School size: A literature review

by Deborah Nusche
Analyst, Directorate for Education and Skills

How big should schools be?

Is bigger better? Or do all the best things come in small packages? For education systems, the question of how school size influences quality and efficiency has long been an important issue. It has become especially pertinent in recent decades, as fiscal pressures and a falling school population in rural areas have meant that countries are looking for the best way for their schools to be effective.

A new OECD working paper on school size policies, published today, shows that there are many ways in which school size may influence learning environments. Small schools make personal contact easy and are often strongly defended by local communities, while larger schools can provide more options to meet a diverse range of interests and needs. Sometimes, due to demographic or geographical challenges, there may be little choice regarding decisions on school size. But disadvantages can be offset, whether through connecting small schools using ICT, or through providing incentives to make larger schools more attractive for teachers and students.

Australia, for example, has used technology to help small schools in rural areas connect with each other through blogging, emailing and engaging in collaborative projects. This has reportedly led to students being more excited about their learning and more motivated to work. In Korea, where over a third of schools have fewer than 60 students, there has been a focus on consolidating schools in rural areas. The process involved many challenges, but due to targeted investments, many of these consolidated schools have thrived and are now attracting students even from urban areas.

In a time of fiscal constraints, consolidating schools is an appealing policy option for governments under pressure to reduce spending. Large schools are often seen as more efficient than small schools: they can use facilities to full capacity, buy large amounts of materials at lower cost and hire support staff to reduce the administrative burden on teachers. But such calculations often forget about hidden costs of consolidation such as increased transportation cost and time for staff and students. There is also evidence that the student body in large schools is often divided with some students taking full advantage of the broader learning options and others not participating at all. Younger students and those from less advantaged families are more vulnerable and may disengage when lacking personalised attention.

While, certainly, there is no “one size fits all approach”, there is much that can be learned from existing research and experience in different countries. This new literature review draws attention to the many factors to be considered when making decisions about school size. What are the transport and cost implications? How can interested parties become informed and engaged in the process? Will the effects of school size changes be different for differing school populations? What type of additional care would be necessary to support those affected by the change? Including such elements in the planning process will help make the full costs and benefits of school size changes transparent and support a constructive debate among stakeholders.

This paper is the first in a series of literature reviews looking at the use of resources in OECD school systems to be published in the coming months.

Links:
School Resources Review - Background papers and studies
Photo credit: Art Supply, An Illustration Different Size of Sharpened Detailed Pencils for Writing or Sketch and Draw A Picture /@Shutterstock

How much do teachers cost?

by Eric Charbonnier and Etienne Albiser
Analysts, Directorate for Education and Skills

Can increasing the salaries of teachers lead to better learning outcomes? Does reducing class size have a positive effect on learning outcomes? Given the current background of tight public budgets, governments seeking to ensure value for money must ask themselves these questions before increasing the salary cost of teacher per student, as teachers account for a major part of education expenditure.

The latest edition of Education Indicators in Focus highlights that the salary cost of teacher per student is a combination of four factors: teachers’ salary, class size, the number of teaching hours in front of a classroom and the number of hours of instruction received by students.

Countries that have similar levels of expenditure on education do not necessarily have similar educational policies and practices. A given level of expenditure may result from a different combination of these factors. One country may pay higher salaries to teachers while another may have smaller class sizes and thus more teachers to pay.

Between 2000 and 2010, increases in the salary cost of teacher per student were mainly influenced by changes in teachers’ salaries and class size
 With the exception of France and Italy, the salary cost of teacher per student at the primary and lower secondary levels increased between 2000 and 2010, and on average it increased by one-third and one-quarter, respectively (for countries for which data is available). In the majority of cases, this increase was due to an increase in the level of teacher compensation (16% at the primary level and 14% at the lower secondary level). The largest salary increases (more than 50%, in constant prices) were seen in the Czech Republic, Estonia and Turkey.

During the same period, class size decreased, by 14% (primary) and 7% (lower secondary), but this was often the result of changing demographics and not of a change in education policies.

