Extensive Reading in English Communication Courses

Ken Schmidt

1. Introduction

In recent years, an increasing number of language instructors have turned to extensive reading as a means of facilititating acquisition among their students. Learners in such programs typically select books from a wide variety of genres and language levels and read them for interest and enjoyment, with minimal post-reading tasks (e.g., book reports, essays)the emphasis being on maximizing input rather than output (Dawson, 1992).

During the 19967 academic year, the author implemented an extensive reading program as a component of his English conversation and English communication courses at Tohoku Gakuin University. It was hoped that this would be an enjoyable, motivating source of plentiful, out-of-class input that would feed into learners' overall language abilities and provide stimulating fodder for communication in the classroom. This overview of the initial year of the program includes discussion of (a) theoretical and practical bases for extensive reading and its use in communication oriented courses, (b) program implementation, (c) activities utilizing reading as a resource for communication in the classroom, and (d) student response based on reading quantity, an exit questionnaire, and book reports. Ideas for future development and research are also offered.

2. Comprehensible Input And Acquisition

Research indicates that second language acquisition can be aided by explicit language study (e.g., rule giving, consciousness raising, vocabulary work) (Ellis, 1990; Schmitt, 1995), and meaningful language use (Brown, 1994; Long, 1990) in interactive contexts (Pica, Young and Doughty, 1987; Swain, 1985). There is, however, strong evidence that the primary requisite for significant acquisition is massive comprehensible input (Krashen, 1988; Nation, 1997). Although many would argue that input alone is not sufficient for gaining native-like fluency in a foreign language (Lightbown and Spada, 1993), few would deny its necessity.

The key concept here is comprehension or understanding. No one learns a dissimilar second language merely by listening to unintelligible talk radio in the L2. Learners must be able to draw meaning from the input they attempt to access (Krashen, 1988; Ying, 1995). Conversely, not everything need or should be understood; "I+1" input, for which the learner occasionally has to infer meaning or wait for more data is seen as ideal for acquisition (Krashen, 1988). With more and more such input, the learner is repeatedly exposed to words, expressions, structures, and aspects of discourse. With each exposure, the learner adds to his or her mental mapping of these features and how they are used in the target language (Ellis, 1995). In other words, learners begin to form ideas of the meaning and usage of new features, while extending and deepening their understanding of more familiar onesjust as learners acquire much of their first language (Krashen, 1988).

What implications does this have for Japanese college students of English? One is rarely if ever called on to understand English in Japanese daily life, and although there are many potential sources of English input (TV, video, radio, newspapers), how much of this content is comprehensible to learners (particularly at lower levels), and how often do they actually access these sources?

Students may receive considerable input through their college English courses, but a great deal of time, even in communicative classes, may be spent in intensive study of relatively short texts (written or spoken) and/or in production through writing and speaking. These activities can play a key role in helping students become aware of language features, build useful strategies, and gain confidence and mechanical facility in the skill areas, and can offer a tremendous "leg up" in making texts comprehensible. They are of limited aid, however, in helping students obtain input in maximal amounts and develop sustained automaticity and fluency in decoding and interpreting this input (Bamford, 1993; Paran, 1996). Students are left with the need to get as much comprehensible input as possible outside of the classroom.

