The Effectiveness of Adaptive Learning

In the current educational system, students are grouped by age and are instructed at the same pace, regardless of their actual level. But what if every student could re-visit concepts they have not yet mastered, or hone in on their unique challenges, whether or not their classmates were learning something different? What if students could progress to more advanced material at a faster pace, even if that material was beyond their grade level?

With adaptive learning, it’s possible!

Adaptive learning means modifying learning lessons and activities to meet the needs of every individual student. It allows them to learn at their own pace.

Adaptive learning apps put adaptive learning into practice by adjusting the activities they present to a student based on the student’s initial knowledge and performance as they go through the app. They are able to give personalized learning experiences to every student, much like a tutor can.

Research has shown that adaptive learning is effective for increasing student achievement (Barrus et al., 2012, Kimberely, 2018, Yilmaz, 2017).

Notably, the Barrus study found that for students who had previously failed high school algebra, just 4 hours per day of individualized instruction with adaptive learning apps yielded a 2-sigma improvement in their test scores within just 2 weeks!

Personalized learning through adaptive learning apps enables active learning, relies on mastery learning techniques, gives students the freedom to work at their own pace, allows one-on-one rapid feedback, and encourages student ownership of learning.

It’s no wonder that at gt.school we love adaptive learning apps!

 

Learning App Architecture: How Adaptive Learning Apps Adapt

But have you ever wondered how adaptive apps actually adapt? How can they personalize learning to each and every student?

To act as virtual tutors, learning apps adapt to the knowledge levels and performance of each student, just as human tutors do,, through the use of artificial intelligence and machine learning.

Adaptive learning apps adapt by:

  1. Assessing student knowledge
  2. Giving guidance and feedback to students
  3. Scaffolding student learning
  4. Personalizing learning paths

So how are they able to do all this? It’s all in the design!

Although the algorithms and design of each individual app are different, in general, a typical adaptive learning app (Figure 1) has four basic components, or modules (Alkhatlan and Kalita, 2018):

  1. An expert knowledge module, which stores knowledge of the subject taught
  2. A student module, which stores the current knowledge state of each student
  3. A tutoring module, which makes decisions about how to teach
  4. A user interface module, which interacts directly with the student

Figure 1: Learning App Architecture

 

Adapted from “Intelligent Tutoring Systems: A Comprehensive Historical Survey with Recent Developments” by Ali Alkhatlan and Jugal K. Kalita, 2018, 1:1, page 5

How Learning Apps Assess Student Knowledge

In order to adapt, apps need to first have a starting point- they need to know what each student knows.

Students are usually either assessed before they begin any learning material, or placed at a grade level (by a teacher or parent) and the algorithm adapts to the students level as they work through the learning material.

If the student takes an initial assessment, the results of that placement test will be recorded in the student module and used by the tutoring module to suggest appropriate lessons and activities for the student.

If the student is placed in a grade level without undergoing an initial assessment, an adaptive learning app will determine the student’s real knowledge level based on how they perform on initial activities. This is done through machine learning, which occurs when the algorithm is able to improve automatically through experience or gathering data.

The student module is then updated accordingly so that the tutor module can continue to suggest appropriate activities for the student, which may be at a higher or lower level than the initial placement level.

How Learning Apps Give Guidance and Feedback

The knowledge module stores all of the educational content: the questions, explanations, and responses. It evaluates the student’s performance by comparing the student’s answers to the correct answers encoded in the module.

The tutoring module, on the other hand, determines how to teach the student, deciding how to present the teaching material to the student and provide feedback to the student.

When a student completes a problem, there is an interaction between the knowledge module and the tutoring module. For example, if the student inputs an incorrect answer, the knowledge module will signal to the tutoring module that an incorrect answer was received. The tutoring module will then decide how to provide feedback to the student: should they be presented with a hint, a partial answer, or a fully guided solution?

How Learning Apps Guide Student Learning

Scaffolding refers to “ guided prompting that pushes the student further along the same line of thinking” (M. T. H. Chi et al., 2001). It is an interactive process between the student and tutor (human or learning app!) that allows the student to gradually take steps to perform the skill entirely by themselves.

Adaptive learning apps scaffold students’ learning by deciding how to present the teaching material to the student. This may be through increasingly difficult problems on the same topic, more specific hints as to how to solve a problem, or partially worked through explanations to problem solutions.

The tutoring module is the driver of scaffolding in an adaptive learning app, acting as an intermediary between the knowledge module (the subject material and actual questions) and the student module (which tracks what the student knows and is ready to learn).

Personalizing Learning Paths

Although feedback and scaffolding are forms of adaptivity and individualization, they are not the same as individualized task selection- this is where the real personalization of learning occurs.

Most adaptive learning apps create a personalized learning path for their students. The student module records the initial knowledge state of the student (usually through a placement test) and then updates that knowledge state as the student moves along their personalized path and masters the material.

The tutoring module personalizes learning paths by gathering information from the knowledge module on the students’ behavior (for example how many questions the student answered correctly on a test) and from the student module on the student’s knowledge state (for example what concepts the student has already mastered). It then determines which activity to give each student next, just as a human tutor would tell a student to watch a video, practice a problem or take a test.

Adapting the Development of Adaptive Learning Apps

As fascinating as the current adaptive process for learning apps is, it’s only the beginning!

At present, developing the algorithms behind adaptive learning apps is difficult, costly and labor intensive.

The challenge with developing adaptive learning apps is that the apps need to know every kind of way to solve a problem, since different students will approach and learn how to solve problems in different ways.

Fortunately, researchers at Carnegie Mellon University have now designed a way to teach computers how to teach, which they say will enable rapid development of future adaptive learning apps (Weitekamp et al, 2020).

The current method of developing adaptive learning apps involves programming solution rules into the knowledge module (which takes a very long time and there is the potential for possible solutions to be omitted by accident).

The new method the researchers have developed uses a machine learning program that simulates how students learn.

A developer can teach the computer by demonstrating several ways to solve problems for a given topic, and then can correct the computer if it responds incorrectly.

The computer then “learns” how to solve that type of problem, and can even generalize to solve the problem in different ways than it was taught, much like a human student might.

The amazing result is that the computer itself writes the solution rules.

So not only is the possibility of missing solutions eliminated, the time taken to develop adaptive learning apps is dramatically reduced.

Personalized learning through adaptive learning apps is about to get a whole lot better!