A Study of Extensive Reading with Remedial Reading Students

Lituanas, P. M., Jacobs, G. M., & Renandya, W. A. (1999). A study of extensive reading with remedial reading students. In Y. M. Cheah & S. M. Ng (Eds.), Language instructional issues in Asian classrooms (pp. 89-104). Newark, DE: International Development in Asia Committee, International Reading Association. A Study of Extensive Reading with Remedial Reading Students Abstract This book chapter reports a study designed to examine the effectiveness of an English-language ER program for remedial students at a public secondary school in the southern Philippines. Sixty first-year students at the school, 30 females and 30 males, who were to be assigned to remedial reading classes constituted the participants in this study. Using a matched-pairs design, each student was first matched with another of similar IQ, sex, socio-economic status, reading level, and past achievement. Then, one member of each pair was randomly assigned to the experimental remedial reading class, and the other member was assigned to the control class, so as to achieve balance on the variables in the two remedial reading classes. A Pre-test - Post-test Control Group design was used. The dependent variable, reading proficiency, was assessed via two instruments: the Informal Reading Inventory (IRI) (Johnson, Kress, & Pikulski, 1987), which yields scores from 0-100 on reading comprehension, and the Gray Standardized Oral Reading Test (GSORT) (Gray, 1967), which measures reading speed and accuracy, and indicates the grade level at which the student is reading. Both instruments were administered twice, once two months before the six-month treatment began and again after the treatment had been carried out. During the six months, both the control and experimental groups received 40 minutes of regular English class daily, plus an additional 40-minute remedial reading class. In their remedial reading class, the control group was taught in the conventional way from a textbook which included lessons on vowel and consonant sounds, minimal pairs, reading and reciting poems, and reading short selections. The only silent reading the control group did - and this infrequently - was of these short selections from their textbook. In contrast, the experimental remedial reading group took part in an ER program, the core of which consisted of students reading texts of their choice and doing a variety of post-reading activities. Post-test scores showed that the treatment group outperformed their control group peers to a statistically significant extent. Introduction The benefits of extensive reading (ER) for both first and second language learners are well-researched and well-known in a wide variety of countries including Asian countries (Anderson, 1996; Coady, 1997; Day & Bamford, 1997; Elley, 1996b; Jacobs, Davis, & Renandya, 1997; Krashen, 1993; McQuillan, 1994; Ng, 1988, 1994a, b, 1995, 1996; Yu, 1993, 1997a, b). However, despite the widely disseminated and strong evidence for the value of ER, implementation has often been infrequent and a less than complete success, especially in poorer countries which suffer from such problems as lack of reading materials, low teacher salaries, and inadequate preparation of teachers to implement ER (Greaney, 1996). Additional constraints on the implementation of ER exist even in countries with more favorable financial conditions. One of these constraints flows from pressure brought by administrators, students, and parents to cover the entire syllabus and to complete every page in the textbook and every exercise in the workbook. Such pressure leaves little or no time for ER, which is relegated to the status of "luxury" or "optional extra" (Yu, 1993). Exam pressure poses another obstacle to ER implementation, especially when these exams measure only discrete skills and when such exams form the only means of assessing student learning and the quality of instruction, neglecting consideration of students' attitudes toward reading or of their ability to deal with large pieces of text. An even more fundamental impediment to greater and more successful use of ER - one that underpins the obstacles discussed above - lies in the belief that the best way for students to increase their literacy skills and to become lifelong learners focuses on part-to-whole instruction, in which students first master the parts of language, e.g., vocabulary and grammar, via direct instruction in these parts before putting the parts together to read whole texts. In contrast, the key belief underlying implementation of ER is that students can best learn the parts of a language indirectly by reading whole texts supplemented by some instruction in the parts (Anderson, 1996; Yu, 1993). Research on ER, including some done in second language contexts in the Asia-Pacific region, provides one means of supporting the efficacy of instruction focused on whole texts (Elley, 1996a; Mason & Krashen, 1997), as we would not expect educators to change their views based solely on the abstract logic of the arguments supporting the use of whole texts. In the present chapter, we report a study designed