Improving Early Grade Reading Outcomes: Aprender a Ler in Mozambique

eTale 2022

The Government of Mozambique has long struggled to improve the low reading levels of children in early grades. A research-based reading intervention was developed and tested in two provinces. This article examines student reading performance from cohorts of second- and third-grade students before and after a 1-year intervention when compared to a control group. The study identifies factors required for successful scale-up of the intervention.

Authors: Shirley Burchfield, Haiyan Hua, David Noyes, & Willem van de Waal

Source: Burchfield, S., Hua, H., Noyes, D., & Van de Waal, W. (2017). Improving early grade reading outcomes: Aprender a ler in Mozambique. In A. Gove, A. Mora, & P. McCardle (Eds.), Progress toward a literate world: Early reading interventions in low-income countries, New Directions for Child and Adolescent Development. 155, 117–130.

In the 1990s (after the civil war), school attendance in Mozambique was difficult due to high poverty levels, impassable roads, and deteriorating or non-existent schools, despite a commitment to improve education.
Enrolment was low (around 40%) and schools lacked books, supplies, and qualified teachers.
Between 2003 and 2010, primary school enrolment increased from 3.3 million to 5.3 million.
With an insufficient number of trained teachers and too few schools to meet the increased demand, most schools operated with double shifts.
One study reported that only 41% of children at Grade 3 could read a single word, and only 6% could read 11 words or more per minute.

Aprender a Ler

The objective of this intervention is to improve reading outcomes for children at Grades 2 and 3 in Zambésia and Nampula provinces by enhancing quality, reducing absenteeism and tardiness, and increasing instructional time.
In 2015, Grade 1 was included in the programme (except for the first 12 weeks of school) when the emphasis is on teaching vocabulary in Portuguese.
The programme provides training, coaching, and technical assistance to strengthen basic education services, and was originally targeted at 120 schools.
The medium treatment group received intensive professional development, culminating in an examination and a certificate.
The full treatment group received school management training aimed at reducing absenteeism and tardiness in addition to training and support.

The study

Research questions:

To what extent have programme interventions improved reading outcomes for children in Nampula and Zambésia provinces?
What factors might assist or impede scale-up efforts?

Sampling included a priori random assignment of 180 schools to either a treatment or a control group in Zambésia and Nampula. Evaluators selected 10 children from each class (3,475 in total). The same schools were sampled at baseline and for midline 1 and midline 2 evaluation, although different students were assessed across the applications.

Components of the intervention

Reading reinforcement programme: provides training, coaching, and teaching materials in Portuguese to teachers in Grades 2 and 3 using a phonics-based scope and sequence.
School director training and coaching programme: provides school directors with training in school leadership, effective communication, giving and receiving feedback, using practical school management tools for data-based decision making, and building community participation.
Institutional capacity building: project staff work with the Provincial Directorate of Education and Culture and Teacher Training Institutes to prepare them to eventually take over programme initiatives.
Measuring impact for evidence-based decision making: the project integrates a rigorous monitoring and evaluation process throughout the programme.

Findings

After the first year of project implementation, students in both the medium and full treatment groups performed significantly better on all early grade reading assessment subtasks (oral comprehension, concepts about print, letter recognition, familiar word reading, oral reading fluency, and reading comprehension) than students in the control group.
Students in full treatment schools scored significantly higher than students in medium treatment schools for most measures.
The treatment group showed the greatest improvement in letter recognition, familiar word reading, and reading fluency.

Implications

Four factors were most likely to have contributed to the results: availability of materials, student and teacher attendance, training and supervision, and language of instruction.
Each teacher in the Aprender a Ler treatment schools received two copies of each of 18 titles of read-aloud books and one copy per child of each of 18 decodable books. In control schools, only 10% of children had any reading books.
One impediment to programme scale-up is high student and teacher absenteeism: 52% in full treatment schools and 62% in control schools for students, and 31% in treatment schools and 34% in control schools for teachers.
The programme selects the strongest and most committed reading coaches and school directors to join a core group of trainers. The success of the scale-up depends on the strength of each district education office.
Inability to understand the language of instruction is an impediment to scaling up the programme. The project has developed teaching and learning materials in three Bantu languages. The purpose is to teach vocabulary and comprehension in Portuguese by using a cross-linguistic transfer model. This new component will likely significantly improve children’s ability to learn to read and enhance their overall academic performance.