Improving education in developing countries is an essential task, and the best indicators of improvement measure performance in reading, mathematics, and other competency areas. But can these measurements be used as incentives to stimulate more rapid improvement in education? There are no simple answers to this question since learning measures pose a myriad of technical challenges. In this CGD Working Paper, visiting fellow Marlaine Lockheed reviews some of these challenges and the effects they could have on measuring the success of progress-based aid programs.
Lockheed addresses three questions about testing in developing countries. One, are valid and reliable measures of student learning are currently available in developing countries? Two, are existing tests used in developing countries capable of registering the changes in educational results called for under progress-based aid? And three, do developing countries have the technical and administrative capacity to undertake annual assessments of learning?
She concludes that existing national learning assessments in developing countries are poorly suited to measuring annual educational progress for both technical and administrative reasons. And existing international and regional assessments, while generally technically superior to national assessments, take place only periodically and therefore are also unsuitable for measuring annual progress.
Lockheed offers suggestions to successfully incorporate measures of learning outcomes into programs for progress-based aid, including using either tests specific to the country or programmatic tests developed by donors; basing measurements on a countries’ progress implementing the various technical steps required for participation in regional or international assessments; measuring non-test-based indicators of educational quality such as the development of concrete performance standards and increases in the amount of learning time; and incorporating multiple indicators of progress into a “country report card,” prepared by an NGO, which would augment test-based results with other observable indicators of education quality that are not complicated by the technical dimensions of testing.