Little change in instruction time and teaching time
With the exception of a few countries, there was little or no change with respect to the two other variables (instruction time and teaching time) during the same period. This may be due to the political sensitivity of reforms in these areas. At the primary level, teaching time increased most significantly in the Czech Republic (200 hours) and instruction time increased the most in Iceland (by nearly 200 hours).

The higher the level of education, the higher the salary cost of teacher per student, with great disparities between countries
Spending on education rises sharply with the level of education. The OECD average salary cost of teachers is USD 2 307 per primary student, USD 2 856 per lower secondary student and USD 3 301 per upper secondary student. In some countries, the differences between the different levels of education is quite small (in Chile and Hungary it is less than USD 50) while in others it is quite important (exceeding USD 2 000 in the Flemish Community in Belgium).

In general, teachers of higher levels of education earn more money than teachers at lower levels. In addition, teaching time generally decrease as the level of education increases (meaning that more teachers are necessary to teach the same number of students).

Wrapping up
Reforms relating to these four factors have an impact on education expenditure and may also affect learning outcomes. However, the link between expenditure and outcomes is not straightforward. PISA results show that between 2000-2009, the performance of 15-years olds did not vary significantly in the majority of countries, regardless of the changes we have seen in instruction time, teaching time, class size and teacher compensation. What is more, changes relating to pedagogy may have an impact on outcomes without necessarily having an impact on expenditure.

The bottom line is that in the past ten years, increasing teachers’ salaries and reducing class size have not led to better learning outcomes in the majority of countries. This raises the question: has all of the additional money been well spent?

For more information
On this topic, visit:
Education Indicators in Focus: www.oecd.org/education/indicators
On the OECD’s education indicators, visit:
Education at a Glance 2012: OECD Indicators: www.oecd.org/edu/eag2012
Chart source: OECD Education at a Glance 2012:  Indicator B7 (www.oecd.org/edu/eag2012)

How does class size vary around the world?

Elisabeth Villoutreix
Communications Officer, Directorate for Education

Class size is a hotly-debated topic and continues to be at the forefront of the educational and political agenda in many countries. Smaller classes are favored by parents and teachers alike. But they come at a price, countries can spend their money only once and money spent on smaller classes can’t be spent on better teacher salaries, more instruction time, better opportunities for the professional development of teachers...

So what's the magic formula? What is the  ideal class size? Is smaller necessarily better?

The latest issue of Education Indicators in Focus shows that at the lower secondary level  among all OECD countries with comparable data, the average class size varies from 20 students or fewer in Denmark, Estonia, Finland, Iceland, Luxembourg, Slovenia, Switzerland (public institutions) and the United Kingdom, to more than 34 students in Korea. The contrast is even more striking in other G20 countries; in China, for example, the number of students per class reaches the 50 students mark.

Class size, together with students’ instruction time, teachers’ teaching time and teachers’ salaries, is one of the key variables that policy makers use to control spending on education. Between 2000 and 2009, many countries invested additional resources to decrease class size; however, student performance has improved in only a few of them.

Apart from optimising public resources, reducing class size to increase student achievement is an approach that has been tried, debated, and analysed for several decades. Some countries like Finland favour smaller class sizes (20 students of fewer) and are among the most successful countries in the PISA study. However, other countries like Korea have much bigger classes (34 students and over) but also feature at the top of the PISA ranking. What other variables than class size may explain the success of countries like Korea?

Findings from the OECD Programme for International Student Assessment (PISA) suggest that systems prioritising higher teacher salaries over smaller classes tend to perform better, which confirms research showing that raising teacher quality is a more effective measure to improve student outcomes.

For more information
On this topic, visit:
Education Indicators in Focus: www.oecd.org/education/indicators 
On the OECD’s education indicators, visit:
Education at a Glance 2012: OECD Indicators: www.oecd.org/edu/eag2012 
On the OECD’s Indicators of Education Systems (INES) programme, visit:
INES Programme overview brochure
Chart source: Education at a Glance 2012: OECD Indicators, Indicator D2 (www.oecd/edu/eag2012)