Executive Summary Executive SummaryExecutive SummaryExecutive Summary Executive SummaryExecutive SummaryExecutive Summary Executive SummaryExecutive SummaryExecutive SummaryExecutive SummaryExecutive Summary
Under the EdData II task order, Strengthening Information for Education Policy, Planning, and Management in the Philippines project (PhilEd Data I), RTI assisted the Department of Education of the Philippines (DepEd) with implementing the first nationally-representative, Early Grade Reading Assessment (EGRA) in Filipino and English in grade 3, and a representative regional sample of first graders in Ilokano (in Region 1). After presentation of the results of the PhilED Data national and regional surveys in mid-2013, USAID agreed to support DepED’s mother tongue-based multilingual education (MTB-MLE) strategy through additional regional baseline EGRA surveys in three languages— Hiligaynon, Maguindanaoan, and Sinugbuanong Binisaya (Cebuano)—along with a follow-on study of Ilokano.
The 2014 Philippines cross-language EGRA study (PhilEdData II) was conducted in four regions to measure reading proficiency in a representative sample of 400 grade 1 (G1) and 400 grade 2 (G1) children in schools where MTB-MLE was being implemented in one of the four following languages: Ilokano, Hiligaynon, Sinugbuanon Binisaya and Maguindanaoan. The study also used information from teacher interviews and classroom observations to construct an index measuring the degree of implementation of three key areas of MTB-MLE reform: (1) Teacher preparation and ability to provide instruction in the mother tongue; (2) Availability and use of a diversity of MTB-MLE materials; and (3) Instructional practice and the use of time during mother-tongue reading lessons.
In November 2013, RTI facilitated an instrument adaptation workshop during which DepED officials, local linguists and curriculum experts worked to develop EGRA instruments for each of the four target languages: Ilokano, Hiligaynon, Cebuano, and Maguindanao; teacher survey and classroom observation instruments were also developed in conjunction with DepED colleagues. In December-January all instruments were finalized, piloted, and rendered into electronic format using the Tangerine® open source software; assessors were trained and data collection was carried out in February 2014 by locally recruited researchers, teachers and other DepED officials.
According to classroom observation data from the study, teachers and students are using MT as the medium of instruction in G1 and G2, both during MT reading lessons and during other subjects. However, not all the necessary supports are in place for teachers to provide the most effective instruction possible in MT. While the majority of schools have MT language teaching materials, in 40% of classrooms not all students have the materials and half of teachers do not have teachers’ guides. More importantly, teachers are not regularly employing techniques that productively reinforce reading in MT. The results on the EGRA in the four languages also suggest some gaps in productive teaching methods that may be associated with achievement. For example, letter-sound identification shows a strong correlation to other word reading tasks, but 10-25% of children, depending on language, could not identify a single letter sound correctly. This is the case for more than half of children learning to read Maguindanaoan.
Some important aspects of MTB-MLE implementation showed strong relationships with student performance in reading. For example, students who often used their mother-tongue book for reading were nearly three times more likely to be in the top 25% of reading performance, than those who never used it. Similarly, students whose teachers often asked them to read in mother tongue were more than three times as likely to be in the top 25%. These findings suggest that if MTB-MLE implementation conditions were better assured across a broader range of schools, reading performance would likely also show broader improvements.
Similar to the findings in the 2013 study, girls are performing better than boys in most languages and subtasks (with the exception of listening comprehension where boys seem to do better). Regional differences are also apparent, which is most likely explained by how well-resourced the schools are and other socio-economic or demographic factors rather than language or MTB-MLE implementation-related factors.