like Cohen’s d. The workshop presents strategies for implementing the “new statistics” approach, highlighting its benefits while acknowledging challenges like transitioning away from entrenched NHST practices.
Topics include
- NHST vs Estimation & Uncertainty Quantification
- p-values and Confidence intervals
- Comparing means (t-test and ANOVA)
- Multiple Regression
- Marginal Effects
The "Generalized Linear Models" workshop introduces participants to modeling techniques for various types of outcome data using R. It begins with a refresher on linear models in R and then explores how different outcome types (binary, count, and bounded) require specific link functions and modeling strategies. Participants learn how to apply generalized linear models to these data types and interpret results using marginal effects. By the end, attendees gain practical skills in fitting and interpreting linear and generalized linear models, with a focus on transparency, flexibility, and statistical rigor in applied research.
Topics include
- Linear models in R
- Outcome types and link functions
- Binary Data
- Count Data
- Bounded Data
The “Mixed Effects and Multi-level Models in R” workshop introduces participants to the principles and practical implementation of hierarchical modeling. It begins with an overview of multi-level data structures,followed by a conceptual breakdown of within- and between-individual variance. Participants will explore how to specify models that include random intercepts and slopes. Real-world examples are used to reinforce understanding and demonstrate the application of multi-level modeling in research contexts.
Topics include
- Introduction to multilevel data
- Fixed vs Random effects
- Random Intercepts vs Random Slopes
- Model Checking