3. Extensive Reading, Input & Acquisition

Krashen (1994) and others make a strong case for extensive reading as an effective and efficient path to obtaining input for acquisition. Ellis (1995) points out that moderate to low frequency words occur much more frequently in written texts than in common speech, thus offering greater opportunity for acquisition. The reader also has time, when needed, to form and confirm hypotheses about meaning and usage. Speech, on the other hand, may pass by too quickly for this to be done. Indeed, avid L1 readers typically display larger than average vocabularies (Ellis, 1995), and proficient L2 learners appear to acquire much of their vocabulary through reading (Krashen, 1989). Janopoulos (1986) found pleasure reading in English the variable correlating most strongly with English writing proficiency among ESL students, while in Tsang's (1996) study, time spent reading proved more helpful to learners' writing (language use and content) than time spent writing. Hafiz and Tudor (1989; 1990), in companion studies in ESL (England) and EFL (Pakistan) contexts, also recorded significant gains in writing proficiency (accuracy, fluency, range of expression) resulting from extensive reading, while Mason and Krashen (1996) reported that students in extensive reading based courses enjoyed greater relative gains in reading speed, writing proficiency, and performance on cloze tests than their counterparts in reading skills/grammar-translation based courses. Both Hafiz and Tudor and Mason and Krashen also observed positive effects on attitudes towards English among extensive readers. In line with this, Robb and Susser (1989), comparing extensive reading based and skills based reading curricula, saw extensive readers improve their reading skills at least as much as the control group, while reportedly enjoying the process much more. Gradman and Hanania (1991) found extensive reading for personal interest and enjoyment to be by far the strongest influence on scores on the TOEFL and its subsectionsincluding Listening Comprehensionand Elley (1991), reviewing a number of empirical studies, reported significantly greater gains in reading, writing, listening, and speaking skills among primary children involved in "book-flood" programs than ones receiving traditional audio-lingual instruction, particularly as assessment was extended over longer periods (one to three years).

While still in its infancy, research on extensive reading indicates that it can be a powerful aid in language acquisition.

4. Extensive Reading in English Communication Courses

A question meriting closer attention at this point is why a reading program should be instituted in a course specifically focusing on oral communication. Why not limit efforts to the targeted skills of speaking and listening?

Two misleading assumptions can underlie such a question. One, regarding production, is that one learns best by doinge..g., the path to becoming a better speaker of English is to speak as much as possible. This is true up to a point. Speaking practice can be a great help in solidifying language items in memory (Stevick, 1996) and in developing mechanical facility, confidence, and strategies for successful interaction (Ur, 1981). Feedback on attempts at speaking can alert learners to both problematic and successful aspects of their interlanguage (Pica, 1994; Schweers, 1995), and speaking allows them to elicit input from others (Krashen, 1988). In an EFL context like Japan, where there may be little opportunity for interaction in English outside of school, it seems logical to maximize communication in the classroom to realize these benefits. The act of speaking, in and of itself, however, does little to extend one's knowledge of vocabulary, grammar, usage, and discourse, which could then be brought to bear both in production and in making further input interpretable. For this comprehensible input is needed.

A second, related assumption regarding input is that oral communication skills and written communication skills are largely exclusive; therefore speaking and listening skills, for example, can only be facilitated by aural input. Certainly, listening through radio, cassette, TV, and/or video brings advantages for acquiring pronunciation and language features specific to spoken English, while simultaneously offering input and practice in a target skill. Extensive, out-of-class listening should be strongly encouraged and the author's classes did include an extensive listening component. However, as the findings reported by Elley (1991) and Gradman and Hanania (1991) suggest, there is no reason that written input, as well as aural input, cannot "contribute to a general competence that underlies both spoken and written performance" (Krashen and Terrell, 1983, p. 131). Candlin (1997) notes that "traditionally, speech and writing have been seen as distinct modes of communication. In recent years, however, there has been considerable research to indicate that rather than existing as separate modes there exists a discursive continuum between them, with shared features of literacy and oracy" (p. 1). Shared features indicate that knowledge of these features should aid competence in both modes. This may be especially true for novels and biographies containing extensive dialog. Gradman and Hanania (1991) also conclude that it is "likely that extensive outside reading helps to improve the level of proficiency in a global sense, enhancing acquisition of grammar, vocabulary, and rhetorical structure" (p. 45).

There seems no need, then, to focus on aural or written input to the exclusion of the other. Both can be encouraged and accessed as dictated by opportunity and individual learning style or preference.