to examine the effectiveness of an English-language ER program for remedial students at a public secondary school in the southern Philippines. The population of the Philippines totals approximately 70 million with a per capita annual income of about US$1000. Gonzalez (1997) provides an overview of the education system in the country based on data for the 1994-1995 school year which show a total of 17,538,049 pupils, with 10,903,529 in 35,671 primary schools (which last for six years), 4,762,877 in 6,055 secondary schools (which last for four years), and 1,871,643 studying at 1,181 colleges and universities. These data exclude students in post-secondary non-degree programmes. While 93% of primary school pupils studied in public schools, only 68% of secondary students and 21% of tertiary level students were in public schools. Class sizes normally ranged from 40-60 students per class. Many classrooms were not able to benefit from electronic teaching aids, as only 51% of municipalities were able to provide distributed sources of power, not to mention the cost of such equipment. However, Gonzalez reports that some affluent schools in urban areas were endowed with electronic teaching devices, including computers. 1. The 1987 Philippines constitution states that Filipino is the national language and that "for the purposes of communication and instruction, the official languages of the Philippines are Filipino and, until otherwise provided by law, English" (cited in Garcia, 1997: 74). Since 1974, the Philippines has had a bilingual education policy. Currently, students study some subjects - Mathematics, Science, and English - in English and other subjects - Social Studies, Values Education, Technology and Home Economics, and Filipino - in Filipino. Although Filipino is the national language, neither it nor English is the first language of many students. For instance, in central and southern Philippines the major L1 is Cebuano, not Tagalog (the basis for the national language). Thus, such students face two mediums of instruction, neither of which is their L1 (first language). Next, we present an overview of ER before proceeding to a description of the study. Drawing on data from studies of Philippines classrooms conducted in the 1980s and 1990s, Gonzalez (1997: 59-61), currently the country's Secretary of Education, offers the following generalizations: 1. No classroom in the Philippines is really monolingual. What happens is continuing code-switching (the local language, Filipino, and English) with bi-medial instruction (the local language and English/Filipino depending on the subject and the language supposed to be used for the subject), ... 2. The proportion of the teacher-talk to pupil-talk is 7:3, with the teacher doing most of the talking -- at all levels. 3. Even at the upper levels, the reduced pupil-talk consists of one or two word or phrasal answers to WH-questions ...; the answers are formulaic and basically fill in the blanks in prefabricated sentences already framed by the teachers' question: ... 4. The pupils seldom ask questions or make comments or requests ... 5. Using Bloom's taxonomy (Bloom, 1956) of education objectives, questions are of the basic factual information type ... 6. The best correlate for achievement in all subjects including language is socioeconomic level. Bearing in mind that broad generalizations fail miserably in attempting to capture the wide diversity of a country of 7000 islands, here are a few. In the 1960s, literature and language were two separate subjects in the Philippines. Now that they are combined, many teachers no longer push students to read books and stories. Instead, they emphasize the teaching of rules of grammar. ER is now sometimes a privilege only of classes of homogeneously grouped fast learners. Otherwise, in the typical class, oral reading may be focused on more than silent reading, and part-to-whole instruction may dominate, with an emphasis on phonics at the lower elementary school level. Extensive Reading: What and Why? Extensive reading can be defined as the reading of large quantities of material for information or pleasure. In extensive reading, the immediate focus is on the content being read, rather than on language skills. Many names have been given to ER programs, including Book Flood, Uninterrupted Sustained Silent Reading (USSR), Drop Everything and Read (DEAR), Silent Uninterrupted Reading for Fun (SURF), and Extensive Reading and Information Literacy (ERIL). Although ER focuses on students reading alone, ER programs can involve group activities (e.g., Daniels, 1994) that motivate students to read more and provide them an avenue for discussing what they have read. ER programs are often beneficially combined with explicit forms of instruction, such as intensive reading. This second type of reading normally involves students reading small amounts of text under a teacher's supervision. Intensive reading focuses mainly on language skills, such as learning specific vocabulary, grammar structures, or reading strategies, rather than on the message of the reading text. As noted above, in intensive reading small amounts of text are