Research Brief Reading Code TZ V5 (2) (1)
- 1. NOVEMBER 2015 READING CODE: Assessing a Comprehensive Readership Initiative in Tanzania OVERVIEW Reading-CODE, a comprehensive readership initiative offered by CODE, works with local partners in Africa to provide culturally-relevant and engaging books that young people will want to read; supports libraries to distribute and care for books; and shares methods of instruction to help teachers engage children meaningfully with books to build their fluency and comprehension—especially their higher order comprehension and critical thinking. The goal of Reading-CODE is not simply to teach reading skills, but to create thoughtful, life-long readers. As it turns out, Reading-CODE programs do a superior job of teaching basic reading skills, too. Reading-CODE programs are active in Ethiopia, Ghana, Mali, Tanzania, Kenya, Mozambique, Liberia and Sierra Leone. Each program promotes certain core approaches to reading and writing, though these may vary from country to country depending on local priorities and traditions. CODE’s partner in Tanzania, the Children’s Book Project (CBP), was founded 25 years ago, and in those years CBP has made possible the publishing of 350 titles of books for children and youth, largely in Kiswahili, in millions of copies. Most of the dozen Tanzanian publishers of books for children are quick to credit CBP and CODE for their survival and even their existence. CBP added teacher training to its repertoire of services, after finding that providing books to teachers was not enough: teachers needed to know how to teach with the books. RESEARCH BRIEF Principal Researchers and Affiliations Charles Temple, Ph. D. Hobart & William Smith Colleges Geneva, New York Firas Elfarr Monitoring and Evaluation Coordinator CODE, Ottawa, Ontario CBP Staff Marcus Mgibili Ramadhan Ali Pilli Dumea Student Assessors University of Dar es Salaam Anania Christopher Chambo Aweso Edson T. Chambili Geniva Kazinja Nshuti T. Appoline Pacho Peter Tel: 613 232-3569 321 Chapel Street, Ottawa ON K1N 7Z2 codecan.org
- 2. Page | 2 PROGRAM PREMISE Quality education requires quality learning materials coupled with quality instruction. A quality book program includes: local authors, local illustrators, local issues gender-balance content that is relevant and worth thinking about design to support pedagogy for fluency and comprehension considerate of children's actual reading levels high standards of production Between 2012 and 2016 CODE and CBP, with funding from the Canadian government and CODE, initiated the Reading-Tanzania Project for 75 of the 105 primary schools in the Kongwa District of the Dodoma Region in central Tanzania to combine these approaches. The aim was to share Reading-Tanzania books in adequate numbers with schools, support school and classroom libraries, and share methods with teachers for teaching children to read and write, using the books. PROGRAM ACTIVITIES Book Production Textbooks can teach skills, but children need engaging, relevant, and varied reading materials that will develop the habit of reading, and grow their language capacity and knowledge of the world even as they inspire their imagination and curiosity. To achieve this there is a need for books that are created by local writers and artists -- books that allow children to recognize themselves and their surroundings and feel at home with the practice of reading. And for this supply to be sustained, local publishers must be part of the book chain. Good books are relevant - - relevant to the child’s reality, reflecting the child’s own environment. They need to be designed on sound pedagogical principles, taking into account reading levels, vocabulary and language and they need to be integrated into an instructional strategy wherein the teacher or librarian has the skills to get the most out of them.
- 3. Page | 3 The Children’s Book Project is not itself a publisher, but sets out specifications for illustrated books in Kiswahili that are needed for children at each grade level, and private publishers compete to supply the books, motivated by the carrot of a guaranteed purchase for successful manuscripts. CBP additionally conducts professional development workshops, often using international volunteer experts, to update the skills of local publishers, editors, writers, and illustrators. For Reading-Tanzania, CBP supported production of 30 new titles, including concept books for emergent readers, patterned books for learners, story books, and informational books. A highlight of the publishing initiative was two books in Kiswahili by a world-famous author, Tollolwa Mollel, originally from Tanzania but now a resident of Canada. CBP provided six titles of “big books,” too, which offer an attractive means to demonstrate concepts of print and are designed for reading aloud. They have been translated into English, French, Portuguese, Twi, Chichewa, Ekegusii and Bambara in addition to the original Kiswahili, and are distributed all around Africa. Library Support. All of the Reading-Tanzania project schools have school libraries, or at least classroom libraries. These are stocked with multiple copies of grade-appropriate books in sets of fifty so that every one, two, or at most three children can share a book. Teacher-librarians are trained to circulate and care for the books, and also to teach good book-handling practices to the children. Many of them conduct after-school library activities for Reading Clubs of children. Some school libraries have sufficient space for whole classes to visit at one time. Others are distribution points from which books are circulated to classrooms to be read.
