How assessment can help tackle learning loss
A blog by Micheal Heffernan, Learnosity
Learning loss has taken on a new dimension in the wake of COVID-19. It’s time for assessment to do the same.
It’s been referred to as “the largest disruption to education in history”, a claim that gains justification as the weeks and months go by.
The impact of COVID on learning has been so extreme that it has induced countermeasures that would have previously seemed radical—maybe even reckless—such as the rapid commencement of “the largest remote learning experiment in history”. For an industry largely beholden to convention and consistency, the words “disruption” and “experiment” are like unwelcome gate-crashers at learning’s immovable feast. While many businesses could fall back on existing tech infrastructure for some measure of continuity during the pandemic, the world of learning waded into uncharted waters. Now more than a year on, COVID’s impact on learning is just starting to emerge.
Patterns of disturbance
A familiar concept to those working in the field of education, learning loss refers to the loss of knowledge and skills caused by extended gaps or discontinuities in a student’s education. This is most often an area of concern following the summer break period.
But the concept of learning loss has taken on a new dimension in the context of COVID. The pressure cooker environment created for learners, educators, and parents has seen them hastily adapt to new ways of living, working, and learning. A national study in the US of over a million students found that the number of elementary school students performing at grade level in both reading and math had deteriorated by 10% and 16% respectively.
A separate analysis of the pandemic’s impact on learning (by management consultant firm McKinsey and Company) supports this, revealing that by the end of the school year, students were an average of five months behind in mathematics and four months behind in reading. The authors of that report claim that unless remedied, the long-term fallout of COVID-caused learning loss could be catastrophic: “We anticipate a potential annual GDP loss of $128 billion to $188 billion from pandemic-related unfinished learning.”
Reassessing learning loss
Such projections fuel very real anxieties around learning loss, but the concept itself has come under increasing fire. Some critics argue that measuring learning loss during a pandemic may do more harm than good by undermining learners’ confidence and unfairly stigmatizing an entire generation of learners as a “lost generation”.
Others have raised concerns about the pandemic’s role in exacerbating pre-existing economic divisions. Studies from the US, Germany, and the Netherlands all converge on the point that learners from low-income backgrounds are less likely to engage in remote learning than their peers from higher-income backgrounds. This is often due to the serious challenges they face at home on issues such as income and job insecurity, as well as a lack of access to the right learning tools—or support in using them.
Given that learners in this group participate irregularly in online learning, standard performance measures offer limited visibility into their progress or learning potential. Indeed, the most academically vulnerable may not be represented in the research at all because they have either disengaged, not been tested, or have dropped off public school enrollment rolls altogether. Without a concerted effort to re-engage learners, it’s estimated that up to 1.2 million learners across grades 8-12 will drop out of the US school system entirely.
Mind the gap: changing expectations
Part of this re-engagement will occur naturally as classrooms reopen and learners rejoin their peers in face-to-face learning. But that alone won’t be enough—either over the shorter or longer terms.
One reason why is that the conversation around pandemic-induced learning loss overlooks the uncomfortable truth that reading and math performance in US schools has been stagnant since 2000. A return to “normality” would be more of a step sideways than a step ahead. Furthermore, as the last year and a half have shown us, infrastructure is critical to continuity. Those who invest in digital learning have more protection against the potential future disruptions. The fear, however, is that those who don’t invest will be left on the wrong side of a fast-widening digital divide.
This is evident from preliminary studies on the effectiveness of remote learning. While teachers in high-poverty schools gave it an average global score of just 3.5 out of 10, their peers in private schools—which tend to have greater access to online tools—awarded it a far more favorable average of 6.2. Should things continue along this trajectory, disparities in learning will become more apparent as early adopters grow more familiar with the capabilities technology affords them. The innovators will move on, leaving others to uphold a status quo no longer fit for purpose.
Creating modern learning experiences
This gap will likely extend into other crucial areas where learning is evolving, such as assessment.
As a yardstick of learner achievement, traditional assessment requirements are both familiar and static. But the landscape is changing—and so too are learner expectations. Technology is now driving assessment toward something much more fluid and dynamic, making it a fundamental building block of modern learning experiences.
Our work at Learnosity advances assessment along these lines. By developing an assessment engine that works in concert with other systems, our aim is to level the playing field by offering instant access to breakthrough capabilities and continuous innovations. These advances are designed to support learning by addressing crucial issues such as student engagement, accessibility, inclusivity, and measurement—all of which have been identified as contributing factors to the problem of learning loss.
Using assessment to mitigate against learning loss
Connect learners with learning
Remote and hybrid learning models offer essential evolutionary pathways for learning but they aren’t without problems.
According to an Education Trust poll, almost 50% of low-income families and 42% of families of color do not have access to the devices they need to participate in online learning. But research shows that while ownership of home computers has unequal distribution, mobile ownership is prevalent across demographic groups. Adoption is increasing in emerging economies too. To engage learners means meeting them where they are. More flexible delivery methods make this possible, offering consistently rich experiences across device and browser types.
Supporting offline assessment delivery also goes some way toward remedying the broader issue of intermittent internet connectivity, making it possible for learners to start assessments online and finish them offline.
Accommodate learner diversity
Education is a fundamental human right. If digital learning is to deliver on its promise to support more learners then it will need to support a diverse range of needs. This presents a challenge of vast scope.
With over a billion people worldwide living with some form of disability and some 7.3 million (14%) of all public school students in the US receiving special education services under the Individuals with Disabilities Education Act, accessibility is a high-priority area.
Making a learning product more accessible means making it more usable; inclusive learning products are simply better learning products. This is why accessibility is simply baked-into Learnosity’s development process. This allows us to meet—and often exceed—accessibility standards by supporting learners with visual, auditory, and motor impairments through functionalities such as:
- Color contrast and adjustable font sizing
- Intuitive keyboard usability for accessible input and navigation
- Screen reader support
- Transcripts, closed captions, and alternate videos
However, accessibility is a moving target in technology and there will always be work to do in removing barriers and reaching more learners. Progressing toward that goal requires sustained commitment.
Make assessment continuous, not just conclusive
Reduced levels of engagement is a recurring theme in the research on learning loss. Zoom fatigue, online distraction, and a disconnection with learning content are listed as impediments to engagement. But the reality is that if you want learners’ attention, you’ve got to earn it.
Technology makes this possible. Swapping in dynamic, interactive experiences for basic MCQs helps revitalize assessment content for modern learners and cater for their individual learning styles and strengths. Additional possibilities such as instant feedback and adaptive testing help further personalize the assessment experience, giving learners a greater sense of ownership and accountability.
The payoff of more engaging, interactive assessment is that learners will use it more often, effectively embedding it into the overall learning process. In turn, continuous assessment improves learning retention, reduces test anxiety, and generates a stream of valuable learner insights.
Information improves decision making. Digital assessment’s potential impact on learning centers on its capacity to measure the context for learning in granular detail, identifying individual needs and illuminating hard-to-reach areas of the learning process.
As learners increase their engagement with assessments, their activity yields further data and insights. These can be used to shine a light on things like learner progress and content performance—both of which inform instruction and support. It can also equip teachers and support staff with information that can identify students who have disengaged from instruction or who are at risk of dropping out entirely.
Furthermore, as educators improve their data fluency, they will also become increasingly adept at building comprehensive student portfolios, thereby moving beyond a focus on learner limitation to better understand and unlock learner potential.