Teaching Analytics: A Clustering and Triangulation Study of Digital Library User Data
Year: 2012
Type: article
Abstract: Teachers and students increasingly enjoy unprecedented access to abundant web resources and digital libraries to enhance and enrich their classroom experiences. However, due to the distributed nature of such systems, conventional educational research methods, such as surveys and observations, provide only limited snapshots. In addition, educational data mining, as an emergent research approach, has seldom been used to explore teachers' online behaviors when using digital libraries. Building upon results from a preliminary study, this article presents results from a clustering study of teachers' usage patterns while using an educational digital library tool, called the Instructional Architect. The clustering approach employed a robust statistical model called latent class analysis. In addition, frequent itemsets mining was used to clean and extract common patterns from the clusters initially generated. The final clusters identified three groups of teachers in the IA: "key brokers", "insular classroom practitioners", and "inactive islanders". Identified clusters were triangulated with data collected in teachers' registration profiles. Results showed that increased teaching experience and comfort with technology were related to teachers' effectiveness in using the IA. (Contains 5 tables and 1 figure.)
Authors: Beijie Xu, Mimi Recker
Language: English
Subfield: Computer Science Applications
Field: Computer Science
Domain: Physical Sciences
SDG: Quality education
Open Access status: diamond
OpenAlex ID: W2166168704