WS F: Practical Applications of the WebApp EduTopics:ECER for Literature Search, Research Information and Research Projects

Since 1998 ECER has been a household name for (European) educational research and especially for emerging researchers. It provides a platform for presentations and discussions of current and past issues from multiple different (disciplinary) perspectives. Its database of over 30,000 contributions over 25 years – including the bibliographic parameters (i.e. authors, EERA-networks etc.) and abstracts – represents a unique window into the field of educational research and it’s (historical) thematic developments. To identify those themes and to enable access and information on the thematic and bibliographic development of ECER the open-access webapp EduTopics: ECER (Christ et al., 2024; url: https://dipf-lis.shinyapps.io/EduTopicsECER/) applied topic modelling (Blei et al., 2003; Griffiths & Steyvers, 2004) to determine the central content-related themes of all contributions ranging over 25 years. This natural language processing method is a parallel clustering approach which (a) identifies terms frequently occuring (together) across documents and (b) clusters documents into thematic groups according to the terms identified in (a). Therefore, it assigns documents probabilities for each topic and enables the identification of topic trends over the years, relevant authors of topics or network/country-topic-probabilities. The app harnesses those possibilities by providing interactive and user-driven manipulable visualizations for the various resulting uni- and bivariate distributions of all combinations of topics and bibliographic parameters.

The focus of this workshop is on the practical application of the aforementioned app for interested researchers. Initially, the applied methods of topic modelling – more specific: latent Dirichlet allocations – as well the different types of clustering and visualizing techniques (for example graph-drawings) are explained in runtime. Following, the participants are given practical instructions to gather information with the app regarding narrow search questions. This is followed by an open practical segment in which the participants can follow their own individual research questions under the guidance of the presenters.

Beyond the methodological and practical introduction into the app and the larger method-family of natural language processing, this workshop provides insight into “new” ways to approach analyses of text collections and into new avenues of research based on the results of such analyses.