Over the past four decades, technology has played a crucial role in shaping the development of adaptive baseline assessments at Cambridge CEM.
I’m sure many of us might find it difficult to remember those times in education when standardised assessments in schools were paper-based, flat tests and often reliant on teacher-marking.
By harnessing the technology to digitise our assessment content, we made it easier for schools to administer our assessments as well as improving student engagement, accessibility and scalability. This new technology also made it possible for us to reach a larger number of students across the world and enabling the collection of real-time data on student performance.
At Cambridge CEM, this ushered in a new era of interactive and dynamic assessment questions that could adapt based on the students’ responses.
The development of our adaptive assessments has been made possible with sophisticated psychometric models like Item Response Theory (IRT), which means we can build adaptive tests that dynamically select items that are appropriately challenging for each test taker.
The advances in data analysis and psychometrics, powered by advanced statistical software and algorithms, have meant that we have an enhanced capacity to process and analyse large volumes of assessment data.
Technology has facilitated the integration of adaptive baseline assessments with personalised learning platforms and schools’ information management systems. By collecting data on individual student performance, our adaptive assessments support schools in the creation of personalised learning paths, offering tailored instructional content and interventions that align with each student's strengths and weaknesses.
Cambridge CEM has evolved from a small research unit based in the north of England working with just a handful of schools, to being part of Durham University, and now is proudly part of the Cambridge family, an organisation whose qualifications, assessments, academic publications, and original research spread knowledge, spark curiosity and aid understanding around the world.
“We are proud of what Cambridge CEM has helped teachers and students to achieve, and excited about its future."
Cambridge CEM now analyses a staggering 43 million data points each year. It is the huge amount of data we gather that gives teachers insight into their students’ abilities and makes our predictions so powerful. It is these predictions that help teachers set ambitious but realistic goals and then create the customised learning pathways to lead to improved outcomes.
Looking ahead, Artificial Intelligence (AI) is expected to play an even more prominent role in the future development of baseline assessments. But will AI become central to the day-to-day routines and processes of teaching and learning?
AI certainly offers many opportunities but should be pursued alongside ethical considerations.
One area of advancement in AI in assessment is around identification and prediction of misconceptions.
Nick Raikes, Director of Data Science at Cambridge University Press & Assessment, explains, “If we ask teachers to label common misconceptions or mistakes in a sample of answers, we might be able to train a model to recognise these so that it could give useful feedback to students attempting the question in future. Machine learning requires quite a lot of training data, though…”
This development in AI could expand our ability to not only assess knowledge and skills but also to identify misconceptions, identify the stage of misconception, predict when mistakes might happen, develop adaptive learning plans, and provide highly specific, immediate, and effective feedback and reasoning about students’ knowledge states.
“Cambridge CEM’s combination of research expertise, data-driven insights and excellence in delivering new technologies means it makes a real difference to educational progress in schools around the world."
Another area in the evolution of assessment may lie in the creation of holistic assessments, which can offer teachers a comprehensive cognitive, non-cognitive and meta-cognitive profile of their students.
Might AI help teachers to identify and measure the specific processes that matter most to better performance for all students across the whole ability spectrum?
Imagine AI-led assessment development that aligns and cross-pollinates the disciplines of psychology, curriculum, psychometrics, and technology which help to analyse students’ motivation, beliefs about learning, their attitudes to themselves and others, their approaches to learning and the varying psychological constructs that explain the variance in achievement.
Imagine an accurate, large-scale, cost-effective, time-saving pathway for truly personalised learning that gives primacy to the role of the educator and enhances the human experience of schooling.
Peter Phillips, Chief Executive of Cambridge University Press & Assessment, said: “We are proud of what Cambridge CEM has helped teachers and students to achieve, and excited about its future. Cambridge CEM’s combination of research expertise, data-driven insights and excellence in delivering new technologies means it makes a real difference to educational progress in schools around the world.”
After 40 years of assessment development, Cambridge CEM’s aims remain unchanged – to continue our commitment to harnessing the best technological solutions, to support teachers and create tools that generate insights and facilitate data-driven decision-making.