Customizing education requires better assessments

By:

Jan 9, 2015

One exciting development that online learning facilitates is the opportunity for greater customizability. Online learning gives students access to a wide selection courses and learning resources beyond those available through their local schools. Additionally, online learning enables blended-learning models that make it possible to customize the time, place, path, and pace of instruction in order to tailor learning experiences to each students’ individual learning needs.

Yet one of the challenges to providing students with a fully customizable education is figuring out how to knit the modular learning experience from various sources together in a cohesive way. When the components of a student’s overall learning plan come from independent providers, those providers must be able to make predictable handoffs. At each stage of instruction, they must be able to confidently expect that students have gained certain knowledge and skills, regardless of where each student received his last segment of instruction.

For example, if students take a sequence of courses from a variety of online and face-to-face course providers, how can the teachers of the courses that come later in the sequence be sure that earlier courses covered all the prerequisite knowledge and skills and that the students have truly mastered them? Similarly, when learning experiences are not all acquired at a single school, how does the school that ultimately graduates a student confidently decide to award a degree?

To some extent, our current system is already designed to facilitate cohesiveness across various modular learning experiences. Commonly used course names and credit hour specifications currently provide the interfaces for determining how individual courses fit into a broader plan of instruction. For decades, schools have used the credit hour system to standardize their courses across different teachers and to accept transfer credits from other schools.

The credit hour system has long been our best working solution for facilitating modularity and customizability in education. Yet a number of breakdowns in the system demonstrate that it is far from ideal. Consider the fact that even with common course names, numbers, and credit hour designations, two supposedly identical courses can be very different learning experiences, depending on the teachers who teach them. When selecting courses for their schedules, high school and college students have a keen ear for figuring out which teachers are easy, which are the most friendly and helpful, and which give the most engaging and memorable lessons.

The letter grades we assign students also point to breakdowns in the predictability of the credit hour system. The reason we use letter grades is that completing the seat-time requirements for a course by attending all the lessons and submitting all the assignments is not a reliable indicator of actual learning. Letter grades are effectively quality control stamps that deal with this unpredictability by indicating the quality of students’ learning after completing a course. (Imagine what it would be like if manufacturing companies took a similar approach. Instead of warranting that their products were free from defects, what if they churned out products with huge variations in quality, and then just labeled them according to their degree of defectiveness before putting them on store shelves?)

Letter grades are practical workaround for the unpredictability of credit hours as signals of learning, but they too have problems when it comes to providing reliable signals. The assessment and grading methods that a particular course instructor adopts heavily bias the letter grades assigned by that instructor. Additionally, the signals of quality learning that letter grades are supposed to provided are frequently muddied by practices such as curving and extra credit.

As one more evidence of the unpredictability of the credit hour system, consider the fact that teachers often spend the first few weeks of their courses reviewing material from previous courses. They do this because they know that many of their students have gaps in their prerequisite knowledge, even though all the students should have completed the prerequisite coursework.

The reality is that credit hours and letter grades are not good signals of actual learning, and this creates some major inefficiencies in our education system. It would be great if we could customize not only the courses in a student’s learning plan, but also the individual learning activities within a course and the pacing of each student’s work through those learning activities. But in order to have that kind of a modular, customizable education system, we need much better ways of measuring and tracking student learning so that smooth interchanges between various learning experiences can happen with predictable success. We need competency-based assessment systems that have enough validity and reliability to garner acceptance and adoption across the entire ecosystem of learning module providers. Only then will we realize the benefits of customizable learning promised by forward-looking innovations in online learning.

As we envision a future education system with modular customizability, we need to recognize that getting to that future will require investment in innovations to develop better competency-based assessment and tracking systems.

Thomas Arnett is a senior research fellow for the Clayton Christensen Institute. His work focuses on using the Theory of Disruptive Innovation to study innovative instructional models and their potential to scale student-centered learning in K–12 education. He also studies demand for innovative resources and practices across the K–12 education system using the Jobs to Be Done Theory.