In 2009, Cambodia’s Ministry of Education, Youth, and Sport (MoEYS) developed a Khmer
language Early Grade Reading Assessment (EGRA) with support from the World Bank. This
assessment was designed to provide the MoEYS with an instrument for assessing curriculumbased
early grade literacy outcomes for students in grades 1-3. However, in the following year,
the MoEYS revised their primary-level Khmer textbooks to reflect a more phonics-based
approach to instruction, returning to more traditional methods of teaching and learning.
Therefore, while the EGRA continued to gain prominence as a measure of reading outcomes,
there were concerns about whether or not the assessment accurately reflected the learning
expectations for early grade students.
In 2015, the U.S. Agency for International Development (USAID) asked RTI International to
conduct an assessment of the education sector and to look specifically at the 2009 Khmer
EGRA. As a result of the findings of this assessment, USAID then asked RTI to review and
strengthen the instrument including a pilot to ensure validity and reliability. This task began with
a review of the Grade 1-3 textbooks to better reflect the realities of materials being used, prior to
conducting a workshop in Phnom Penh in August 2016. The primary focus of the workshop was
to review the objectives and purposes of EGRA, critically examine the instrument, make the
necessary revisions, and conduct a field test to address any major issues with changes that
were made. This was then followed by a full pilot of the newly revised/created instrument and
rigorous analyses of the results. The EGRA strengthening workshop and pilot activities were led
by RTI’s Jessica Mejia, Jonathan Stern, and Keely Alexander, with support from Dr. Ok Amry of
SBK Research & Development (a subcontractor to RTI).
The results of the pilot show an overall strong instrument with high internal consistency and
validity. The results also show a good grade progression across the majority of subtasks, as
well as consistency across reading passages. However, some inconsistencies were found,
especially in the listening comprehension and dictation subtasks. While the new instrument is a
significant improvement upon the original version (in terms of EGRA best practices and
appropriately reflecting learning expectations from the current curriculum), there are several
recommendations to be considered before using the EGRA at scale.