- 4. Page | 4 Teacher Training. Reading-Tanzania’s training is organized as the Mbinu Saba, or “Seven Methods” of instruction. The methods are set out in a 100-page guidebook in both English and Kiswahili. The methods are: 1. Introducing students to literacy, addressing emergent literacy concepts—because children with little exposure to reading and writing need a sensible introduction to the purposes and nature of literacy in order to have a context before they begin learning new skills. 2. Phonological awareness, because half the equation of literacy is being aware of the units of spoken language—words, syllables, and phonemes—that are represented by units of print. Reading-Tanzania uses word games and graphic representations to make children aware of the sounds in their language. 3. Phonics, because research clearly shows that in order to be readers, children must be able to “crack the code.” But “the code” must be understood in the context of real reading, so Reading- Tanzania teaches phonics by contextualizing lessons in a whole-part-whole approach, beginning with reading a meaningful text, focusing on some of its parts, and reading meaningful text again. 4. Reading Fluency, because research shows that children who can read words quickly and accurately are better readers with more concentration available for understanding, just as practiced drivers can enjoy the scenery or plot better routes around obstacles. 5. Comprehension, using a three-part model with dozens of strategies for teachers to use before reading, to arouse prior knowledge and elicit curiosity and purpose; during reading, to follow text structures, guide an active search for meaning, and teach strategies of inquiry; and after reading, to teach students to reflect upon, interpret, debate, derive lessons from, and remember what they learned. 6. Vocabulary, because words are flashlights that illuminate different aspects of experience and enable children to notice and think about things. In Tanzania, where primary school is conducted in Kiswahili, even though children speak many other languages at home, teaching vocabulary and other language features is recommended as part of every lesson. 7. Writing, because literate people are producers of messages in print, too; and learning to write reinforces reading and language skills. The teachers’ mastery of the training is managed via a set of training standards and rubrics that are used during regular monitoring visits by project staff. Those who complete all of the training—three four day workshops—and receive satisfactory ratings during monitoring visits are certified as Reading-Tanzania teachers.
- 5. Page | 5 ASSESSMENT METHODOLOGY A team of six assessors were recruited from advanced students of Kiswahili at the University of Dar es Salaam and trained by Dr. Alison Preece of the University of Victoria and Dr. Charles Temple of Hobart and William Smith Colleges in using the Reading Tanzania assessment tools during two sessions totaling four days, one in August and one in November, 2015. The assessors traveled to the Kongwa region to conduct the Reading Tanzania assessment under the supervision of CBP staff, plus Dr. Temple, an international literacy consultant; and Firas Elfarr, CODE’s Monitoring and Evaluation Coordinator. The assessors tested a total of 104 children from 13 randomly chosen schools from the Kongwa District that were participating in the Reading-Tanzania project. For a comparison group, the District Education Officer of a nearby school district graciously allowed the assessors to test 48 children from six randomly chosen schools that had not participated in the Reading- Tanzania project. Children in the control schools came from the same ethnic groups as the project schools. The control schools were matched to project schools in terms of their “rurality”: schools that were located in a town, near a town, or far from a town. To evaluate students’ reading and writing, the assessors used a test developed for the Children’s Book Project by Dr. James Hoffman from the University of Texas-Austin and Dr. Misty Sailors from University of Texas-San Antonio. There were two levels of the test, one for standard or grade 2, and one for standard 4. Four children (two girls and two boys) were randomly selected for testing from each school.
