Adaptive Learning: Personalized Learning on a Large Scale

By Dr. Henry E. Schaffer, Dr. Karen R. Young and Dr. Maria T. Gallardo-Williams

Students’ success in their college courses is often a reflection of their engagement and participation during the semester, and it requires instructor feedback as well as sustained practice. One challenge to this is the reality of resource allocation in some higher education settings. Delivering personalized feedback to individual students is one of the big challenges that educators tasked with the responsibility of teaching large classes face today. Adaptive learning is a tool that can be leveraged to address this issue. It incorporates aspects derived from various fields of study including computer science, AI, education, and psychology.

Adaptive learning is the use of computer algorithms to deliver customized resources and learning activities to address the individual needs of each learner. The presentation of educational material can be adjusted to students’ learning needs, as indicated by their responses to questions and tasks. In this way, it turns the students from passive consumers of information into partners in the educational process, since the adaptive learning algorithm is used to deliver information according to student performance in the course. In its simplest form, adaptive learning requires clear concept maps of the content that students are expected to master, coupled with assessment items that can be mapped to the nodes in the course concept map. From homegrown strategies (1,2), to incorporation in the LMS (3), to commercial products (4,5), there are a large number of adaptive learning solutions available today. However, they all require resources, whether money or time and effort or all of these and so the instructor must balance this with benefit.

Important considerations include whether the textbook and services must be purchased by the campus and/or the students, the monetary costs, if any, and the amount of time required from the instructor. The instructor time required should offer “economies of scale”. That is, the instructor’s time and effort required for this computerized feedback should not increase with an increase in the size of the class, as this allows support of the students in large classes. Perhaps more important is the support for learning given to each student. Ideally, the student will utilize the concept maps of the content along with the feedback to improve learning. The instructor can help motivate this by utilizing the time saved to help all the students appreciate the course content in terms of their own development and careers and so increase motivation and, therefore, engagement.

In our own experience using an open-source adaptive learning system (FormAssess, see references 1 and 2) we found that the instructor and students thought that it was helpful to student learning. The instructor appreciated that it allowed for providing each student individualized feedback. This had not been possible previously because of the large class size. The students appreciated that it allowed them to see strengths and weaknesses in their understanding of course topics and it increased their confidence that they could do better in the course going forward.

References

  1. Schaffer, H., Young, K.R., Ligon, E.W., & Chapman, D. (2017). Automating individualized formative feedback in large classes based on a directed concept graph. Frontiers in Psychology, 8. doi: 10.3389/fpsyg.2017.00260 Note: has a link to the Open Source software described.
  2. Young, K.R., Schaffer, H.,James, J.B., & Gallardo-Williams, M.T. (in press). Tired of failing students? Improving student learning using detailed and automatic individualized feedback in a large introductory science course. Manuscript accepted for publication in Innovative Higher Education. doi: 10.1007/s10755-020-09527-5.
  3. Moodle – https://moodle.org or https://wolfware.ncsu.edu
  4. https://realizeitlearning.com/
  5. https://www.dreambox.com/adaptive-learning