Advantages And Disadvantages Of Learning Analytics

- Prediction: It is often used to predict when a participant is expected to drop from an online course. This could be done by analyzing a student behavior, exam performance, and video skips. Storing numerous records of previous students’ activities based on specific modules help researchers predict the prospective action, such as dropping out of a course or detecting students, who are at risk. Furthermore, learning analytics can be used to predict the performance and motivation of the students.

- Recommendations: These are generally suggestions. For example, a MOOC provider may recommend learning materials to students based on their previous registered courses. These recommendations can also be generated to suggest a student answering a specific

Papamitsiou and Economides (2014) emphasize some weak points of learning analytics in their work. For example, misinterpretation of results because of human decision-making factors is one of the weak points in learning analytics. Hence, it would be useful to focus on the reports that were issued before the decision phase, not on the decisions made. Since heterogeneous data sources are utilized in learning analytics, a dictionary to define the data as the representation is not yet available. Therefore, the problem of data representation is another weak point of learning analytics. It is observed that learning analytics often include quantitative research and its results. Qualitative research only constitutes a minority. On the other hand, Khalil, Taraghi & Ebner (2016) listed the disadvantages of the learning analytics as

One of the areas where the application of learning analytics is needed is open and distance learning. Learners in open and distance learning perform learning activities and a set of interactions electronically and leave behind some electronic traces after each action. These traces are a prominent source of data for enhancing learners' academic performance, estimating prospective academic steps, improving open and distance learning platforms, and testing teaching competencies as well. Analyzing large amounts of data from learners containing millions of numbers, using traditional methods makes it challenging to get answers to some questions. Hence, the use of learning analytics seriously supports the decision makers and practitioners in achieving the desired objectives. Despite all the positive aspects, some questions and problems in the implementation of learning analytics still exist:

• Although the confidentiality of personal information is very popular nowadays, it is not yet clarified which personal information would be used in the learning analytics and how to avoid including this mistaken information in analysis and reporting. In addition, there is no standard commitment for institutions to comply with, on what information will be shared, for which purposes and how the information of learners will be shared and with which