Lotta Larsson, Senior Advisor in the Department for Population and Welfare Statistics, Statistics Sweden
This blog was also published by Norrag.
As the Inter-agency and Expert Group on SDG Indicators (IAEG-SDGs) meets in Vienna from 9-12 April 2018, a perspective from Sweden illustrates the challenges even the world’s most advanced statistical systems face in producing the education data needed to monitor and achieve the global education goal.
The Scandinavian countries are often held as a model for other countries to follow on almost any area of development you can name, from poverty reduction to health and well-being. From an international perspective, Sweden is a country with a high quality education system.
In 2013, however, the PISA results showed that the average scores had declined from previous heights to below the average for OECD countries. This started discussions on the quality of the education system at the primary and lower secondary levels in Sweden. Since 2013 the country’s PISA results have improved and it is now – once again – at or above the OECD average for mathematics, reading and science.
Among others, the discussions in Sweden have focused on three aspects of the education system:
- The need for stronger measures for children with special needs in the earliest grades of schooling. And that, in turn, means developing the skills of teachers to work with these children;
- Teacher shortages in the future and working to make teaching more attractive, including increasing teacher salaries and improve the working environment for teachers in schools; and
- The question of how to tackle inequality between schools – more resources to schools with poorer results can be one way to reduce inequality.
In 2017, Statistics Sweden published a detailed analysis of the country’s implementation of Agenda 2030, based on all the available data and results. The analysis found that, from an international perspective, Sweden has good access to statistics on education. But this does not mean that we face no challenges in gathering the detailed and disaggregated statistics that are needed.
We are, for example, seeing falling response rates in surveys of individuals in Sweden, such as the labour force survey, the survey on living conditions for adults and children and the adult education survey. This is a real worry, and reinforces the need to provide incentives for participation, as well as the need to try different methods, using mixed modes of data collection, non-response follow ups and dissemination of information about the surveys to the Swedish population.
Right now, disability status is not readily available from registers or administrative sources. We need comparable definitions of disability status across different sample surveys – and plans are underway to ensure this comparability so that children and adults with special needs become more visible in the surveys on the labour force and living conditions.
When we get to other variables, such as the highest levels of completed education among parents of pupils who were born abroad, we have another problem. We have register-based data on some aspects of their lives, but a large amount of information about the parents’ highest completed education in the registers is missing. This is a worrying gap, given the close links between the educational performance of children and the educational level of their parents.
Statistics Sweden is working collaboratively with several national organizations and authorities that are engaged in monitoring progress towards Sustainable Development Goal 4 (SDG 4) on education. These include the Ministry of Education, the National Agency for Education, the Swedish Higher Education Authority and the Swedish Council for Higher Education.
The national statistics on teachers are of good quality but there is still a need for a national strategy regarding indicators at a disaggregated level for teacher salaries, teachers for children with special needs and teacher training. This involves working to align the national definitions to global definitions of teacher indicators.
The focus is firmly on equality in Sweden at the moment. Take SDG Target 4.5, for example: eliminate gender disparities in education and ensure equal access to all levels of education and vocational training for the vulnerable, including persons with disabilities, indigenous peoples and children in vulnerable situations. Statistics Sweden is working with national stakeholders to develop a national list of education indicators based on global and thematic education indicators in Agenda 2030.
This involves disaggregating indicators by native/foreign born, country of birth, disability status, socioeconomic background and income. It is possible to disaggregate the indicators based on administrative sources to a detailed level, which is not always the case when indicators are based on sample surveys. The number of respondents will be too small to produce the statistics at a detailed level.
Statistics Sweden is working over the long term to be able to produce the disaggregated statistics that are needed. For example, there is information on the year of immigrants’ arrival to Sweden in our administrative registers, but the quality and coverage have to be ensured. This is crucial information to ensure equitable education for recent immigrants.
We are also pursuing more data on schools where a large proportion of pupils have parents whose education has been limited, who have low incomes and who were born in another country or have only recently arrived in Sweden – vital information for the allocation of resources.
And on SDG Target 4.c – increasing the supply of qualified teachers – the focus is on more disaggregated data on the number of trained and qualified teachers, teacher salaries and pupil/teacher ratios, zooming in on both the municipal and school level.
There is still much work to do both here in Sweden and in many other countries to gather the data needed for the world’s education goals. But we’re working hard to get this done.