read, as a good deal of time is spent on using the text as a vehicle for teaching language and reading skills, whereas with ER large amounts of text are read. Also, the texts are usually at students' instructional reading level, i.e., they need some help from dictionaries, teachers, or other sources to understand the text, whereas with ER the texts are at students' independent reading level, i.e., while students may not understand every word, they can comprehend the text on their own. As ER and intensive reading should be combined, school timetables can be set up so that students spend some time reading silently, some time on activities based on the materials they have read during extensive reading, and some time devoted to direct reading instruction. [For more details on setting up ER programs, especially in L2 (second language) classrooms, Day & Bamford (1997) is an excellent recent book, which can be supplemented by ideas from the authors of the collection edited by Jacobs et al. (1997).] The following advantages have been proposed for ER (Yu, 1993): 1. Increased knowledge of the world. 2. Enhanced language acquisition in such areas as grammar, vocabulary, and text structure. 3. Improved reading and writing skills. 4. Greater enjoyment of reading. 5. Higher possibility of developing a reading habit. 6. Opportunities to individualize instruction (Nolasco & Arthur, 1988). Rationales for these proposed advantages of ER range from the common sense - we learn to X (in this case, read) by doing X (in this case, reading) - to the currently more esoteric - e.g., chaos theory (Larsen-Freeman, 1997) which postulates that dynamic, complex non-linear systems such as human language are self-organizing, given sufficient input and feedback, and reading provides one source of such input and feedback. A more common scholarly explanation of the benefits of ER argues that the human brain contains innate potential for language learning of both L1 and L2s. This potential is known as language acquisition device or universal grammar (Chomsky, 1965; Cook, 1988). The large quantities of meaningful and comprehensible input provided by ER activate that potential, thereby fostering language acquisition, as learners induce the rules of grammar and other language elements, such as spelling, from the data they receive in their environment (Krashen, 1993). This innate ability enables young children to gain mastery of most of their first language's rules and a good deal of its vocabulary regardless of their socio-economic status and intelligence. We generally agree with this nativist view, and feel that the same processes come into play for the learning of second languages, but we also see the possible benefit of what interactionist theorists (Larsen-Freeman & Long, 1991; Swain, in press) have proposed, i.e., that while comprehensible input is an absolutely crucial condition for second language acquisition, it may by itself not be sufficient. The effectiveness of ER may be further enhanced by such means as students engaging in activities in which they talk and write about what they have read and will read. This talking and writing can help make the reading more comprehensible and may provide a means for students to "infect" each other with the joy of reading. Talking and writing also push students to move from the receptive language competence needed for reading to the more demanding productive competence required for speaking and writing. Additionally, interactionists learners can benefit from a small amount of explicit language instruction in the overall context of an instructional programme featuring large quantities of comprehensible input by such means as ER. Extensive Reading: How? Experts on ER (e.g., Yu, 1993) suggest the following characteristics for successful programs. 1. A large selection of reading materials to suit various reading levels and interests. 2. Time set aside for students to read during school. 3. Teachers who: a. read silently along with students and tell students about what they read, b. read aloud to students, c. teach reading skills, d. ask students to share with their classmates about their reading, and e. monitor students' ER progress. 4. Engaging post-reading tasks, ones which do not take away from the joy of read and that do some or all of the following: a. allow students to "advertise" to peers the texts they have enjoyed, b. help teachers and students check students' progress, c. provide students with some check and demonstrate their understanding, d. encourage students to apply and develop their understanding of concepts and issues addressed in their reading in a variety of ways, including via art, music, and drama. From our own observations and from talking with colleagues in various Asian countries, we feel that much good work in ER does takes place. For examples of such programmes, see the collection edited by Jacobs, Davis, and Renandya (1997) which contains chapters describing successful ER programmes in a number of countries in Southeast Asia and elsewhere around the globe. For instance, Lie (1997) describes ER among Indonesian university students, Smith (1997) explores the