- 6. Page | 6 At both standards 2 and 4 the children in the project schools outperformed the children in the control schools on all measures. At standard 2, the differences were especially pronounced in the areas of comprehension and fluency, as can be seen in Figure 1. Fluency and comprehension instruction are the main focus of the Reading-Tanzania project. These skills are developed through select instructional methods the teachers are trained to use, and supported by books in which the children can practice and develop those skills. The magnitude of the differences was impressive, however. Children in the project schools understood twice as much of what they read and were three times more fluent. FIGURE 1 8.5 9.2 4.2 4.3 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0 Boys Girls ReadingComprehensionMean Scores Project Schools Control Schools 1.81 1.88 0.58 0.75 0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40 1.60 1.80 2.00 Boys Girls FluencyRatingMeanScores Project Schools Control Schools STANDARD 2 TEST: Students from standard 2 were tested individually on these measures: Letter recognition Syllable reading Reading short sentences Storybook reading: word recognition Storybook reading: fluency Storybook reading: comprehension
- 7. Page | 7 Children in standard 2 also outperformed the control group of children on letter recognition, syllable reading, and word reading (See Figure 2). These skills are included in the Reading-Tanzania training, but not given so much emphasis as fluency and comprehension, because these more basic skills are so heavily emphasized in the LANES and EQUIP programs, Tanzania’s two national literacy projects ongoing at the moment. FIGURE 2 Table 1 shows the scores for children from project schools and control group schools for standard 2. 1 Mean, or average. 2 Standard deviation. 46.2 46.6 27.3 29.0 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 45.0 50.0 Boys Girls Meanno.ofWordsRead Accurately Project Schools Control Schools Table 1: Performance of Standard 2 Children Standard 2 Scores Project Schools Control Schools Boys (n = 26) Girls (n = 26) Boys (n = 12) Girls (n = 12) μ1 σ2 μ σ μ σ μ σ Letter recognition, lower 24.5 / 25 1.1 24.6 / 25 0.7 17.0 / 25 4.2 16.2 / 25 5.7 Letter recognition, upper 24.8 / 25 0.7 24.8 / 25 0.4 17.4 / 25 5.1 18.3 / 25 4.4 Syllable recognition 16.8 / 17 0.6 17.0 / 17 0.2 13.3 / 17 4.6 13.0 / 17 4.7 Sentence reading 14.6 / 15 2.0 15.0 / 15 0.2 10.1 / 15 5.4 11.8 / 15 4.5 Passage reading 46.2 / 47 2.2 46.6 / 47 1.1 27.3 / 47 15.1 29.0 / 47 14.5 Fluency 1.8 / 2 0.4 1.9 / 2 0.3 0.6 / 2 0.5 0.8 / 2 0.6 Reading comprehension 8.5 / 10 1.4 9.2 / 10 1.1 4.2 / 10 2.3 4.3 / 10 1.9
- 8. Page | 8 STANDARD 4 TEST Students from standard 4 were tested partly in small groups and partly individually on: Vocabulary, by means of matching a target word with its opposite Listening comprehension Silent reading comprehension Writing (composition, mechanics, and combined) Oral reading accuracy Oral reading comprehension Word recognition The smaller standard deviations in the project students’ scores are partly due to a ceiling effect, since many of the children had perfect scores on items of the test. But the larger standard deviations in the control group’s scores—many times larger than those of the project students— spell trouble for both teachers and children in the control schools. By standard 2, all children should know the letters of the alphabet, read simple syllables, and simple words in context. When there are many children who cannot do these things, especially in large classes, the teacher’s challenge to teach, and the children’s challenge to learn, is greatly intensified. As in standard 2, at standard 4, children in the project schools out performed children in the control schools on all measures, with the more pronounced differences in comprehension and writing (See Figures 3 and 4).