On the practical side, two advantages of extensive reading are portability and simplicity; it is possible almost anywhere with no extra equipment. More important is the wide variety of titles, genres, and language levels easily available from ELT publishers (e.g., Oxford, Longman, Heinemann, Penguin) at reasonable cost (averaging around 600 per book). Commercially available tapes and videos, on the other hand, can be prohibitively expensive and materials graded to match a variety of levels can be difficult to find. In-house listening materials can be very time consuming for teachers to produce themselves; many hours of work went into producing one 90 minute cassette for each of the author's classes (hardly enough to be called "extensive"). Fortunately, companion tapes for graded readers are increasingly available, and these will become part of the author's program as funds allow.

A final, strong reason for maximizing high interest input in any form is its potential as a resource for communication in the classroom. As students spend much of their time in a communicative classroom interacting, they need something to communicate about. Common sources are the students' own thoughts, feelings, and experiences. Reading exposes students to new experiences and points of view that can then be shared and discussed. Extensive reading, therefore, not only contributes directly to acquisition, but aids skill development by providing content for meaningful interaction in the classroom.

5. Materials For Extensive Reading

5.1 Authentic vs. Simplified Materials

Given a conviction that extensive reading can be a valuable aid in facilitating acquisition, the next consideration is what type of materials to point students towardin particular, whether to use authentic materials written for the general market or simplified (graded) materials produced especially for language learners, as suggested above. Some educators discourage the use of non-authentic texts, claiming they are inferior as models of language and lack important cues for interpretation present in most authentic texts (Haverson, 1991). These objections may be valid, particularly for very advanced learners, but are probably outweighed for many Japanese university students by their need for enjoyable, comprehensible texts that provide practice in fluent reading and allow readers the sense of success and accomplishment that comes with understanding and finishing real, foreign language books (Bamford, 1984; Hafiz and Tudor, 1989; Hill and Thomas, 1988).

Authentic texts can be a useful tool in language classes, even at lower levels, when text and task type allow readers to be successful (e.g., scanning a TV guide for show times, reading a newspaper article for gist). Moreover, learners must develop skills and strategies for dealing with authentic texts when interest or career dictate. However, this type of work falls in the realm of intensive, task-based readingnot extensive reading. The time required to decipher texts that are incomprehensible without considerable dictionary work and multiple rereadings precludes the stated goal of massive amounts of input and limits its helpfulness in developing reading fluency (Bamford, 1993).

In some ways, moreover, criticism of simplified materials for extensive reading misses the point. The goal, of course, is that learners would be able to read anything that interest or need suggests, but towards that goal, graded reading materials can serve as a bridgeproviding comprehensible input, skills practice, and increased confidence leading to eventual fluent handling of authentic texts. In any extensive reading program, students must be encouraged to raise the level of their reading material as their developing abilities allow.

5.2 Interesting, Enjoyable Materials

Materials selected for extensive reading must also be interesting and enjoyable. One reason is motivational; the more students are interested in and enjoy the reading they do, the less it seems like work and the more reading they are likely to do. Narrative texts (e.g., novels, short stories, biographies) are inherently motivating in this way, engendering a desire to "see how it comes out" and compose the bulk of most extensive reading libraries. Program administrators must take care, however, to include a wide variety of genres to suit the varied interests of students.

Quantity of reading is not the only factor in input. Quality of attention to input is critical to what is actually taken in and the depth at which the input is processed. Gass and Selinker (1993) make a distinction between comprehensible and comprehended input. A text may be at an appropriate level for effective decoding and interpretation; it may be comprehensible. But if the reader does not actually comprehend the text, no input can be said to have been received or processed. Most readers have had the experience of coming to the end of a passage only to find that they had been visually following the text, but mentally had wandered to a completely different topic. A text may be eminently comprehensible, but this does little good if it is never read or simply glazed over. If engrossing for the reader, comprehensible input will be comprehended.

6. Program Implementation

In light of its potential for facilitating acquisition and serving as a resource for communication in the classroom, an extensive reading program was implemented in all of the authors' English Conversation (first and second year) and English Communication (third year) courses.

