Prerequisites:

• Basic statistical knowledge, including an understanding of regression analysis
• Basic proficiency in R programming

Who should enroll:

This course is ideal for students and researchers interested in performing regression analysis in R, whether for academic projects, research, or data analysis tasks.

Concept of the course: 

This one-day course offers a comprehensive introduction to regression analysis using R, focusing on both linear and logistic regression techniques. Participants will explore the fundamental principles of regression modeling, learning how to build, interpret, and evaluate models effectively. The course emphasizes a hands-on approach, guiding participants through real-world applications with practical coding exercises in R.

Throughout the session, participants will gain experience in essential tasks such as data preparation, variable selection, and model fitting. They will learn how to interpret regression coefficients, assess statistical significance, and evaluate model performance using appropriate diagnostic measures. Additionally, the course will cover common challenges in regression analysis, including multicollinearity, outliers, and model assumptions, providing strategies for addressing these issues.

By the end of this course, participants will be able to:

• Understand the fundamentals of regression analysis
• Implement linear regression models in R
• Implement logistic regression models in R
• Interpret regression coefficients and model fit statistics
• Diagnose and improve model performance

The workshop is application-oriented and based on the “learning-by-doing” concept. The basics of R or statistical methods and their purpose will not be discussed in the course. At the end of the course, we will dedicate time to your questions. 

Other:

The program should be installed on participants’ own laptops:

• R: https://cran.uni-muenster.de/
• RStudio: https://rstudio.com/products/rstudio/download/
• MS Teams: https://www.microsoft.com/en/microsoft-teams/download-app

Trainer:

Dr. Katherine Ogurtsova (Katherine.Ogurtsova@hhu.de) is working in the Institute of Occupational and Social Medicine, in the working group of Environmental Epidemiology, at University Clinic in the Heinrich Heine University. Her primary qualification includes statistical methods in medicine, epidemiology, and public health with the main focus on R-programming and methodological issues.

Prerequisites:

• Good general PC knowledge
• Programming skills is an advantage
• Basic statistical knowledge is required

Concept of the course:

“R” (https://www.r-project.org/) is a programming language and application environment for statistical analyses and graphics. R is a free GNU project software and can be installed and used free of charge. Unlike many commercial systems (e.g. SPSS, SAS), the R-project is constantly developed by leading scientists and the wide statistical community. All procedures and functions are visible, i.e. the source code can be called up and checked at any time. There are a lot of packages that cover literally all statistical questions and methods.

The seminar gives a first impression of the R functionality and how to deal with a scripting language. The workshop is application-oriented. Standard procedures in R are shown and trained by means of examples and own calculations on a learning dataset. During the first two days the basic knowledge of R is given. The third day could be interesting for advanced users as well.

The participants are required having basic statistical knowledge before being enrolled. The statistical methods and its purpose will be not discussed in the course but only their application in R.

Learning aims:

• Operate with objects, vectors and matrices. Import and export from/to SPSS and Excel sheets. Basic data management and data types. Basic arithmetic operations, data handling and transforming.
• Making simple and complex graphics, running simple and sophisticated descriptive and inferential statistical analysis.
• Understanding how the complex statistical analysis can be performed in R (linear and logistic regressions).

Course content:

• Brief introduction to R, the concept of R and the basic functions and conventions
• Working with RStudio
• Practical exercises with simple calculations and application procedures
• Getting to know simple basic vocabulary of the script language
• Reading complex data records
• Data sorting and data retrieval
• Explorative data analysis: hypothesis testing, confirmatory data analysis
• Creating custom functions
• Creating graphics: simple and complex (gglopts)
• Creating descriptive and inferential statistical procedures (power analysis).
• Basic regression analysis in R, diagnostic techniques for the quality of regressions, statistical testing, diagrams.

Other:

The program can be installed on participants’ own laptops:

• R: https://cran.uni-muenster.de/
• RStudio: https://rstudio.com/products/rstudio/download/

YouTube how-to-do for Windows 10: https://www.youtube.com/watch?v=_2sewGCA0y4&ab_channel=BecomingaDataScientist

YouTube how-to-do for Mac OS: https://www.youtube.com/watch?v=LanBozXJjOk&ab_channel=DataSciencewithTom

Please install R first and then RStudio. The programs must be granted administrator rights in Windows 7 or higher operating system versions. Installation on Mac OS is also possible.

Speakers:

1. Dr. Katherine Ogurtsova

Dr. Ogurtsova is working in the Institute of Occupational and Social Medicine, in the working group of Environmental Epidemiology, at University Clinic in the Heinrich Heine University. Her primary qualification includes statistical methods in medicine, epidemiology, and public health with the main focus on R-programming and methodological issues.

2. Dr. Ralf Schäfer

Dr. Schäfer is a psychologist and a head of Psychological laboratory at Clinical Institute for Psychosomatic Medicine and Psychotherapy at University Clinic in the Heinrich Heine University. Dr. Schäfer’s research interest are psycho-physiological questions related to processing emotions and an investigation into the effectiveness of special psychotherapeutic therapies such as the effects of stress management training.

Prerequisites:

• Basic statistical knowledge
• Basic proficiency in R programming
• Familiarity with data structures
• Some experience in statistical analysis using R is an advantage

Who should enroll:

• Students are interested in the data handling skills
• Students who want to create advanced visualizations and smart report of their data analysis
• R users eager to explore the capabilities of tidyverse and ggplot2 packages

Concept of the course: 

This workshop is designed for individuals who already have a foundational understanding of statistical analysis in R and are eager to elevate their skills to the next level. During the workshop, we will focus on data manipulation using a pipe function and will create complex graphs with ggplot2 package. We will follow the standard flow of a descriptive statistical analysis and data visualization, and will learn how to produce professional Markdown and HTML reports for results presentation. 

Key Topics:

• Tidyverse Essentials:
  • Understanding the philosophy behind the tidyverse and packages “dplyr”, “tidyr” and other.
  • Chaining operations with the %>% operator.
  • Mastering the art of clean, logical and tidy data structures.
• Introduction to ggplot2:
  • Crafting informative visualizations.
  • Customizing plots for publication-ready graphics.
  • Creating complex plots with facets and layers.
  • Utilizing themes and color palettes for consistent styling.
• Markdown and HTML Reports:
  • Creating dynamic and reproducible reports with R Markdown.
  • Converting R Markdown to HTML for sharing and presentation.

By the end of this course, participants will be equipped with the knowledge of how efficiently handle, clean, manipulate, visualize data and create the summary reports. 

The workshop is application-oriented and based on the “learning-by-doing” concept. The basics of R or statistical methods and their purpose will not be discussed in the course. At the end of the course, we will dedicate time to your questions and your visualizations that you want to create or improve. 

Other:

The program should be installed on participants’ own laptops:

• R: https://cran.uni-muenster.de/
• RStudio: https://rstudio.com/products/rstudio/download/
• MS Teams: https://www.microsoft.com/en/microsoft-teams/download-app

Trainers:

Dr. Katherine Ogurtsova (Katherine.Ogurtsova@hhu.de) is working in the Institute of Occupational and Social Medicine, in the working group of Environmental Epidemiology, at University Clinic in the Heinrich Heine University. Her primary qualification includes statistical methods in medicine, epidemiology, and public health with the main focus on R-programming and methodological issues.