establishment of an ER programme in a Brunei secondary school, and Cockburn, Isbister, and Sim-Goh (1997) explain a buddy reading programme in Singapore in which older primary school students promote reading among schoolmates from lower grades. However, a gap often exists between, on the one hand, what theory and research indicate would be beneficial to learning and what actually is implemented and sustained in classrooms, on the other hand (Rodriguez-Trujillo, 1996). Despite the success stories mentioned in the preceding paragraph, sustained, well-run programs are more often the exception than the rule. Effective ER programs seem especially scarce for lower achieving students, as many educators express the view that such students lack the desire and skills to read extensively. Thus, further research is needed to develop and test situation-appropriate ER implementation with lower-achieving students. We now state the research questions used in the present study, one which investigated an attempt to engage a group of these lower-achieving pupils in ER. Then, the methodology used in the study will be described. Research Questions Two research questions were formulated: 1. Will there be a significant difference in the pre-test reading proficiency scores of the control group (students who do not participate in an ER program) and the experimental group (students who do participate in an ER program)? 2. Will there be a significant difference in the post-test reading proficiency scores of the control and experimental groups? The second question was the one of interest. The first one was set in order to test whether the randomization procedures used before the study began had succeeded in yielding control and experimental groups that were indeed matched as to initial reading proficiency. Method Participants Students at a public secondary school on the island of Mindanao in the southern Philippines participated in the study. The two-story school boasts clean, beautiful grounds, and has received an award for being the most effective secondary school in Region X. However, the school lacks a gymnasium and AV room, and the library is housed in a dilapidated building. School enrolment stood at more than 2800 for the 1997-1998 academic year, with an average of 52 students per class. Ninety percent of the students at this school come from low-income homes where reading materials tend to be scarce. Many of them do not live with their own families. Instead, they live with other families who pay for their schooling in return for work around the house and elsewhere. Indeed, some students even support their families by working at night. For instance, they might sell "balut", eggs that are about to hatch. Thus, many have little time or energy after school for academic tasks, and without an in-class ER programme they are likely to do little reading. Most classes last 40 minutes except for Science and for Technology and Home Economics which last 80 minutes. In addition to a bulletin board, every classroom is enlivened by various corners. The Filipiniana corner features displays on Philippines heroes, tourist spots, and folk dances. Other corners focus on science and on drug prevention. Drug abuse is a problem among a small number of students including some who participated in the study. The study was conducted over a period of six months from September 1996 to January 1997. In September, 60 first-year students at the school, 30 females and 30 males, who were to be assigned to remedial reading classes constituted the participants in this study. Their ages ranged from 12-18. Using a matched-pairs design, each student was first matched with another of similar IQ, sex, socio-economic status, reading level, and past achievement. Then, one member of each pair was randomly assigned to the experimental remedial reading class, and the other member was assigned to the control class, so as to achieve balance on the variables in the two remedial reading classes. Procedure A Pre-test - Post-test Control Group design was used. The dependent variable, reading proficiency, was assessed via two instruments: the Informal Reading Inventory (IRI) (Johnson, Kress, & Pikulski, 1987), which yields scores from 0-100 on reading comprehension, and the Gray Standardized Oral Reading Test (GSORT) (Gray, 1967), which measures reading speed and accuracy, and indicates the grade level at which the student is reading. These instruments had been used previously by local researchers in the same area of Philippines and were found to be useful. Both instruments were administered twice, once two months before the treatment began and again after the treatment had been carried out. The pre-test was administered by the first author with assistance from other teachers at the school, while the post-test was administered by other teachers, in an effort to lessen experimenter bias. The treatment lasted six months. During that period, both the control and experimental groups received 40 minutes of regular English class daily, plus an additional 40-minute remedial reading class. The first author conducted both remedial classes, whereas two different teachers taught