6.1 Obtaining Books

In order to select appropriate books, students' reading levels had to be determined. During the first class meeting, students read a handout containing photocopies of the first page from books at all six of Bamford's (1993) levels (Table 1) and reported which they felt comfortable reading, but with some degree of challenge. The author then ordered readers based on present levels and anticipated progress. This was a very rough estimation process, but the resulting distribution of books seemed to fit the needs of the students through the year.{1} Books ordered ranged from simplified versions of popular novels (e.g., Forrest Gump) and classics (e.g., Pride and Prejudice) to original works, biographies (e.g., Ghandi), and books on current issues. Vocabulary levels ranged from 300 basal words to unsimplified works. Available funds allowed for a ratio of just under two books to one student.

Over the course of the year, however, this ratio proved to be barely sufficient. Students were only able to borrow one or occasionally two books per week and still leave a reasonable selection in the classroom for others to choose from. Readers finishing books at midweek were often left with nothing to read, even though they had opportunity and desire. These missed chances for acquisition also lessened the probability of students developing a habit of free time reading in Englishthe best hope that this learning strategy would make a real, lasting difference in their eventual attainment in English, beyond the end of the term, or even beyond graduation. A book to student ratio approaching six to one would come closer to allowing optimal chances for acquisition, while retaining an attractive array of titles for display in the classroom.

Table 1. Graded reader series arranged by level (after Bamford, 1993, p. 71)

[Editor's Note: This table has been omitted. It has been superceded by the David Hill chart available on this site.]

6.2 Processing Books

After books were received, the front cover of each was marked with a round, colored seal indicating reading level (green = level 1 (easiest), blue = 2, yellow = 3, orange = 4, red = 5, pink = 6 (unsimplified)). The number of reading pages (considering page size and deducting for space taken up by illustrations, exercises and titles) was then quickly estimated and recorded on the colored seal.{2}

6.3 Program Administration

In each subsequent class session, books were set out on tables, and the last five minutes of class were reserved for book selection. (These periods also proved opportunity for personal chats with students and wrapping up class business in a less hurried way.) Students received reading points for each book based on their year in school, the book's level, and its number of reading pages (Table 2). Reading targets for the year were set depending on year in school: 280 points for first year students, 360 for second, and 480 for third.{3} Targets yielded weekly goals of around 15, 23, and 30 points per week, respectively.

Students were informed that although high level books brought more points per page, reading at a comfortable level would be faster, yielding more reading points and greater enjoyment. On the other hand, reading needlessly simple books would likely cut enjoyment without yielding significantly more points than more challenging, but still comfortable, material. It was hoped that this system would encourage readers to find a good level and to raise this level as their growing abilities allowed.

Students kept track of their own reading on Personal Reading Records (Table 3). Their ratings of each book (G=good, F=fair, P=poor) were valuable in reordering books for the following year.{4} Reading times helped students and instructor track how much (or little) time they had actually invested. Difficulty ratings also helped students gauge appropriate reading levels.

Initially, a check-out system was used in which students recorded the books they borrowed and check-out and return dates. The extra time this consumed, however, seemed counterproductive, and it was dropped, allowing students to freely borrow and return books. To date, book loss does not appear to have been excessive, but an inventory at the end of the current school year will reveal whether this honor system is sustainable.

6.4 Instant Book Reports

For each book read, students completed an Instant Book Report form (Fig. 1). These were termed instant reports because students were encouraged to spend no more than 15 minutes per report before going on to start their next bookthe emphasis being on input rather than output.

Reports consisted chiefly of a two to three sentence summary and a three to four sentence free personal response. The summary gave students an opportunity to review the story mentally and demonstrate general understanding, while getting practice in giving a brief synopsis. The free response section called for students to go beyond simple comprehension and employ this understanding in further analysis or reflection. Responses ranged from judgements of a story's strengths or weaknesses, to thoughts on how themes arising in a book affected the student personally, to memories elicited by a

particular scene, to questions regarding cultural or historical background. Particularly for some students hesitant to speak up in class, responses gave them a chance to display rich thought lives and real expressive ability in English that the instructor might otherwise rarely have seen.

