SUPERVISOR: pantelis papadopoulos
topic
Learning analytics has been widely used to transform vast amounts of data to pedagogically meaningful information. The context and goals of such uses vary significantly with studies focusing on how to monitor learning performance, support decision making, enhance peer interaction, improve self-learning, provide knowledge visualizations, and identify/predict students’ behavior. For teachers, the evaluation of a student audience state occurs copiously and continuously by combining information from (on-site and online) interactions and activities to identify learning needs. For students, the analysis of one’s self is a demanding task that requires developed metacognitive skills. As it happens, though, there is a lot of information that is not readily available or manageable for human actors. Data could be collected from a wide range of sources: educational tool logs, social media, online platforms, learning management systems, online communications, wearables, biometrics, and, of course, from (self-provided) data on surveys, quizzes, and assessment instruments, etc.
At a lower level, these sources could provide data on fundamental variables. Examples of such variables for an individual would be: number of posts submitted, length of individual’s post in words/sentences, response time to a peer’s post, timeline of key-terms mentioned in a group’s discussion log, etc. Likewise, variables for a group could include: number of discussions occurred in the group, size/type/length of products in the group, the size of the group, the aggregated profile of the group in terms of expertise/background/goals, etc. Based on the context, the list of fundamental variables can be further extended or restrained, while at the next level these variables would be combined into meaningful metrics. For example, a metric could provide information for an individual as a discussant, combining variables related to the number/size/quality of posts this individual has made in group discussions, the average response time, the number of discussions that have been initiated/resolved by the individual, etc. Finally, at a higher level, these metrics could be combined to answer more complex questions, such as “what is the contribution of the individual in the group?”. To answer this question, learning analytics would have to refer to metrics that would analyze the individual as a discussant, an author, a producer, an administrator, etc. As the same information can be presented in many ways, the use of helpful representations can also be based on the overall scope of learning analytics.
Some overarching research questions are: (a) how can learning analytics identify and provide meaningful information on the interaction between individual activity and the group’s progress, (b) what type of information could be more productive in personalized and group feedback, and (c) how should students and teachers be equipped for utilizing the feedback effectively?
METHOD
Depending on the research question and the targeted audience (individual students, groups, teachers), the study may be based on a comparative analysis of different study conditions or the impact of appropriate intervention within the same condition.
references
Aldowah, H., Al-Samarraie, H., & Fauzy, W. M. (2019). Educational data mining and learning analytics for 21st century higher education: A review and synthesis. Telematics and Informatics, 37, 13–49.
Van Leeuwen, A., Janssen, J., Erkens, G., Brekelmans, M., (2014). Supporting teachers in guiding collaborating students: effects of learning analytics in CSCL. Computers & Education, 79, 28–39.
Janssen, J., Erkensa, G., Kanselaara, G., & Jaspersa, J. (2007). Visualization of participation: Does it contribute to successful computer-supported collaborative learning? Computers & Education, 49(4), 1037–1065.