the regular English classes. The school had no ER program, and it is not common for teachers to have initiated their own. The regular English class followed the same syllabus for both groups, while the remedial reading class varied. A typical reading class period - to the extent that there is a typical reading class period - at the school in which the present study was conducted begins, like all classes, with a prayer, followed by the class and teacher exchanging greetings. The teacher then asks students to open their textbooks to a given page. Next, the teacher might introduce the topic of the reading text and/or ask students to skim or scan the reading passage. The teacher then reads aloud the text, unlocking difficult vocabulary and grammar points. Next, students are asked to answer questions to check their comprehension. This involves silent reading of at least parts of the passage. An integrated skills approach is used, so that in any given lesson in the textbook (it takes several classes to cover one textbook lesson) students are listening, speaking, reading, writing, and studying grammar. In their remedial reading class, the control group were taught the conventional way from a textbook which included lessons on vowel and consonant sounds, minimal pairs, reading and reciting poems, and reading short selections. The only silent reading the control group did - and this infrequently - was of these short selections from their textbook. In contrast, the experimental remedial reading group took part in an ER program. The core of the ER program consisted of students reading texts of their choice and then doing a variety of post-reading activities. The female students tended to choose fiction, whereas the males often preferred non-fiction, such as news and feature articles from magazines. Care was taken that student chose books that were at their independent reading level. Obtaining materials took a good deal of effort and time, but a barely sufficient collection was assembled from students themselves, fellow teachers, the school library, and donations of money or materials from the community (Lituanas, 1997). [One method that was not employed was for teachers (Guadart, 1994) and students (Davidson, Ogle, Ross, Tuhaka, & Ng, 1997) to write ER materials.] In the experimental group's remedial class, students spent about 45% of the time doing silent reading (about 20 minutes per lesson), with another 45% spent on pre- and post-reading activities (mostly post-reading) which included attention to students' problem areas in reading. The remaining 10% of class time was spent on classroom management, including disciplining unruly students. The teacher used various techniques to encourage students to read more and to increase their reading skills, such as: 1. Reading aloud by the teacher. 2. Asking students to predict what a story was about using such clues as the title, cover, and illustrations. 3. Giving brief summaries/reviews of materials she had read and enjoyed. 4. Asking students to summarize for the entire class material they had read and enjoyed. 5. Chatting with individual students about what they were reading or had read. 6. Monitoring students' progress in ER and involving them in such monitoring. While students were reading silently, the teacher would: 1. Read on her own (10%). 2. Assist students to select reading material (10%). 3. Help students, e.g., by answering questions and by sitting beside students who had difficulty recognizing words and guiding them (80%). Ideally, the teacher would have spent a much larger percentage of the time reading on her own as a model for students. However, given the difficulty of obtaining ER materials suitable to students' interest and reading levels, it seemed more important to spend time on the two other activities. Post-reading activities included: 1. Answering higher order thinking questions, as part of a game called "Book Wheels" (Jacobs, 1993). 2. Role play. 3. Retelling. 4. Mock interviews in which one student portrayed a character in the story that other students would then interview. 5. Adding new words encountered while reading to a personalized vocabulary notebook (Kweldju, 1997). These post-reading activities provided one means of attracting the less-diligent students to ER, because they enjoyed the stories that were related by their classmates who had done the reading and follow-up tasks. Nevertheless, the participation of these few lessdiligent students remained unsatisfactory. Table 1 provides an overview of how ER was implemented with the experimental group. Table 1 Guidelines for ER Programs and How They Were Implemented in the Current Study ER PROGRAM GUIDELINES IMPLEMENTATION 1. Large selection of materials for Materials were obtained by the students various reading levels and interests and teacher from a variety of sources 2. Time set aside for students to read 45% of class time was reserved for during school silent reading, and pre- and postreading was designed to encourage students to also read at home 3. Teachers who encourage students to The teacher read silently while student read read and talked about what she read, read