- 9. Page | 9 FIGURE 3 FIGURE 4 8.0 8.0 4.5 4.3 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0 Boys Girls ReadingComprehensionMean Scores Project Schools Control Schools 5.6 6.0 3.9 3.3 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 Boys Girls WritingScores(Compositionand Mechanics)Combined Project Schools Control Schools
- 10. Page | 10 At standard 4, the Reading-Tanzania project put heaviest emphasis on reading fluency, reading comprehension, and writing—so it is not surprising that in those areas the children in the project schools showed the larger advantage over the children in control schools. Children in project schools outperformed children in control schools on all other measures, too. The standard deviations for oral reading accuracy and word recognition were far greater in control schools, which again raises the concern that teachers in those schools will be challenged to reach the many children who are “not getting it,” especially in large classes. Note that the performance of boys and girls in both standards 2 and 4 was tracked separately. There was no appreciable difference in the performance of the two genders in either the project schools or the control schools Table 2 shows the scores for children from project schools and control group schools for standard 4. Table 2: Performance of Standard 4 Children Standard 4 Scores Project Schools Control Schools Boys (n = 26) Girls (n = 26) Boys (n = 12) Girls (n = 12) μ σ μ σ μ σ μ σ Vocabulary (opposites) 6.2 / 9 2.3 6.7 / 9 1.9 5.0 / 9 2.0 4.8 / 9 2.3 Listening comprehension 2.3 / 3 0.8 2.5 / 3 1.1 1.5 / 3 0.8 1.6 / 3 0.8 Silent reading comprehension 2.4 / 3 0.6 2.2 / 3 0.6 1.6 / 3 0.7 1.6 / 3 0.5 Writing composition 3.0 / 4 0.8 3.2 / 4 0.6 2.3 / 4 0.7 1.9 / 4 0.9 Writing mechanics 2.6 / 4 0.9 2.8 / 4 0.7 1.6 / 4 0.5 1.4 / 4 0.7 Writing scores combined 5.6 / 8 1.5 6.0 / 8 1.1 3.9 / 8 1.0 3.3 / 8 1.5 Oral reading accuracy 128.3 / 130 3.6 128.4 / 130 2.3 117.7 / 130 14.1 114.2 / 130 15.6 Oral reading comprehension 8.0 / 10 2.0 8.0 / 10 1.7 4.5 / 10 3.3 4.3 / 10 2.5 Word recognition 14.7 / 15 0.8 14.7 / 15 0.7 11.8 / 15 2.7 11.8 / 15 2.6
- 11. Page | 11 DISCUSSION On the basis of this assessment it was demonstrated that Reading-Tanzania has largely succeeded in its mission of creating capable readers and writers. Reading fluently, reading with understanding, and writing coherently and correctly are the core abilities of such readers and writers. Another key variable in the comprehensive readership model was beyond the scope of this assessment and that was the amount of actual practice readers and writers put in: that is, do students in the project schools show evidence of having the habit of reading and writing? That variable will be measured in a separate survey; but the circumstantial evidence points to a positive answer: first, because under the Reading-Tanzania grant the Children’s Book Project had produced thirty titles of highly attractive books for children of primary school age—and these supplemented the scores of such titles the CBP had produced before. Second, visits to school libraries showed students engaged in reading and discussion activities during regularly scheduled class visits to the libraries (in those schools whose libraries’ space permitted), and “Library Coaches” were able to show us whole boxes full of little books and other writings produced by children in their after school Reading Clubs. Shortly after Reading-Tanzania began in 2012, the Tanzanian Ministry of Education and Research began two large initiatives to train teachers across the country in methods of literacy instruction. One, LANES3 (Literacy and Numeracy Education Support), is funded by the government of Sweden, and the other, EQUIP-Tanzania4 (Education Quality Improvement Program for Tanzania), is funded by DFID. Both show the influence of the popular American literacy initiative, EGRA5 , in their focus on teaching very basic literacy skills, essentially the “Big Five” skills6 identified by the National Reading Panel7 commissioned in the 1990’s by the Bush Administration in the United States. Whatever the merits, there are two main criticisms of the EGRA approach. One critique is that the emphasis on phonological awareness and phonics was taken out of context. When the National Reading Panel recommended an emphasis on those skills, the point was to encourage American teachers to add them to a repertoire that was already very rich. Their classrooms were supplied with children’s 3 http://www.swedenabroad.com/Pages/StandardPage.aspx?id=70248&epslanguage=en-GB 4 http://www.equip-t.org/ 5 http://www.rti.org/pubs/bk-0007-1109-wetterberg.pdf 6 These are phonological awareness, phonics, fluency, vocabulary, and comprehension. 7 http://nationalreadingpanel.org/Publications/summary.htm
