Reading Programs that Work
Elements of Success
Education—and its bedrock, literacy—is foundational to human development and critical to a country’s economic growth. A 10 percent increase in the share of a country’s students attaining basic literacy translates into a 0.3 percent higher annual growth rate for that country (Hanushek and Woessmann, 2009). Conversely, children who do not attain reading skills at the primary level are on a lifetime trajectory of limited educational progress and therefore limited economic and developmental growth (Annie E. Casey Foundation, 2010). Yet, according to a 2014 UNESCO report, one in four young people (or 175 million youth) in developing countries are unable to read a single sentence.
Designing high-quality primary grade reading programs is essential to ensuring a pipeline of learners who remain in school, achieve educational success and contribute to economic growth. Because reading is a skill that must be carefully and explicitly taught, designing such programs is complex. Contextual analysis plays an important role in determining the capacity of a country’s education system to deliver quality reading services, where the gaps are, and how each gap should be addressed in a given context to improve reading outcomes in that context.
Contextual analysis should focus on the following areas, with activities addressing any gaps in these outcomes:
Enabling Policies and Standards
First and foremost, evidence-based policies on the language used for reading instruction must be in place. The use of familiar languages in reading instruction is crucial, because one’s ability to read builds on one’s ability to speak. Experts estimate that to learn a complex language such as English well, a beginning reader needs a spoken vocabulary of at least 3,000 words. For a student learning to read in a language in which he/she is not familiar, building such a broad vocabulary base takes years. Therefore, the most cost-effective and efficient policy for beginning reading instruction is to deliver it in a language that the learners can already use fluently and understand.
Good data on language use by children should drive policies about the languages in which young children are first taught to read. (Note that these policies may be specific to the earliest grades; they do not have to be policies that determine the language of instruction for all years of study). Several promising methodologies have been developed in recent years to assist in determining what language young learners easily communicate in at home and with each other when they begin school. Recent language mapping funded by USAID has increased the efficacy of early reading instruction by ensuring a better language matching between teachers and students.
Additionally, a government’s curriculum needs to provide enough time devoted to reading in particular, and very clear guidance on reading instruction. Often, reading is written into the curriculum as one of four skills in the “language arts” (reading, writing, speaking, and listening). Such language arts classes are often as short as 30 to 40 minutes, which leads to very little time for children to practice reading during the day. Experts estimate that least 60 minutes per day of reading instruction are required to acquire literacy in a familiar language (Shanahan, 2013; Reading Rockets; Fielding, Kerr, and Rosier, 2007; Gumm and Turner).
Finally, teacher assignment policies need to be developed with students’ need for high-quality reading instruction in mind. Teacher placement decisions should be based on reliable data about both teachers’ and students’ language skills, so that young learners and their teachers all have at least one language they can use to mutually communicate. Additionally, teachers should only be assigned to the early primary grades if they have received explicit training in the teaching of reading. Likewise, pre-service faculty that have not themselves taught reading in primary school should not be assigned to train prospective teachers how to teach reading. Policies such as these will make teachers’ jobs easier and students’ acquisition of reading skills more efficient, creating a virtuous cycle.
High-Quality Texts and Materials
Learners who cannot read need books designed for reading instruction in order to learn. Not just any book will do. In low-income countries, and in many languages, there is a near-absence of the right kinds of books for reading instruction. In many contexts, books given to beginning learners are too difficult, too few and not written in languages the children understand.
At a minimum, books designed for students learning to read should be written in a language that is familiar to them and be linked to a reading curriculum. For beginners, these materials should be decodable (the text includes only symbols, the sounds for which the learners have already studied and are able to practice reading out loud), and leveled (the texts introduce new vocabulary, syntax, semantics, or grammar in a structured way). Making the texts easy, accessible and engaging will also contribute to positive learning outcomes.
Decodable: Decodable books are those in which all the letter-sound combinations presented have been previously studied and memorized. A complete frequency analysis of how often each letter of the alphabet is used in the words of a given language is the foundational tool for designing decodable books.
Leveled: A leveling system will consider the structure, semantics and vocabulary level of a book and place books on a spectrum from easiest to hardest, using specific criteria such as complexity of expression, syntax and theme or concept.
