Mastery learning: learning all the way to mastery

Most corporate courses end in the same way, with a final quiz you pass by reaching a minimum score, a certificate that arrives shortly afterwards and the feeling of having ticked a box. As a rule nobody stops to ask whether the person has really understood the topics on which their answers were shaky, because the system is built mainly to record that the test was passed, and it leaves in the shadows how much the learner actually took in. Mastery learning starts from a different premise and tries to put that understanding back at the centre, building on a principle so straightforward it can seem obvious, namely that you move on to a new topic only once you have properly learned the one before it.

An idea with a long history behind it

The concept has deep roots and was around long before training software existed. It was formalised in the late 1960s by the educational psychologist Benjamin Bloom, who believed that the great majority of students could reach a high level of mastery, provided they were given the time, the feedback and the chances to correct their mistakes that suited their own needs. This conviction became famous through a 1984 paper in which Bloom compared three ways of teaching, namely the traditional classroom, mastery learning applied to a group, and one-to-one tutoring.

Bloom summed up the experiment with a figure that would become well known, according to which students tutored one to one with mastery learning techniques achieved results two standard deviations higher than those of a conventional class. That number referred to the combination of individual tutoring and mastery, and it has since been much debated and never reproduced with the same clarity, a caution worth keeping in mind. The part that matters here holds up well over time, however, because mastery learning on its own, applied to a whole class through tests, feedback and correction, showed a gain of close to one standard deviation by itself, and later meta-analyses have confirmed the picture. The one by Kulik and colleagues, built on more than a hundred controlled studies, found positive and consistent effects of around half a standard deviation on average, with the clearest benefits going to the students who started from a weaker base.

Stripped of its academic language, the method comes down to an orderly sequence of steps that repeat for as long as they are needed. You teach a piece of content, you check it with a test, you give feedback on the parts that did not work, you work on correcting the errors and you check learning again, and only once a threshold of mastery has been reached do you move on to the next topic.

Why it works so well

The strength of the method lies in the way it treats error, because it regards a mistake as the most useful piece of information in the whole journey, a kind of compass that points precisely to where it is worth going back to work. Returning to a question that was answered incorrectly, understanding the reasons for the slip and trying again corresponds to that active retrieval which research on memory has long identified as one of the most effective study activities of all. The studies by Roediger and Karpicke on the testing effect have shown that calling a piece of knowledge back to mind fixes it far more firmly than simply rereading it, and the feedback that accompanies correction, according to John Hattie's research on feedback, is one of the most powerful levers available to anyone who teaches.

There is also a feature that makes this approach particularly well suited to the world of work, where on subjects such as safety or compliance it becomes hard to accept that anyone is left behind. The meta-analyses show that mastery learning helps above all those who start from a less solid base, narrowing the distance between people who had already grasped a concept and people who needed a second pass to get there. In a traditional classroom that second pass rarely finds any room, because the lesson moves on at the pace of the syllabus and those who struggle accumulate gaps along the way. With mastery learning the pace instead adapts to the person, and the final certificate goes back to corresponding to a genuine command of the subject.

The question of scalability

Bloom himself described his own result as a problem, and not out of modesty. The most effective condition, namely one-to-one tutoring, is also the most expensive and the hardest to extend to everyone, since it would require a dedicated teacher for every single student. Group mastery learning was his partial answer, a way of approaching those results with the resources of an ordinary class, and it remained costly to run by hand, because preparing alternative tests, marking them, putting together remedial materials and giving fresh assessments to each person involves a considerable amount of work.

This is the point at which technology has changed things. A digital platform can carry out automatically what would be unsustainable for a single teacher, because it can keep track of every answer, recognise the points where each person ran into difficulty, present those same topics again and check learning once more, all without multiplying the workload. A good deal of the research of the past forty years on learning systems and digital tutors grew precisely out of the attempt to make Bloom's insight scalable at last, and mastery learning is today among the things that software handles with particular effectiveness.

Mastery learning inside Evolve

Evolve, AWorld's Learning Experience Platform, brings this principle into its learning paths, and it does so above all through the mechanism of the final assessment. Along a path the learner meets short readings interspersed with quizzes, which perform the active retrieval function described by the research, and at the end the final assessment takes up again all the questions that were answered incorrectly during the journey, presenting them once more with unlimited attempts until they are passed. It is an almost literal translation of the mastery learning cycle, because in this way no one moves on carrying their own gaps along, and each person returns to exactly the content they had not understood until they have command of it, at a pace that adapts to the individual and with a final certificate that attests to real understanding.

This mechanism fits together coherently with the other elements of the platform. The distribution of content over time, which lies at the heart of microlearning, gives everyone the room to consolidate before moving forward, and gamification provides the motivation needed to complete a path that, by asking for genuine command, is more demanding than a course to be skimmed distractedly. The dashboard, finally, shows whoever manages the training how each person is progressing and, above all, which topics required more attempts, a valuable piece of information for understanding where the content or the processes need to be clarified. <!-- INTERNAL LINK: article on scalable 1:1 tutoring / the two sigma problem -->

Frequently asked questions

What is mastery learning? It is an approach to learning in which you move on to a new topic only after reaching a level of mastery on the previous one. It rests on a cycle of teaching, testing, feedback, error correction and a further test, and it was formalised by Benjamin Bloom in the late 1960s.

Does mastery learning really work? Research confirms it with a fair degree of confidence. Meta-analyses, such as the one by Kulik and colleagues covering more than a hundred studies, show positive and consistent effects on learning, on average around half a standard deviation, with the clearest benefits for those who start from a weaker base.

What changes compared with a course that ends in a quiz? In many courses the final quiz serves to certify that the test was passed, and the mistakes made along the way are never revisited. Mastery learning instead provides for those same mistakes to be corrected and re-checked until they are overcome, so that completing the course corresponds to genuine understanding rather than a barely sufficient score.

How does Evolve apply mastery learning? Through the final assessment of its paths, which presents the questions answered incorrectly along the way with unlimited attempts until they are passed. In this way no one advances with unresolved gaps, and the final certificate attests to a real command of the subject.

Giving learning the time it needs

What sets apart a course that genuinely teaches lies in large part in the way mistakes are handled along the way. Setting them aside in order to reach the certificate sooner is the surest way to find them again, untouched, at the moment they really matter, whereas taking them up patiently, correcting them and checking once more takes more time and more care, and for that very reason it turns information that was once merely heard into a competence that lasts. It is a possibility that technology has made accessible to any organisation that wants its training to leave a mark, well beyond the narrow circle of those who could once afford a teacher devoted to every single person.

If you want to see how to build paths that people complete while genuinely reaching mastery, discover Evolve and talk to our team.

Sources

• Bloom, B. S. (1984). The 2 Sigma Problem: The Search for Methods of Group Instruction as Effective as One-to-One Tutoring. Educational Researcher, 13(6), 4-16.
• Kulik, C. L. C., Kulik, J. A., & Bangert-Drowns, R. L. (1990). Effectiveness of Mastery Learning Programs: A Meta-Analysis. Review of Educational Research, 60(2), 265-299.
• Guskey, T. R., & Pigott, T. D. (1988). Research on Group-Based Mastery Learning Programs: A Meta-Analysis. Journal of Educational Research, 81(4), 197-216.
• Roediger, H. L., & Karpicke, J. D. (2006). Test-Enhanced Learning: Taking Memory Tests Improves Long-Term Retention. Psychological Science, 17(3), 249-255.
• Hattie, J., & Timperley, H. (2007). The Power of Feedback. Review of Educational Research, 77(1), 81-112.