Applied Quantitative Analysis in R for Social Researchers

Introduction to R for Social Researchers is a practical course that will help you use R for real research. Rather than focusing on programming for its own sake, the course emphasises building robust, reproducible workflows for social science research, with a particular focus on data wrangling and data visualisation in R.

Learning outcomes

• Understanding the logic of R, including the main types of objects used in data analysis.
• Import and export data from common file formats used in social research.
• Apply data wrangling in R, including selecting variables, merging datasets, and recoding variables.
• Develop proficiency in data visualisation in R, using ggplot, facets, and appropriate statistical transformations.
• Create weighted summaries and visualisations for data arising from complex survey designs.
• Carry out core statistical analyses in R, including descriptive statistics, t-tests, chi-square tests, correlations, and regression models.

By the end of the course, you will be able to work comfortably in R and apply these skills directly to your own research projects.

Learning methods

The course combines over six hours of video content with hands-on practical exercises, using clear explanations and video demonstrations in R and RStudio. Learning is grounded in realistic, real-world survey data, including the European Social Survey.

Additionally, each section includes quizzes, practical exercises with worked solutions, and access to a dedicated discussion forum where you can ask questions, clarify concepts, and receive support as you apply the material to your own data.

Who this course is for

This course is for PhD students and early-career researchers who want to develop practical and transferable skills in R. Moreover, it is particularly appropriate for social scientists transitioning from SPSS, Stata, or Excel, as well as for policy, education, and other applied researchers who need modern, transparent, and reproducible data workflows. By focusing on core research tasks—data wrangling, data visualisation, and statistical analysis—the course provides a strong foundation for working with complex social data. It also prepares you to progress to more advanced topics, such as mediation and moderation analysis.

Testimonials

Thoroughly recommend anything taught by Alex.

Michelle Degli Esposti, Postdoctoral Researcher, University of Oxford

I owe Alex’s 2018 course my ‘re-birth’ as an R data manager and analyst.

Teresio Poggio, Research Assistant, University of Trento

Understanding why and when relationships between variables occur is central to social science. This course provides a complete, hands-on introduction to the two frameworks that allow researchers to distinguish between a mediating vs moderating variable.

Mediation analysis examines how one variable influences another through an intermediary variable. For example, by investigating how education affects health through income or wellbeing. Moderation analysis examines when or for whom an effect is stronger or weaker. For instance, it can show whether the relationship between stress and health depends on gender or social support. Together, these methods reveal the structure of causal processes and provide deeper insights than simple regression models.

The course begins with the conceptual foundations of mediation vs moderation. It then progresses to advanced applications, including path analysis, structural equation modeling (SEM), longitudinal mediation, multi-group analysis, and causal mediation frameworks. In the process, you will learn to use R packages such as lavaan, mediation, and interactions to estimate, visualise, and interpret these models.

Each lesson combines video lectures, illustrated slides, and clear explanations with practical exercises based on real-world data — ensuring that all examples reflect realistic social research. Practical sessions guide you step-by-step through model estimation, diagnostics, and interpretation, helping you to distinguish between mediation vs moderation. Every exercise includes downloadable datasets, solutions, and R scripts, allowing you to replicate analyses and learn by doing. Additionally, quizzes and a community forum help you deepen your understanding of mediation and moderation.

Learning outcomes

• Understand the logic and assumptions of mediation and moderation analysis.
• Can separate a mediating vs. moderating variable.
• Estimate and interpret interaction and mediation models using R.
• Learn how to estimate moderation with the Johnson–Neyman method.
• Build path models and SEMs to capture direct, indirect, and total effects.
• Extend analyses to longitudinal and causal mediation frameworks.
• Conduct and interpret sensitivity analyses to assess robustness.
• Visualise and communicate results effectively through plots and model diagrams.

A course designed for you!

This course is designed for researchers, students, and analysts in the social, behavioural, and health sciences who already have a basic familiarity with regression analysis and want to better understand mediating vs. moderating variables. Whether you are working with survey data, experiments, or observational studies, you will gain the skills to design stronger studies, interpret results rigorously, and understand the mechanisms that drive social and behavioural outcomes.

Join the course to build a clear understanding of mediating vs moderating variables, and learn how to apply these techniques to your own research using transparent, reproducible, and data-driven methods. On the way, move your career forward by improving your statistical skills and getting better at using R for social research.

Testimonials

Alexandru Cernat is the best quantitative analysis teacher I have experienced. His courses are well thought out, and this one is the exact same. I feel my knowledge of mediation analysis has grown substantially, and the practicals accompanying the lectures are really useful.