Learning time: The gap between intentions and reality

It’s common knowledge that students need time to learn. Schools give students time in the classroom to engage with the subject matter; teachers assign homework to encourage students to spend more time analyzing, studying, and solving problems.

Indeed, for over half a century we have had scientific evidence to back up what seems like common sense – namely that different students need different amounts of time. Some students grasp the meaning of a certain math problem more quickly than others; some students memorize their foreign language vocabulary in no time, while others need to study words over and over again.

Why, then, do we still lack the necessary research findings to guide policymakers, principals, teachers, parents, and students in determining how much time is needed to learn, for example, mathematics? While strict (quasi-)experimental studies varying only learning time show positive correlations between learning time and learning outcomes, larger, international studies find only small and inconsistent relationships. Consequently, findings can be summarized as “small at best” and “typically inconsistent.”

A literature review reveals that the simple idea of “learning time” is used in vastly different contexts across studies and that, in fact, it is quite difficult to formulate a simple, straightforward definition of the term. Education systems typically define the duration of mandatory learning time, e.g. the number of mandatory years of schooling.  The literature also talks about intensity, e.g., the number of lessons per week or per school year.

To complicate things even further, students learn not only in school, but also by doing homework, studying on their own, attending cram schools, working with a tutor, or simply playing educational games with their peers.

Allocated learning time does not correspond to actual learning time

To organize this multitude of possible uses of the term “learning time,” it is important to keep two things in mind. First, learning time can be formal (organized in school or private lessons) or informal (private study and learning apps), and second, it involves a variety of agents and stakeholders at different levels – school systems, schools, classrooms, and individual students.

The curriculum or administrative guidelines specify a certain amount of formal learning time, for example four math lessons per week for 10^th grade students over the course of a school year that lasts for 183 school days. Ideally, sufficient public funding is then allocated to allow schools to be in session for those 183 days and to employ enough math teachers to provide four math lessons per week for all 10^th graders.

“The more time a student has to engage in meaningful learning activities with relevant curricular content, the higher the chances that substantial learning gains will be achieved.”

Unfortunately, in real life there are numerous reasons why – contrary to the system’s best intentions – schools may be unable to provide the prescribed amount of instruction. Some factors – weather conditions, for example – may affect all schools in a given region, while others, such as shortages of various kinds, may only affect individual schools.

Still other factors, such as organizational tasks or disciplinary issues, lead to differences in the amount of instruction provided in one classroom relative to another. Finally, students themselves are a factor as well: Not all of them are equally engaged in school and individual learning.

So there is a gap between the amount of learning time that is originally allocated and the time students actually spend learning. This gap varies across countries, regions, schools, classrooms, and individual students. It’s no wonder, therefore, that the learning time allocated at the systemic level is hardly correlated with students’ learning outcomes.

Comparing different measures of learning time

To test whether a measure that takes into account different aspects of time loss is more closely related to test performance than one that does not, my team and I looked at learning time with the help of data from the 2012 wave of the OECD’s Program for International Student Assessment (PISA). Students and principals answered several questions about possible reasons for lost learning time.

We first correlated students’ test performance with a measure of the number of math lessons they typically have per week and the duration of those lessons (allocated learning time). We found the expected low correlations. But as we corrected this measure of learning time for time lost for various reasons, the remaining measures of learning time became more and more strongly related to student test performance. This pattern can be observed not only for individual students, but also for schools and countries. The figure below illustrates this.

In other words, student test performance is positively related to learning time, but this is best predicted using a measure that has been adjusted for various types of time loss. Moreover, average student performance in a given school (or even country) is positively correlated with the average amount of learning time reported by students in that school (or country). Again, however, this is more easily seen if we first adjust our measure of learning time for possible time losses at different levels of the education system.

Preventing the loss of learning time could result in substantial learning gains

The study leaves us with two messages. First, it is important for practitioners as well as school administrators and policymakers to be aware of possible reasons for a loss of learning time and to limit such losses to an absolute minimum. The more time a student has to engage in meaningful learning activities with relevant curricular content, the higher the chances that substantial learning gains will be achieved. Time loss can occur at different levels and for multiple reasons.

Second, this study is a reminder that as researchers, we need to be precise in our definitions, compare relevant circumstances, and revisit our measures to make sure that we deliver sound and meaningful results that can be translated into practical advice and recommendations.