USE CASES FOR LEARNING ANALYTICS IN SELF-LEARNING COURSES
Stuttgart Media University (GERMANY)
About this paper:
Conference name: 9th annual International Conference of Education, Research and Innovation
Dates: 14-16 November, 2016
Location: Seville, Spain
Abstract:LeMon is a tool implementing the idea of classroom response systems (CRS) using student’s own smartphones. At the ICERI 15, the results of a practical usage of this tool and an extension to a self-learning module have been presented. The learning process using the self-learning module consists of the following steps for each topic of a course:
1. Self-assessment: the student starts with a self-assessment using LeMon. Depending of his success rate, the further steps of the learning process are recommended to him by the module:
a. Theory: If he answered less than 50% of the questions correctly, he has to repeat the theory of the topic at first.
b. Exercises with several levels: If he answered more than 50% of the questions correctly, dependent on the exact quota of correct answers, different levels of exercises are recommended (level 1: repetition of the course topics, level 2: comprehension, level 3: transfer).
2. Learning success assessment: At the end, an additional closure assessment is performed using LeMon. Contrary to the first self-assessment, the questions of the second assessment can only be answered by entering some key words or some longer texts.
Using the self-learning module, information about the student’s learning behavior is being collected. Based on this and using some learning analytics approaches, two use cases are considered for a further development of the self-learning module:
1.A personalized recommendation component for the steps after the self-assessment:
Learning analytics will be used for the recommendation component. Hence, the recommendations for a student will be based on personalized information about him, e.g.:
a. The student’s motivation degree based on the intensity he uses the self-learning module,
b. The student’s learning strategy based on the recommended and the real learning path,
c. The student’s time investment based on the time periods in which the student uses the self-learning, e.g., continuously or only shortly before some due dates.
2. A recommendation component for learning teams:
Learning analytics can also be used for a recommendation of learning teams. The question is here, how to bring together two students to achieve the best learning success (students with a similar level, students with different levels, small or bigger teams).
A second question is about how the students met each other in the real word, because the self-learning module is used in an anonymous way.
Two possible approaches will be discussed in this paper:
a) Introducing a personalized mode of the self-learning module: the students have to login using their real name. However, this can be only done on a voluntary base.
b) Introducing a chat component: The recommendation module only establish a chat connection between the students which match together. The students can then exchange their personal data via the chat and also meet in real world, but they won’t know the criteria why they were brought together by the tool.
In the proposed paper, the learning analytics approaches will be described in more details. Also the approaches for the establishing of learning teams will be discussed.