Alexandra Krebs: Identifying User-Types and Types of Historical Narrations in the “App into History” using K-Means Clustering

The more the digital transformation of society progresses, the more digital aspects of learning and teaching come into focus. Unfortunately, in the case of historical learning in schools, the status quo of analog teaching forms has mostly persisted. The few attempts at digital historical learning often restrict themselves to simple quizzes or knowledge-based tasks. Furthermore, studies about digital historical learning processes are lacking. Only a handful of works have been carried out in this field until now. 

To tackle this issue, the project “App into History” addresses digital history learning as a process of historical thinking based on the concept of historical narrativity (Arthur Danto, Hans Michael Baumgartner, and Jörn Rüsen). The modular educational platform enables students to develop historical questions and investigate them using historical sources from digital archives, with several digital tools, and other narrations on the web to tell their own stories while taking a position in controversial debates inside society. In the “Story Modus Bethel,” for example, students investigate for whom a new street in Bethel (Bielefeld, Germany) should be named and write an expert report for the city council. To accomplish this, they carry out a historical research project in the app using digitized historical sources from the Bodelschwingh Foundation that deal with the history of eugenics, “euthanasia,” and forced sterilization. 

Additionally, two mixed-method studies focusing on user behavior (a pilot study with university students and the main study with high school students) were executed to investigate the historical learning processes and narrations inside the application. The central research question is how users narrate history in the application and which other assistance tools should be developed for different user types. For this purpose, qualitative and quantitative data were collected and analyzed using qualitative content analysis, logfile analysis, and computer-based cluster analysis (machine learning). 

The presentation will first introduce the conception of the app and the pragmatic configuration of the individual tools. The second part focuses on the empirical study. Study design and evaluation methods will be presented and discussed, especially with regard to the implementation and application of the K-Means clustering algorithm for identifying user types and types of historical narratives. 

Vortrag im Rahmen des „Digital History“-Forschungskolloquiums
Zeit: Mittwoch, 14. Dezember 2022, 16-18 Uhr
Ort: Zoom-Konferenz (Zugang auf Anfrage oder via Mailingliste)

---

---