Description

Statistical practice is shifting away from null hypothesis significance testing toward estimation, uncertainty, and effect interpretation using intervals. However, this shift is still poorly reflected in everyday research practice, creating a gap between common scientific inference and recommended best practices. This course addresses that gap and consists of two parts:

 Part I: Conceptual foundations

The first part has a strong conceptual (non-mathematical) focus. We start from first principles, examining the logic, assumptions, and goals of statistical inference. Hypothesis testing and estimation-based approaches are compared, their strengths and limitations discussed, and widespread misconceptions clarified. Emphasis is placed on effect sizes, uncertainty, and interval estimates as tools for scientific reasoning.

 Part II: Practical data analysis

The second part focuses on hands-on data analysis using the R statistical environment. Participants will learn practical approaches to estimation and interval-based inference, with particular emphasis on resampling methods such as the bootstrap. Real-world examples are used throughout, and participants are encouraged to work with their own data.

Learning Objectives

• Think critically about how statistics are commonly used in research and better understand what statistical results do (and do not) tell us
• Use estimation and confidence intervals to describe and interpret effects, rather than relying on yes/no significance tests
• Recognize and avoid common mistakes in the analysis, interpretation, and presentation of data
• Make better choices in study design and data analysis by understanding key assumptions and sources of uncertainty
• Apply practical tools, including bootstrap methods, to analyze real data and draw well-reasoned conclusions

Prior Knowledge

No prior knowledge of statistics or R is required, though helpful. The main requirement is an interest in learning how to gain meaningful insights from data while properly accounting for uncertainty.

Assessment

Attendance and active participation during the course days (16 hours). To get the credit points, you must hand in an assignment that you work on at home (preparation work of 14 hours). Assignment details will be explained during the course. The finished assignment is due no later than three weeks after the course has ended.

Registration

Via email to phd-duw@clutterunibas.ch