Artificial Intelligence for Learning Analytics and Instructional Design Steps: An Overview of Solutions
Artificial intelligence methods are getting frequently used in research and development in learning analytics, which is aimed at analyzing data collected during learning to enhance its results. The same aim is relevant for instructional design models, the most widely applied is ADDIE model, which cu...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
National Research University Higher School of Economics (HSE)
2022-12-01
|
| Series: | Вопросы образования |
| Subjects: | |
| Online Access: | https://vo.hse.ru/article/view/16518 |
| Summary: | Artificial intelligence methods are getting frequently used in research and development in learning analytics, which is aimed at analyzing data collected during learning to enhance its results. The same aim is relevant for instructional design models, the most widely applied is ADDIE model, which cuts course design into steps. The first two research fields are criticized for a weak connection to teaching practice, while the third lacks evidence-based and measurable nature. This literature review aims to show the prospects of bringing the three fields together. The theoretical analysis of the paper covers AI definition, its techniques and methods, areas of application in educational setting, the definition of learning analytics, its borders with other fields, spheres of application, definition and the essence of instructional design, as well as the concept of ADDIE model which frames the practical analysis of the review. Forty-three articles were included in the final sample. The solutions described there correlate with the tasks of the instructional design steps and are systematized according to them. It was found that the least number of solutions described in the literature were assigned to the analysis, design and evaluation steps, more articles were assigned to the development step, and the largest number of papers considered the application step. It can be due to the difference in the availability of data at different ADDIE steps. The weak focus of teachers on methodological reflection at the assessment step also may play a role. These deficiencies open up the opportunities for future research and developments. To push these solutions forward it is crucial to elaborate on the models, to move from anecdotal experiments to a wide-scale practice, and to enhance required competencies among the faculty. The questions and conclusions presented in the article help to set a new pedagogically-oriented framework for discussions of AI and learning data analytics. |
|---|---|
| ISSN: | 1814-9545 2412-4354 |