aloud and had students predict what would happen next, asked students to share with classmates about what they read, and students and teachers monitored student progress 4. Engaging post-reading tasks Games, role play, retelling, mock interviews, vocabulary notebooks were used Data Analysis T-tests were used to compare the pre-test scores of the control and experimental groups on the GSORT and the IRI to establish if they were indeed roughly equivalent on the dependent variable as it was operationalized in this study, i.e., the two measures of reading proficiency, before the study began. T-tests were used again to compare the posttest scores of the two groups on the two instruments to see if the treatment might have been associated with any difference in reading proficiency. A familywise alpha level of .05 was set for all t-tests. With degrees of freedom of 29, the critical value was approximately 2.67. This higher than normal critical value for a .05 alpha level (compared to the normal 2.04) was used to compensate for the fact that four t-tests were done. Results Table 2 shows the pre-test results on both measures of reading proficiency. The t-tests suggest, in answer to the first research question, that there were no significant differences between the control and experimental groups prior to the inception of the six-month experimental period. Table 2 t-tests Comparing Control and Experimental Groups on Pre-tests of Reading Proficiency Proficiency Test n Mean s.d. 30 7.89 4.81 t IRI Control 0.20 (n.s.) Experimental 30 7.61 5.19 30 2.55 0.73 GSORT Control 0.86 (n.s.) Experimental 30 2.38 0.83 n.s. = nonsignificant Table 3 shows the post-test results on both measures of reading proficiency. The mean post-test IRI score for the control group was 12.28 compared to 32.57 for the experimental group. On the GSORT, mean score for the control group was 3.96 compared to 5.25 for students in the ER group. As mentioned in the Procedure section, GSORT is an indicator of the grade level at which the student is reading. Since students in our study were in Grade 7, they should have scored at least 7 on the GSORT. However, as Table 2 indicates, the ER students were still reading at 5.25 level, somewhat below their expected level. Table 3 t-tests Comparing Control and Experimental Groups on Post-tests of Reading Proficiency Proficiency Test n Mean s.d. 30 12.28 5.77 IRI Control t 6.72* Experimental 30 32.57 14.80 30 3.96 0.88 GSORT Control 5.31* Experimental 30 5.25 1.16 * Significant at p < .05 The t-tests suggest, in answer to the second research question, that after the six-month experiment, significant differences existed between the control and experimental groups in terms of reading proficiency. As a follow-up statistical procedure to measure the amount of variance in scores on the two reading tests accounted for by the independent variable (whether students were in the control or experimental group), eta squared tests were run. Results showed that the experimental treatment accounted for 61% of the variance in IRI scores and 49% of the variance in GSORT scores, an average of 55%, a very strong association (Hatch & Lazaraton, 1991). Discussion The results of the current study suggest that a well-conducted ER program may be able to make a significant impact on reading proficiency, even when students are of below average level and where reading materials are scarce. However, it should be noted that despite their impressive gains, students in the experimental group were still reading below grade level, as indicated by their GSORT scores1). Perhaps, continued extensive reading would be part of the necessary elements of a program for overcoming this deficit. From a research perspective, the two groups would, ideally, have continued the control and experimental treatments for a longer period to see if the effect remained and if the experimental group continued their progress. Also, the research design would have been improved had follow-up been done to investigate whether the ER programme was associated with more out-of-class reading by students after the programme had ended. Constraints on the first author's time, unfortunately, did not permit this. However, the first author continues to use ER and to inform other teachers about it, both at her school and elsewhere. The relative success of ER with the remedial students in this study may have important implications. Early lack of success in reading often leads students down a slippery slope to failure in other academic areas, low academic self-image, low motivation to study, high likelihood of dropping out of school, high delinquency rates, and poor career prospects after leaving school (Goodlad, 1983). Further, instruction for such students may sometimes be of lower quality, focusing on drills and other lowerorder thinking tasks, as teachers may have inappropriately low expectations for what these students can achieve (Oakes, 1985). Many intervention programs have been implemented to meet the needs of such low achieving students. Successful programs focus on early and intensive intervention, and use well-researched pedagogy (e.g., Clay, 