When schools went completely virtual due to COVID-19, it exposed existing cracks in our education infrastructure. Part of the problem was a lack of tools and connectivity—schools and students needed tablets, laptops, and a reliable household internet connection.
But the other problem the pandemic exposed was with the way we teach. Despite the capabilities of digital education tools, we aren’t leveraging them to their full potential. Here’s how we can harness the power of digital learning tools to make the education system work for more students.
The problem with the current learning ecosystem
The current classroom model treats all students as if they start with the same level of knowledge and learn at the same pace. Teachers stand in front of a whiteboard and lecture students, who sit passively in neatly arranged desks or in lecture halls. This “factory-model” approach hasn’t changed much in over a century, despite sweeping innovations in technology and new insights about how we learn. Plus, the nature of work and the kinds of jobs needed to solve today’s and tomorrow’s complex problems have changed dramatically. The current classroom model leaves many students ill-equipped for the challenges they’ll face.
Even as schools adopt education technology, teachers lack training for digital environments, and the traditional classroom model doesn’t translate well to digital formats. As a result, we based our approach to computer-based instruction on a disjointed ecosystem of capabilities:
- We’ve effectively established protocols to map out what information students should learn at each level of their education.
- We understand how to use assessment tools like standardized testing to test individual levels of mastery.
These advances allow us to establish and measure how individual students progress, but we haven’t changed the learning paradigm to accommodate those insights. To drive the transformation needed, these need to be integrated.
Adaptive Instructional Systems can transform the classroom experience…
Adaptive instruction turns the traditional model on its head and focuses on tailoring education resources to each student’s individual needs. A study by Fulcrum Labs found students’ course completion rates were 15% higher than with traditional online instruction. Adaptive instructional systems (AIS) enable us to leverage existing computer-based instruction practices to make adaptive instruction more effective and efficient.
We can already map out specific domains of instruction and measure proficiency on an individual level. AIS can close the loop and make it possible to apply those insights to the learning experience. For example, using artificial intelligence and machine learning algorithms, AIS can offer recommendations based on students’ behaviors and assessment scores. If a student fails to complete a lesson, the AIS may suggest confidence-building exercises to motivate them to finish.
… but there are a few obstacles
However, current AIS solutions are limited. Most are created to cover a particular topic and thus are too specialized, such as technical training for mechanical engineering students. As a result, the AIS works well for the creator’s domain or area of specialty—in this case, the college’s engineering department—but it doesn’t translate to other areas of knowledge. That limits AIS adoption.
There are other obstacles when it comes to expanding AIS adoption. Since many solutions do not scale, the results are limited even within its domain. On top of that, AIS solutions are in their infancy, which means there are no agreed-upon standards. Without accepted industry-wide standards, concerns about privacy, interoperability, and ethics, evaluation will remain. Despite these hurdles, the potential for AIS is enormous. Working out issues around standards and integration can drive sweeping change in the classroom model.
What a fully realized, collaborative AIS can look like
Combining existing computer-based instruction practices with machine learning (ML) algorithms and artificial intelligence (AI) can create a more tailored experience for students and enable teachers to work more flexibly.
The student can enter the AIS and create an avatar (for protecting personally identifiable information or PII). An AI agent may greet the student and guide them through the lessons or tasks they need to complete for the day. The persistent learner model constantly collects data on the student’s activities and aptitude levels to understand what drives (or hinders) their learning. AI and ML algorithms can feed the student recommendations based on that data, and the teacher can use that information to provide targeted instruction as they float between students.
An AIS that can do this does require multiple tools on the backend, such as a group composition engine and course authoring tools. And while those tools may themselves be standalone applications, the AIS can provide the framework to bring them together and create a seamless experience for both the student and teacher.
This approach makes AIS more accessible and allows these systems to be applicable across multiple competencies and domains. So, we might use an AIS initially developed to train mechanical engineers for the entire engineering department—or even for unrelated disciplines across the institution. The more students who have access to integrated AIS, the more we can leverage their full potential value.
Bringing these technologies to life, of course, requires partners who understand the concepts behind computer-based instruction and have expertise in creating customized education software. JBS Solutions has extensive experience developing custom software solutions for educational institutions of all sizes.
To get your project started, contact us today for a free consultation.