Why smaller classes sizes don’t improve learning
By Fenton Whelan and Richelle George
Smaller class sizes mean more teachers, which almost always means less learning.

There is a pervasive myth that smaller class sizes are beneficial to students

According to the OECD, reducing class sizes is one of the most popular, widely implemented and extensively funded reforms for improving learning in schools.

In the summer of 1996, the state of California implemented what was, at the time, the most expensive education reform in history by introducing incentives to encourage schools to reduce class sizes in kindergarten and the first three grades of school to a maximum of 20 students (down from 29) in each class. Within three years, the reduction in class size had largely been implemented across the school system.


California has not been alone in reducing class sizes. In fact, almost every school system in the developed world has reduced its student-to-teacher ratio over the past two decades.

The truth is, smaller class sizes do little to improve student outcomes  

The problem was that the initiative did not have much impact on how much Californian students learned at school. The state’s own evaluation of the initiative concluded that while there had been some improvements in educational performance in California over the period as a whole, the “attribution of gains in scores to class-size reduction is not warranted”.

In fact, reducing class sizes often results in lower student outcomes. When class-size reduction is implemented across a school system as a whole, it significantly reduces teacher quality. School systems which have smaller classes need to hire more teachers. The more teachers a school system has, the less it can pay each one. Countries with smaller class sizes require more teachers, so they are also less able to be selective about the teachers they hire.

Among developed countries, varying class sizes explains more of the disparities in student performance between school systems than any other policy choice. While countries with the smallest class sizes are often among the lowest performing, those with the largest class sizes are among the best.

What does this mean for improving education globally?

Education outcomes are not improved simply by shrinking classes and hiring more teachers. The evidence consistently shows that what really matters is not whether there are 25 or 15 students in a classroom, but rather whether that classroom has a good teacher. Studies show that within the range of class sizes typical in developed countries, any effects which variations of class sizes may have are completely dominated by the effects of teacher quality.


The difference between how much a child learns each year when placed with an effective teacher compared to an ineffective one is equivalent to up to a year of schooling. Any positive impact induced by class-size reduction has been shown to be much less. For example, the Student Teacher Achievement Ratio project (the biggest experimental study conducted into the effects of class-size reduction) concluded that, when all other variables are constant, students in the study’s smallest classes achieved an increase in performance over their peers which would be equivalent to only ten additional days of schooling per year.

About the Authors

Richelle George is a research expert with expertise in public health, gender and development. Fenton Whelan founded Acasus, he has more than a decade of experience in public health and education development.

Many reform efforts continue the status quo or apply radical solutions to common problems. In our experience the best approach is far..
Learn more
The science of detecting false data
By Fenton Whelan
19 November 2019
Leaders of health and education systems often rely on datasets that are inaccurate or incomplete.
Learn more
Case for intervention in private schools
By Fenton Whelan, Jonny Barty & Max Balestra
05 August 2019
Large numbers of children in developing countries attend private rather than government schools.
Learn more