Until recently education systems focused almost entirely on measuring access to and completion of school instead of what students take away from their schooling experience. Since 2000 there has been a pronounced shift to measuring learning outcomes, with more and more countries assessing student learning in national, regional and international assessments.
The desirability and feasibility of developing comparable measures of learning outcomes has thus become a global issue, vigorously debated among policy analysts, donors and researchers. The challenges of using a global metric to improve learning are also the theme of a two day symposium this week, sponsored by the (US based) Comparative and International Education Society (CIES) entitled: The Possibility and Desirability of Global Learning Metrics: Comparative Perspectives on Education Research, Policy and Practice, at which the GEM Report will be present.
This blog series will discuss questions about global monitoring of learning outcomes, how they can be measured and whether and how information from assessments is used to enable learning.
Arguments in favour of global measures of learning
Proponents of cross-country comparable measures of learning outcomes argue they are long overdue. Unlike the health sector, education advocates have lacked solid measures of learning gains from having attended school. The plight of millions of poor and marginalized children, who fail to master basic competencies during primary school grades, while recognized, fails to garner sufficient policy attention and threatens the achievement of the ambitious sustainable development agenda. Global measures of learning would help to address this.
Just as improved international data on out of school children helped maintain pressure on governments to ensure that all children complete a full primary school cycle, global monitoring of learning outcomes would push governments to prioritize learning and ensure that all children acquire core knowledge and master basic skills.
Comparable measures of learning can also promote a culture of transparency, and evidence-oriented policy making. They can contribute to public debates of desired learning outcomes, and improved international partnerships. They can also help countries develop their capacity for analysing results and assessing a wider range of skills.
Arguments raising concerns about developing comparable measures of learning
Many have raised concerns over the nature and value of global measures of learning. For example, shifting attention and resources to globally comparable measures of learning would inevitably reduce the range of learning outcomes that countries monitor. The exclusive focus on basic or minimum proficiencies in reading and numeracy, which are more amenable to measurement, risks marginalizing the value of a wider range of subjects and competences and can weaken national curriculum priorities. Moreover, measuring learning only in terms of literacy and numeracy skills gives a slanted portrait of the breadth of benefits that school brings to children. Significant changes in student attitudes, values and behaviour, which are critical outcomes of education, would likely go unmeasured and unnoticed.
Cross-country measures of learning, whether intended or not, inevitably lead to country rankings, whose value remains problematic. League tables on learning can discourage country participation in assessments that contribute to effective policy reform over time. Also, learning contexts are diverse, which makes it difficult to develop and interpret comparable measures of learning both in language and mathematics.
The costs of comparative learning assessments, which can be a substantial burden for poorer countries, are likely to result in requests for funding support from international aid agencies.
Finally, while large-scale assessments are useful in tracking system-level performance, evidence is limited on how useful they are in guiding teacher training and classroom practices and improving learning outcomes over time.
How can the two sides of the debate be bridged?
The adage ‘don’t value what you measure, measure what you value’ can still unite the international education community under the common cause of improving education quality. Student proficiencies in reading and mathematics, which represent key foundational skills, are highly valued. They can provide signposts of a well-functioning education system. Data on reading and numeracy skills can also be used to examine how learning contributes (or not) to different development outcomes. However, measuring proficiency in these areas requires sensitivity to national and linguistic contexts. Measuring these and other learning outcomes should be an ‘open source’ project developed in a collaborative and transparent manner.
Coordination will be needed to efficiently allocate financial resources for improving the comparability of learning measures across diverse contexts. The recently established Global Alliance to Monitor Learning, coordinated by the UNESCO Institute for Statistics, which aims to harmonize assessment frameworks and standards and help coordinate capacity-building efforts, is a welcome initiative. A further initiative by the Global Partnership for Education, Assessment for Learning, is also being proposed as a platform whose main aim would be to finance these capacity-building priorities and strengthen national learning assessment systems.
We look forward to hearing the views and debates that emerge from the CIES Symposium this week. The issues and challenges around this topic are of special interest to the GEM Report, especially as we collect and analyse evidence related to different accountability mechanisms, including learning assessments for our 2017 Report.