Almost as good as individual tutoring

Singapore is a nation where coaching and tuition are part and parcel of a child’s education. There are more than 500 tuition centres and Singaporeans reportedly spend more than S$800 million a year on tuition. But does this work? Are there alternatives?

Well-know education psychologist Benjamin Bloom, in a now-classic report with his doctoral students, showed that when students are individually coached and tutored, they perform much better (by 98 per cent, on average) than those who received regular classroom instruction.

This is a significant and striking observation that gets to the heart of learning. The implication is that it is possible to greatly extend and improve learning beyond traditional classroom lectures, and that under the right circumstances, most students have the potential to perform at a very high level.

One important point to note is that much of current education focuses on how students do under a system of teaching (usually lectures), by grading and ranking their performance on tests. Students who receive coaching can be expected to do better than their peers.

But in the context of medicine — and I would dare say for most other professions — the goal of learning is competence, not one’s relative ranking. In a medical school like ours, the focus is not just on grading a student but on getting them competent to practice medicine; it matters little that they perform better than their peers if they do not achieve competence.

MENTORING WITHIN A TEAM

Medical students learn best when they are part of the clinical environment and are, therefore, individually mentored and tutored. To do that, we place students in clinical care teams — that is, teams of doctors and nurses who take care of patients. The students get to watch and learn both the science and art of medicine.

They begin to pick up the culture as well as the knowledge, but also critically, the application of knowledge in diagnosing and managing patients. They also learn bedside manners, how to elicit the history and nature of patient complaints and problems, technical skills in performing patient examinations; as well as critical thinking skills in ordering and interpreting tests and then using this information to diagnose, treat and manage the patient.

They also learn how to communicate and educate patients and families. They learn how to work with their team in a manner which optimally serves the patient’s interests. The learning happens not only from the main physician or consultant on the team but also, more usually, from fellow members, senior students, house officers, resident physicians, registrars and nurses.

This kind of learning is more than individual tutoring; it includes a fair amount of mentoring by being part of a team managing patients.

Clearly, this form of environment for learning, with major individualisation, works for medical school learning. But can it be applied to small class sizes and in other settings? Could we scale this type of learning? If so, how?

MASTERY LEARNING

Bloom proposed a partial solution to improve performance by using simple questioning and testing. He suggested that feedback tests by themselves could improve performance. He called this type of instruction “mastery learning”.

He and his group conducted a study comparing three groups receiving regular classroom instruction, mastery learning and tutoring. The regular classroom group and the mastery learning group comprised about 30 students each. The tutoring was for up to three students at a time. The results were impressive.

As expected, the students with individual tutoring performed best, but the mastery learning class was also much better than the group that received regular classroom instruction: Remarkably, 70 per cent of students in the mastery class performed at the same level as the top 20 percent of the regular class. In this case, testing and examinations are used as a tool for learning, not as a tool for grading.

Physicist and Nobel laureate Carl Wieman, who has a passion for research based educational methods, in a recent article in Science magazine compared a very simple method called “deliberate practice” to conventional teaching.

“Deliberate practice” begins with students getting multiple-choice questions on a particular concept, which they then discuss in small groups and answer. Their answers are used by the instructor in a short class discussion to correct and improve their understanding.

Similar in intent to “mastery learning”, this was implemented by his post-doctoral student and a graduate student over one week (three times, for an hour each time) for a section of an introductory physics course. The professor continued to teach the other section using the regular setup.

The results were dramatic: After this very brief and relatively small change, the students who took part in the deliberate practice session performed twice as well as those who attended the conventional course, in a multiple-choice test of the material.

This is another powerful piece of evidence that simple feedback methods which assess and reinforce concepts greatly improve learning. Many variations of this can be adapted for different settings.

Such deceptively simple studies hold great promise, showing that a simple feedback system of teaching is better than traditional classroom lectures — and almost as good as individual coaching or tuition.

ABOUT THE AUTHOR:

K Ranga Krishnan is Dean of the Duke-NUS Graduate Medical School Singapore. A clinician-scientist and psychiatrist, he chaired the Department of Psychiatry and Behavioural Sciences at Duke University Medical Centre from 1998 to 2009.