This blog is the third in a six-part series on the KIX discussion papers commissioned by the GPE Secretariat to inform the design and implementation of the GPE Knowledge and Innovation Exchange (KIX). The post highlights relevant thematic outcomes in the 2019 Results Report and from GPE’s ongoing country-level evaluations.
Without data on whether and how students are learning, it is impossible to ensure quality education. With the drive for education of quality front and center of the Education 2030 agenda, the issue of learning assessment is an increasing priority for developing countries and the partners that support them.
Despite a proliferation of assessments in many countries, including national examinations, sample-based assessments and the everyday classroom assessment practices deployed by teachers, developing countries face challenges in ensuring that assessment data effectively informs policy and practice.
This is why strengthening learning assessment systems is a priority area for GPE’s Knowledge and Innovation Exchange (KIX).
Comments
Please bear in mind a very simple issue: many learners are learning through a second language which they do not speak well enough to use as a medium of instruction. This can radically depress what they are able to learn. Thus assessment systems cannot measure what they would be capable of learning if they were working through a language in which they are fluent. Secondly, assessment in a second language is very unreliable. Learners possess good curricular knowledge which they are unable to express in the second language, the language of assessment. This means that assessment is not only unfair to learners in that it prevents them from demonstrating knowledge which they possess; it is also unreliable as a measure of the achievement of the whole cohort: classroom assessment as well as public examinations fail to show what learners can achieve. This issue of medium of instruction and epecially medium of assessment is a first order issue: whatever technical means are used - as described in this paper - to assess learning, they will be straightforwardly inaccurate if the language of assessment produces skewed results. It is a hugely significant matter. But it is constantly, as in this paper, overlooked. Can we please start taking it seriously.