- 12. Page | 12 books, visual aids, and other resources, and the teachers themselves were skilled at using naturalistic approaches to teaching literacy, such as immersing children in print and encouraging children to compose meaningful messages with drawings, scribbles, and invented spelling—in keeping with the dictates of “emergent literacy.” The context of Tanzanian schools, especially in the rural areas, is a different world from the context that gave rise to the “Big 5.” In sum, this critique suggests that without paying more attention to emergent literacy, without being careful to offer a meaningful context to skills instruction, and without providing engaging books to attract children to reading and offer them practice, the emphases on the “big 5” will not be fully successful even in achieving their own limited objectives. After the staff and consultants of the Reading-Tanzania became aware of the emphases of the LANES and EQUIP-T programs the Mbinu Saba methods of the Reading-Tanzania project were expanded to include methods for nurturing children’s emergent literacy, and meaningfully contextualizing phonics instruction. The other critique of the heavy emphasis on the “Big 5” lower order literacy skills is that even if teaching the emphasis is successful on its own terms, the goal is not sufficiently ambitious. In the West, studies by the OECD8 link five levels of adults’ literacy to their levels of employment, income, political participation, and other quality of life indicators. Not surprisingly, those with the highest levels of literacy are the ones who derive the most benefits. What is surprising, though, is that people need to attain fairly high levels of literacy to achieve the higher incomes, better jobs, greater awareness, less need for public assistance, and greater political participation—as well as better health and lower rates of incarceration. Those whose skills are limited to decoding words accurately to retrieve explicit messages remain far behind those who have the power to apply what they read, make inferences, create interpretations, detect biases, and negotiate between competing messages. A leap of faith is required to assume that putting most of teachers’ emphasis on developing children’s phonological awareness and phonics skills will eventuate in their becoming insightful and critical readers. Reading-CODE takes a different approach. Reading-CODE does not view reading so much as a set of skills, but rather as a set of life-long habits and ways of thinking. Reading-CODE strives to create literate citizens who collectively form a literate culture. Toward that end, Reading-CODE produces and shares books, rooted in the local culture, that young people will want to read again and again and will want to talk about. And it equips teachers with the means to help students become fluent readers and deep thinkers. The data suggest that the Reading-CODE approach is working. 8 http://skills.oecd.org/skillsoutlook.html
- 13. Page | 13 RECOMMENDATIONS On the basis of the evaluation of the Reading-Tanzania project reported here, we are confident in recommending the following. 1. Reading-Tanzania methods, Mbinu Saba, should be shared as widely as possible in Tanzania. As noted above, the Tanzanian Ministry of Education and Research is currently embarked on two large- scale teacher training efforts, both intended to reach most of the teachers in Tanzania with methods for teaching basic literacy skills. Many teachers trained in Reading-Tanzania already are playing lead roles in both programs. We recommend that Reading-Tanzania methods be “piggy-backed” onto those initiatives. To that end, CODE should publish a user-friendly version of Mbinu Saba, the Reading-Tanzania guidebook of methods, in sufficient quantities to be used by large numbers of teachers. 2. Many of the children’s books published under arrangement with the CBP are already being purchased by EQUIP for distribution through the schools. We recommend that teachers’ guides be written for many of the books as a way of enhancing their effectiveness in the classroom, and as a vehicle for spreading good practices for teaching literacy more broadly. 3. Children learn to read by reading, and having accessible and engaging books available in large numbers for children is a valuable way to promote literacy. We recommend that an agency within the Ministry of Education and Research, in collaboration with the CBP and TEN/MET coordinate efforts of donors, so that adequate numbers of trade books can be published and made available to children in every standard in the primary schools. 4. Some 20 schools in the 75 schools in the Reading-Tanzania program have specially-funded enhanced libraries. These have more ample book shelves for displaying books attractively, carpets for seating whole classes of children, provision of additional numbers of books, and a paid Library Coach to conduct after school literacy activities and to coach Reading Clubs. These appear to be enormously valuable, though their evaluation was not part of this review. We recommend that they be expanded as far as possible through the primary schools of Tanzania.
- 14. Page | 14 ACKNOWLEDGEMENTS The authors wish to recognize documentary, photojournalist Eliza Powell for her company and stunning photography over the course of this research mission in Tanzania. Thanks as well to Audrey Roberson, Assistant Professor, Department of Education, Hobart & William Smith Colleges for her work supporting the data analysis. Copyright © 2015 by CODE 321 Chapel Street Ottawa ON K1Z 7N2 codecan.org Good instruction is important, but children learn to read by reading. That is the guiding principle of CODE, Canadian Organization for Development through Education. CODE has worked for over 55 years helping national partners make locally-written children’s trade books available in local and national languages to children and youth in many countries in Africa. CODE also supports school libraries in primary and secondary schools, and trains coaches to help teachers acquire sound methods for teaching reading and writing.