By Vitomir Kovanović, Srećko Joksimović and Dragan Gašević
Schools of Informatics and Education, The University of Edinburgh
1 Problem statement
One of the widely used theoretical models that describes different dimensions of online learning is Community of Inquiry (CoI) framework (Garrison, Anderson, & Archer, 1999). In short, CoI framework defines three important presences that shape online learning experience: i) Cognitive presence, ii) Teaching presence, and iii) Social presence. The central component of the model is cognitive presence, which is associated with development of students’ critical thinking and construction of meaning through collaborative inquiry process. To assess the levels of three presences researchers either use quantitative coding schemes for manually coding student discussion messages, or post-course survey instrument. As such, it is not possible to use CoI framework to provide insights into the social learning processes as they unfold and was primarily used for post-course analyses.
2 Proposed solution
The goal of the proposed project is to enable use of CoI framework for continuous monitoring of student discourse. Using text mining and text classification techniques (Aggarwal & Zhai, 2012), we aim at developing an automated content analysis system which would label each student discussion message in accordance with CoI coding scheme. The focus of the proposed project is on the cognitive presence, as it is the central construct of the framework and typically, the one of the most importance.
There are several direct benefits of the proposed project:
Firstly, the development of automated content analysis system would enable continuous monitoring of student discussions. This automated codes could then be used to develop novel learning analytics visualizations and dashboards that provide insights into the student learning processes.
Secondly, the development of automated content analysis system would enable the easier and wider adoption of Community of Inquiry framework by both educational researchers and practitioners. As CoI framework was originally used by educational researchers, it requires trained coders and a labor intensive and expensive manual coding of discussion messages. The automated system would lessen the burden for CoI adoption by providing a mechanism for easy and fast message coding.
Finally, the automated coding system would provide insights into the cognitive presence construct by providing low-level operationalization of the framework. As CoI coding scheme was intended to be used by educational researchers, it requires an extensive background knowledge for its interpretation and use. The development of automated coding system would as a side product provide a better operationalization of CoI coding scheme which is also a valuable theoretical contribution of our study.
3 Current progress
Our initial investigation (Joksimović, Gašević, Kovanović, Adesope, & Hatala, 2014; Kovanović, Joksimović, Gašević, & Hatala, 2014) of automating CoI coding scheme showed a potential for the proposed solution. Using Cox-metrix (Graesser, McNamara, & Kulikowich, 2011) and LIWC - Linguistic Inquiry and Word Count (Tausczik & Pennebaker, 2010) tools, we identified a set of highly relevant linguistic properties that are predictive of different levels of students’ cognitive presence. However, our study showed the need for more contextualized set of predictors that would model structure of threaded discussions and students’ progress through different phases of cognitive presence. The Jics project funding would thus help us to further progress in development of automated coding system, expanding our understanding of cognitive presence and enable for broader adoption of CoI model by educational researchers and practitioners.
Aggarwal, C. C., & Zhai, C. (2012). Mining Text Data. Springer.
Garrison, D. R., Anderson, T., & Archer, W. (1999). Critical Inquiry in a Text-Based Environment: Computer Conferencing in Higher Education. The Internet and Higher Education, 2(2–3), 87–105.
Graesser, A. C., McNamara, D. S., & Kulikowich, J. M. (2011). Coh-Metrix Providing Multilevel Analyses of Text Characteristics. Educational Researcher, 40(5), 223–234.
Joksimović, S., Gašević, D., Kovanović, V., Adesope, O., & Hatala, M. (2014). Psychological characteristics in cognitive presence of communities of inquiry: A linguistic analysis of online discussions. The Internet and Higher Education, 22, 1–10.
Kovanović, V., Joksimović, S., Gašević, D., & Hatala, M. (2014). Automated Content Analysis of Online Discussion Transcripts. In Proceedings of the Workshops at the LAK 2014 Conference co-located with 4th International Conference on Learning Analytics and Knowledge (LAK 2014). Indianapolis, IN. Retrieved from http://ceur-ws.org/Vol-1137/
Tausczik, Y. R., & Pennebaker, J. W. (2010). The Psychological Meaning of Words: LIWC and Computerized Text Analysis Methods. Journal of Language and Social Psychology, 29(1), 24–54.