1996; Slavin, Madden, Karweit, Dolan, & Wasik 1992). However, unlike the two early intervention programs referenced in the preceding sentence that are typically used with lower primary school students, the program described in this chapter was carried out with secondary school students. These older students may be more difficult to reach, as they have suffered many years of reading failure. ER, for the many reasons reviewed in the introduction to this chapter, certainly appears to belong in intervention programs for students with reading difficulties. In the present study, we saw a significantly greater improvement in proficiency for those students who participated in the ER programme implemented by the first author. This program, we would like to stress, followed the guidelines for effective ER mentioned earlier in this chapter. How this was done was displayed in Table 1 in the Procedures section above. Teachers will often need support to implement ER, support in the form of teacher development workshops and follow-up coaching, time for teachers to assist each other on ER implementation, administration backing for devoting time to silent reading, funds to purchase reading materials for class and school libraries, and help from students' homes to encourage them to make reading a habit. Lack of such support is a key reason why, despite the apparent success of an ER program with one group of students at their school, many other teachers at the school where the present study was conducted have remained reluctant to initiate ER with their students. In conclusion, students who are not currently skilled, enthusiastic readers face unnecessary and serious obstacles to realizing their potential contributions to themselves, their families, and to society in general. In this information age, they will be shut off from the power gained through obtaining and providing information and from the splendor and inspiration of good fiction. Thus, educators need to create and implement programs to help students who fall behind in reading. The accumulated wisdom embodied in the current study and the many which came before it strongly suggests that ER can play an important role in helping students gain in their level of reading skill. Reading skills and the benefits that flow from them are essential if students are to become people who, to paraphrase Friere (1970), use the word to know and change the world. References Anderson, R.C. (1996). Research foundations to support wide reading. In V. Greaney (Ed.), Promoting reading in developing countries: Views on making reading materials accessible to increase literacy levels (pp. 55-77). Newark, DE: International Reading Association. Chomsky, N. (1965). Aspects of the theory of syntax. Cambridge, MA: MIT Press. Clay, M. (1996). Reading recovery (updated ed.). Auckland: Heinemann Education. Coady, J. (1997). L2 vocabulary acquisition through extensive reading. In J. Coady & T. Huckin (Eds.), Second language vocabulary acquisition (pp. 225-237). Cambridge: Cambridge University Press. Cockburn, L., Isbister, S., & Sim-Goh, M.L. (1997). Buddy reading. In G.M. Jacobs, C. Davis, & W.A. Renandya (Eds.), Successful strategies for extensive reading (pp. 65-80). Singapore: SEAMEO Regional Language Centre. Cook, V. (1988). Chomsky's universal grammar. New York: Basil Blackwell. Daniels, H. (1994). Literature circles: Voice and choice in the student-centered classroom. York, ME: Stenhouse. Davidson, C., Ogle, D., Ross, D., Tuhaka, J., & Ng, S.M. (1997). Student-created reading materials for extensive reading. In G.M. Jacobs, C. Davis, & W.A. Renandya (Eds.), Successful strategies for extensive reading (pp. 144-160). Singapore: SEAMEO Regional Language Centre. Day, R. R., & Bamford, J. (1997). Extensive reading in the second language classroom. New York: Cambridge. Elley, W. (1996a). Lifting literacy levels in developing countries: Some implications from an IEA study. In V. Greaney (Ed.), Promoting reading in developing countries: Views on making reading materials accessible to increase literacy levels (pp. 39-54). Newark, DE: International Reading Association. Elley, W. (1996b). Using Book Floods to raise literacy levels in developing countries. In V. Greaney (Ed.), Promoting reading in developing countries: Views on making reading materials accessible to increase literacy levels (pp. 148-163). Newark, DE: International Reading Association. Friere, P. (1970). Pedagogy of the oppressed. New York: Seabury Press. Garcia, E.A. (1997). The language policy in education. In M.L.S. Bautista (Ed.) English in an Asian language: The Philippine context (73-86). Sydney: Macquarie University. Gaudart, H. (1994). Selecting readers: Children's choice. In M.L. Tickoo (Ed.) Research in reading and writing: A Southeast Asian collection (pp. 63-78). Singapore: SEAMEO Regional Language Centre. Gonzalez, A. (1997). Language classrooms of tomorrow in the Philippines: Dreams and realities. In G.M. Jacobs (Ed.) Language classrooms of tomorrow (55-66). SEAMEO Regional Language Centre: Singapore. Goodlad, J. (1983). A place called school. New York: McGraw-Hill. Gray, W.S. (1967). Gray Oral Reading Tests. Indianapolis, IN: