Blog

Research Methods at the Weizenbaum Institute: Survey Results

Methods Lab

In December 2022, the Methods Lab conducted an internal survey to map out the methodological experiences and needs at the Weizenbaum Institute. Thanks to everybody who participated! We have identified specific demands and requests at the institute. Even though there already is extensive expertise for a large variety of methods and tools, many Weizenbaum scholars also expressed a wish for additional support and knowledge-building in, for instance, the following areas:

  • Data collection: Automated observation (e.g., logging, tracking), Automated content analysis, Web Scraping, API-based data collection, and Eye-Tracking
  • Data Analysis: Network Analysis, Deep/Transfer Learning, Natural Language Processing, and Classification Methods
  • Software/Tools: R, Python, and Network analysis software

With these results as our polaris, we in the Methods Lab have embarked on the expedition of developing a future methods training and consulting program suited to your needs, which we will announce shortly. In the meantime, the results of the survey hopefully serve as a launch pad for networking amongst the scholars at the Weizenbaum Institute.

Software Review: BRAT Rapid Annotation Tool

Christian Strippel

Our Methods Lab group lead and WI research associate, Christian Strippel, has written a software review of the BRAT rapid annotation tool, co-authored by Laura Laugwitz, Sünje Paasch-Colberg, Katharina Esau, and Annett Heft. The review is published in issue 4/2022 of Medien & Kommunikationswissenschaft. Read the article here.

In the context of interdisciplinary collaboration, especially with colleagues from computer science, communication and media research has for some time been confronted with a wide range of research software with which it has had little prior experience. In addition to programming lan­guages such as Python or R, these include specific tools for text analysis that represent an alterna­tive to previous variants of computer-assisted content analysis. With the brat rapid annotation tool (BRAT) we present such an alternative in this paper and review it against the background of our experience in using it. BRAT is a web-based open-source text annotation tool that was developed by an international team of computer scientists about ten years ago. The article introduces the tool and its most important features, presents examples for its use in qualitative and quantitative content analyses on the basis of three case studies, and finally evaluates it with regard to potentials and difficulties for the field.

A few notes on the Methods Lab

Methods Lab

Dear all, 

Welcome to the digital baptism of the Methods Lab blog. This blog will keep you informed about our work, future workshops, events, and other resources and materials that may be useful to you in your upcoming research.

As a unit, we are committed to three principal tasks: training, consulting, and research. We aim to assist you with all your methodological questions, issues, and needs, no matter how large or small, and to coordinate expertise at the institute. Think of us as a hub, a metaphorical Rome, if you will, where all your methods-related queries, and (non-)knowledge have a space to converge. If you have any thoughts, suggestions, or concerns, don’t hesitate to contact us – we will always lend you an ear. 

At the start of December, we asked you to participate in a survey in order to give us an overview of your expertise and needs regarding data collection, analysis, and software. With the help of the results, we have created a preliminary training program tailored to your wants and needs. To everyone who participated: thank you!

On that note, we are delighted to announce that our first official workshop will take place at the beginning of March. Besides that, we have two more workshops planned for spring.

So stay tuned for further announcements about many exciting things to come! We look forward to beginning this new chapter with you.