It started with music. It started with the watching, observing, analysing and assessing of the instruction of music. It started with a plan to find out why someone might excel in music. Or rather, it started with an intense desire to sneak a look at the secrets of expertise.

Almost 30 years later (I started this journey in Berlin in 1987), and deliberate practice, the concept that resulted from those early studies, which I and others have built upon, is well established.

It is curious to find myself cited, quoted and sometimes misrepresented. It happens a lot, it seems, in education. To have my ideas explored, discussed and used by teachers is a great compliment, but I have little control of how my work is described. The concept of deliberate practice has become quite popular, but in that process of popularisation, the meaning of the concept has often lost many of the crucial attributes that made it distinctly different from other types of practice activity.

Tuning up training goals

It is time for a correction, I think. So I will try to explain this concept again. I will try to hopefully help teachers see deliberate practice afresh, first hand, with all of its essential attributes.

So, music. The concept of deliberate practice was developed around the instruction of music. Here, the teacher will have a student perform for them. The teacher will then identify where particular improvements can be made and devise specific training goals that are designed to address those issues.

Perhaps the phrasing of a specific section was sloppy; the teacher would design tasks to improve the student’s phrasing work.

Armed with these tasks, the student goes away to practise. They follow the teacher’s direction for training, and each time they perform they can hear remaining differences and are motivated to continue to work on the training goals. They reflect, problem solve, perform and repeat. Eventually, they reach a point where they feel the goals are met and they return to the teacher to perform. The teacher listens, analyses once more and a new set of training activities are devised. The cycle starts again.

This is deliberate practice. My work - and indeed the work of others who have built upon this concept - has shown how it can be applied to virtually any learning task and facilitate a journey towards expertise. Yet so few learning environments are set up to work in this way.

Schools, for example. This type of one-on-one training, where teachers design personalised training that the student can engage in with immediate feedback, is rare. Most practice activities are based on a single teacher organising activities for a group of students. There is no individualised training schedule, no cycle of reflection, problem-solving, performance. Often, a teacher will give a student individualised feedback on a particular generated answer, but the learning activities have to be generic, relying on the approach that one size fits all.

That’s not to say that schools have not tried. Many a school and many a teacher have attempted to make instruction more deliberate (conscious and intentional), and many a researcher has attempted to assess the impact of these changes. Unfortunately, the interpretation of the concept of deliberate practice has often been inaccurate.

We know this as Macnamara, Hambrick and Oswald (see references, 1) have tried to evaluate the effects of “deliberate practice” on performance extensively, and some of the studies they looked at involved schools.

They concluded that the accumulated amount of practice was significantly related to attained performance, but that substantial individual differences were left unexplained.

A closer analysis of their 2014 meta-analysis showed that their estimated correlations underestimated the relation between attained performance and our definition of deliberate practice because these authors had included mostly studies that met only some of our criteria for deliberate practice (see references, 2). The included studies relied on practice activities that were neither targeted at improvement of specific factors of performance nor guided by a teacher in an individualised way.

Another big problem with some of the included studies was that practice was not always linked directly with final performance.

So let me try to describe how deliberate practice might work in a classroom.

One of the main criteria is for the classroom to be set up in such a way that students can repeatedly perform tasks and get immediate feedback in an individualised way. Hence, if a student is just listening to a teacher with everyone else, lecture style, or listening to other students solve a problem, this does not meet the criteria for deliberate practice.

To meet the criteria for deliberate practice, a teacher (or a computer-based system) should analyse each student’s performance individually, so that each of them can be assigned appropriate training goals selected based on their attained skill. These goals should be explained individually and a series of training tasks would be selected. The target performance for a given instructional unit may be similar for all students - a piece of analytical writing, for example - but the training to get there and the individual practice along the way are likely to be different for each student.

Solutions in progress

Admittedly, the school system is not currently set up to do this. You often have one teacher, 30 kids and limited time.

There is some promise in computer-based technology as a solution here, but it is still a work in progress. To try to find a solution, I worked with Deans for Impact on a document describing how teachers can practise with a purpose and incorporate many of the aspects of deliberate practice (see references, 3).

As for specific pedagogy in the classroom best suited to this model, we don’t know what works. That’s an important point: deliberate practice may work better through direct individualised instruction or it may work better through a carefully crafted individual project-based task, or it may be best to combine these approaches or do neither. We don’t yet know.

It is also important to realise that engaging in deliberate practice requires full concentration, which cannot be sustained for a limited time per day, especially for students in lower grades. The ideal school day must therefore be a combination of different types of learning activity, where deliberate practice can be only a fraction of a six-hour school day.

One approach for figuring out successful schedules for activities during a school day involves identifying teachers who are exceptionally able to improve their students’ performance during a school year. In a project led by Angela Duckworth, we are trying to identify these teachers and uncover the methods and organisation of learning activities designed by these teachers, and then help other teachers to adopt them. I am excited to see what we will find.

In the meantime, teachers should experiment. They are the masters of their classroom and they know their students and their context best. And deliberate practice provides an effective tool for helping students improve their performance.

This type of practice shows how important the teacher is for assessing individual students’ skills and helping them find the most effective learning activities to keep improving. It reaffirms how important the teacher is to effective learning. It is not possible to conceive of a great educational system without great teachers. That’s why teachers will remain so fundamental in any future educational system.

Anders Ericsson is professor of psychology at Florida State University