Bobbs-Merrill. Greaney, V. (Ed.). (1996). Promoting reading in developing countries: Views on making reading materials accessible to increase literacy levels. Newark, DE: International Reading Association. Hatch, E., & Lazaraton, A. (1991). The research manual: Design and statistics for Applied Linguistics. New York: Newbury House. Jacobs, G.M. (1993). Book wheels. In R. R. Day (Ed.), New ways in teaching reading (pp. 5-6). Alexandria, VA: TESOL. Jacobs, G.M., Davis, C., & Renandya, W.A. (Eds.). (1997). Successful strategies for extensive reading. Singapore: SEAMEO Regional Language Centre. Johnson, M.S., Kress, R.A., & Pikulski, J.J. (1987). Informal reading inventories (2nd ed.). Newark, DE: International Reading Association. Krashen, S. (1993). The power of reading. Englewood, CO: Libraries Unlimited. Kweldju, S. 1997. Measuring English department students' vocabulary size, and developing an individualized tutoring model by incorporating vocabulary to extensive reading course (Research report). Singapore: SEAMEO Regional Language Centre. Larsen-Freeman, D. (1997). Chaos/complexity science and second language acquisition. Applied Linguistics, 18(2), 141-165. Larsen-Freeman, D., & Long, M.H. (1991). An introduction to second language acquisition research. London: Longman. Lie, A. (1997). The reading and writing connection: Community journal. In G.M. Jacobs, C. Davis, & W.A. Renandya (Eds.), Successful strategies for extensive reading (pp. 161-170). Singapore: SEAMEO Regional Language Centre. Lituanas, P.M. (1997). Collecting materials for extensive reading. In G.M. Jacobs, C. Davis, & W.A. Renandya (Eds.), Successful strategies for extensive reading (pp. 25-29). Singapore: SEAMEO Regional Language Centre. Mason, B., & Krashen, S. (1997). Can extensive reading help unmotivated students of EFL improve? I.T.L. Review of Applied Linguistics, 117-118, 79-84. McQuillan, J. (1994). Reading versus grammar: What students think is pleasurable for language acquisition. Applied Language Learning, 5(2), 95-100. Ng, S.M. (1988). Research into children's language and reading development. Singapore: Institute of Education. Ng, S.M. (1994a). Changing the English language curriculum in Brunei Darussalam. International Journal of Educational Development. 14(4), 361-370. Ng, S.M. (1994b). Improving English language learning in the Upper Primary levels in Brunei Darussalam. In Tickoo, M.L.(Ed.) Research in reading and writing: A Southeast Asian collection (pp. 41-54). Singapore: SEAMEO Regional Language Centre. Ng, S.M. (1995). Language teaching and bilingual literacy at pre-primary and primary levels in Hong Kong. In Consultation Document by the Education Commission Report No. 6, Part 2. (Annexes) Hong Kong. Ng, S.M. (1996). Innovation, survival and processes of change in the bilingual classroom in Brunei Darussalam. Journal of Multilingual and Multicultural Development, 17(2-4), 149-162. Nolasco, R., & Arthur, L. (1988). Large classes. London: Macmillan. Oakes, J. (1985). Keeping track: How schools structure inequality. New Haven, CN: Yale University Press. Rodriguez-Trujillo, N. (1996). Promoting independent reading: Venezuelan and Colombian experiences. In V. Greaney (Ed.), Promoting reading in developing countries: Views on making reading materials accessible to increase literacy levels (pp. 109-129). Newark, DE: International Reading Association. Smith, R. (1997). Transforming a non-reading culture. In G.M. Jacobs, C. Davis, & W.A. Renandya (Eds.), Successful strategies for extensive reading (pp. 30-43). Singapore: SEAMEO Regional Language Centre. Slavin, R.E., Madden, N.A., Karweit, N.L., Dolan, L., & Wasik, B.A. (1992). Success for All: A relentless approach to prevention and early intervention in elementary school. Arlington, VA: Educational Research Service. Swain, M. in press. Integrating language and content teaching through collaborative tasks. In New insights for the language teacher. C.S. Ward and W.A. Renandya (Eds.). Singapore: SEAMEO Regional Language Centre. Yu, V. (1993). Extensive reading programs--How can they best benefit the teaching and learning of English? TESL Reporter, 26(1), 1-9. Yu, V.W.S. (1997a, March) How students respond to an extensive reading programme. Paper presented at the annual meeting of TESOL (Teachers of English to Speakers of Other Languages), Orlando, USA. Yu, V.W.S. (1997b, March) The impact of a self - access reading scheme on students' reading habits, attitudes and second language development. Independence (IATEFL [International Association of Teachers of English as a Foreign Language] Learner Independence SIG), p. 18. Acknowledgements: The authors wish to thank Ms Teresita Taladua and Miss Rebecca Alvarico, the two teachers of the study's participants in their regular classes, who generously gave of their time and insights; Dr Angel Pesirla of Cebu Normal University for his insights into Philippines education; Director Edwin Goh, SEAMEO Regional Language Centre, who consistently supports classroom-based research; and Mr Donald Teo, SEAMEO Regional Language Centre, who skillfully and diligently coordinated the communication between the authors of this report.