Achieving a 1:1 ratio of books to learners is likewise critical; no child should have to share books in order to practice reading. In programs with few financial resources, prioritizing giving every child his/her own copy of a single book is more important than spending money to have a large variety of titles that children must then share (Global Book Fund Feasibility Study, 2016).
Effective Teachers and the Role of Coaching and Mentoring
Teachers must be trained on how to specifically teach reading. Their training must be as practical as possible (less theory, more practice). This is achieved by introducing teachers to the actual materials and routines they will use in classrooms, giving them ample time to practice using these materials, and providing them with an understanding of how to build from the study of phonics toward increasingly complex differentiated instruction (Reading Rockets, What Else Matters).
After they receive initial training in using a carefully designed (i.e. scripted, explicit) program of instruction for reading, teachers should also be coached and mentored on the effective use of that program in their own classrooms. Such coaching and mentoring should be supportive (never punitive), offering more praise than criticism, with no relation to a teacher’s pay or grade status in his/her system. Coaches should identify one to two areas for additional practice or improvement per visit and should offer specific, actionable feedback related to the reading program the teacher is learning to use. This is likely to require, at least at the outset, a minimum of two visits per teacher per month.
Continuous Assessments Inform Instruction
Children who are learning to read must be assessed by having them read out loud. The ability to match sounds to symbols, to decode with ease, to read fluently and to comprehend what is read are critical sub-components of reading and must all be monitored as a learner moves from being a non-reader to being a reader.
It is worth remembering that there is no point in conducting any type of skills inventory or student testing if there is no curriculum, printed program and teacher training routine for introducing actual reading instruction into schools. It is useless to measure progress in skills that are not actually being taught.
Finally, all testing and evaluation should be conducted using the paradigm that all children can learn to read. The only reason for collecting testing data is to determine how close or far a student is from reaching that goal. It should never be to stigmatize children as unable to read; testing and evaluation should examine how well a particular instructional intervention supports each student’s progress, and changes to facilitate reading mastery for all should be made to that program and to teachers’ practices based on the testing data available.
Regular Reading Practice Outside of School
Reading is a skill. The more it is practiced, the better a learner will become at it. Reading practice outside of school—at home with family and in community gathering places, for example—increases learners’ attainment of reading skills and helps foster a culture of literacy. In contexts where there is a lack of reading proficiency, families and communities can still help children gain literacy skills such as oral fluency, listening comprehension and sense of story by speaking with children and telling them stories.
Literate family members can be provided with tools, training and guides to help learners practice their reading, and illiterate family members can be provided strategies and tips on engaging with learners to encourage them to practice reading. Home-to-school routines and materials, where a learner is expected to bring something home from school every day and accomplish a task with a family member for reading practice, have been shown to improve learner outcomes. And at-home routines for illiterate families, such as oral storytelling time, or listening together and discussing a radio broadcast meant to increase vocabulary, have also been shown to support literacy development (Menendez, 2015).
Systematic Monitoring, Evaluation, Research and Learning (MERL) to Inform Adaptation
Reading programs must monitor and evaluate learners’ attainment of the five essential sub-skills of reading instruction (phonemic awareness, phonics, vocabulary, fluency and comprehension). Note that each of the key elements of a high-quality reading program must be designed, monitored and evaluated to increase learners’ attainment of these sub-skills.
Monitoring and evaluation works best when it is implemented throughout a program lifecycle. This means planning for it. This also means identifying through a system diagnostic and/or literature review where the key gaps are in expert understanding of what works in a particular context and targeting research, monitoring, evaluation and learning efforts to address those gaps and then taking the time to pause and reflect and then adapt based on emerging evidence.
USAID endorses the use of validated data collection instruments and processes in its programs. All attempts should be made to identify and adapt validated instruments to a particular context before creating new, untested instruments. USAID has developed many tools that might be useful in the monitoring and evaluation process, including the CLA Toolkit, the EGRA Toolkit, the EGRA itself, guidance on which assessment instruments are best suited for which purposes, an oral language assessment module, a classroom observation tool, a How-To Note on Collecting Data on Disability, a Framework for Gender Inclusion, and a How-To Note on Cost.
Effective reading is a necessary precondition for skill development in all academic areas. Countries only stand to benefit by investing in high-quality primary grade reading programs.