Can we afford not to measure learning?
A new paper from the UNESCO Institute for Statistics (UIS) shows that the costs associated with learning assessments are marginal compared to the benefits arising for students, education systems and society at large.
Learning is at the top of the Education 2030 agenda and the focus of five of the seven targets of Sustainable Development Goal 4. A growing number of countries are assessing learning through national, regional and citizen-led initiatives. However, the results are not internationally comparable and therefore cannot be used to monitor progress globally.
To bridge this gap, the UIS is building a Global Alliance for Learning to ensure that quality data are used to track progress and formulate policies to improve the learning outcomes of all.
Instead of developing a new assessment, the Alliance will link existing assessments through a common measurement framework. Getting more countries to regularly conduct assessments is one step towards this goal, but how much would it cost?
According to the new UIS paper, testing would cost pennies in relation to the total amounts governments are spending per student while providing critical data for improving student achievement.
To regularly assess learning in the early grades of primary school and at the end of secondary, it costs about US$110 per student in a sample size of about 3,000 (for TIMMS, PISA and EGRA). The UIS paper considers these costs in relation to the total amounts governments are actually spending on each student who completes secondary education. For example in sub-Saharan Africa, the estimated cost per secondary graduate would be a mere US$ 0.000139. Plus, boosting learning outcomes has a well-documented impact on GDP.
Clearly, there are many factors that interfere with a student’s learning and successful completion of school, which are beyond the scope of the education system. But at a minimum every government can benefit from accurate data to answer some key questions: Who is learning? What are they learning? When are they learning it?
Viewed this way, it becomes clear that the cost of not assessing learning far outstrips the relatively small expense of increasing participation in country-level testing.
Cost assessments per year per student (ISCED 2-3) in terms of actual government expenditure per graduate (ISCED 1, 2, 3)