Quarto for reproducible analyses and academic writing (English)
Elen Le Foll (Department of Romance Studies, University of Cologne)
Have you ever found yourself copying and pasting the results of your analyses from one programme to another while writing a paper or thesis? Not only is it frustrating to have to do this every time you update your analyses, it is also very error-prone!
This hands-on workshop introduces a reliable, reproducible way to combine text, statistical analyses, tables, figures and cross-linked references in a single publishing format: Quarto. As a multi-language format, Quarto can be used with R, Python, Stata, and many other programming languages, and is also a great tool for academic writing even if you don’t code. It enables you to seamlessly export your research to a variety of formats, including PDF, Microsoft Word and Powerpoint, LibreOffice, and HTML.
To take part actively, please bring your laptop along and ensure that you have updated versions of R, RStudio, and Quarto installed (see detailed instructions). No prior experience is required. If you’d rather just come to listen and try things out at home, that’s fine, too.
The pitfalls of p-values: Reflecting current practices in linguistics (Englisch)
Elen Le Foll (Department of Romance Studies, Universität zu Köln)
In this workshop, we will explore the limitations and common pitfalls of p-values in linguistics research. Through interactive discussions and hands-on exercises, we will gain a deeper understanding of how p-values can lead to both false positives and false negatives, and how Questionable Research Practices (QRPs) such as multiple testing and optional stopping can exacerbate these issues. We will discuss the future of statistical practices in linguistics, including alternative approaches to statistical inference.
This workshop is designed to help you critically evaluate current research practices and consider how to improve them. Whether you’re already running mixed-effects regression models in R or just starting out on your statistics learning journey, this workshop aims to create a collaborative and supportive environment in which we all strive to improve the rigour, reliability, and robustness of future linguistics research.
No prior knowledge or software installation is required, and there will be plenty of time for questions and reflection. I look forward to exploring these important topics with you!
Attention: Unfortunately, this workshop had to be cancelled!
Hands‑on stats clinic (English and German)
Annika Schiefner (University of Amsterdam)
PhD students in linguistics often face messy or heterogeneous data, uncertain coding choices, and a multitude of plausible analytic routes. In this workshop, we will get you the support you need to tackle these issues, so please bring your laptop and your data questions and let’s get going.
You will get a short orientation to common statistical analyses and then work in small groups to develop the optimal approach for your own project. I will support you to find your own analytical approach, meeting you where you are at. For some, that might mean preparing a data cleaning and analysis plan, for others, it means sketching the R code for your analyses. While it won’t be feasible to run every full analysis in the session, we will plan achievable next steps to keep your project moving.