6.5 Evaluation

Performance in the reading program was a part of each student's final course grade. Chief factors were meeting the minimum reading target and completion and quality of Instant Book Reports. Extra credit was given for every 50 points gained beyond the target. One examination each term also consisted of a one-to-one, teacher-student oral interview regarding a book the student had read.

7. Reading As A Resource For Classroom Communication

7.1 Pre-reading Activities

As students began selecting books, they went through a series of activities intended to stimulate interest in reading, put students into physical contact with the books, and generate classroom interaction:

1. Students were given a "Reading and You" questionnaire eliciting information on reading habits and experiences, as well as attitudes toward reading for pleasure and as a path to language learning. After completing it at home, students shared their responses in small groups.

2. The first day books were brought into class, students were introduced to English vocabulary for a variety of genres in literature. In this "Titles & Genres" activity, pairs then scanned through five bookslooking at titles, illustrations, chapter headings, and blurbs (short introductions on the back cover)to determine their genres. After returning all books to the front of the room, students formed new groups of three and took turns describing the books they had just examined (no titles!) while their partners tried to locate the books in question. Finally, several students introduced interesting books they had found to the rest of the class.

3. For "Blurb & Title Match," students worked in pairs. Partner A and partner B each received a different set of blurbs taken from books in the class library, along with identical title lists. While A read aloud a blurb from his list, B searched for a likely title. After agreeing on a title, B read another blurb to A, and so on. Through this listening activity students had opportunity to identify interesting books, and, since blurbs are written at the same language level as the books, to further gauge their appropriate reading levels.

7.2 Reporting on Reading

After the program was underway, periodic in-class activities allowed students to discuss the books they had been reading. These provided fluency practice and the opportunity to bring language and content from books into play in the classroom. Students generally appeared to enjoy sharing stories and opinions of the books and many recorded in their class diaries that they had picked up good book recommendations.

1. For "The Best Book I've Read," students worked with partners, and using their completed Instant Book Reports for support, discussed their favorite books so far (e.g., "What's it about? What did you like best about it? How did it end?"). After an appropriate time interval, new pairs were formed, and the activity was repeated, but this time with no access to written reports. Students were pleased to find that they were able to function without support and many could interact more freely when not tied to the written page. After repeating this process for a third time, several students shared interesting books they had learned of that day.

2. "Book Review Round Robin" functioned much as the previous activity, but was based on short book reviews (positive or negative) written at home.

3. In "What's Happening Here?" each student made a drawing of a scene from a book. Groups of three then used these drawings as starting points for talking about the stories, e.g., "What's happening here? Is this scene near the beginning or end of the story? What happened before this? Why did you choose this scene?"

4. In "Telling the Truth About Extensive Reading" (Dubravcic, 1996), students worked in teams of three, each sharing brief book summaries before settling on a title of interest that had only been read by one member. The two students who had not read the book then had ten minutes to elicit as much information as possible about plot, characters, setting, and so on. Facing another team, all three members then claimed to have actually read the book, and members of the facing team questioned them to determine who had, in fact, read it. Points were earned for correct guesses. After reversing roles and repeating the process, each group went on to face another team.

5. For other helpful book-based communication activities see Helgesen (1997), Day (1993), and Day and Bamford (in press).

8. Student Response

Any analysis of an acquisition oriented program begs the question of what effects the program had on the participants' acquisition. In the first year, however, efforts were concentrated on getting the program up and running successfully rather than implementing a controlled research scheme. It is hoped that this will be practical in the near future. In the meantime, student response to the program yielded valuable information.

8.1 Reading Quantity

Student response as expressed in reading quantity is summarized in Table 4. Average reading points for all year levels exceeded the minimum targets; 7280% of students met them, and 3045% exceeded them by at least 50 points. Average weekly reading times were encouraging, especially for third year students, who averaged over two hours of English reading per week.

As a main focus of the reading program was acquisition, however, the value of these numbers is largely dependent on judgements of the quantity of input needed to yield significant acquisition. Little research has been done, but Nation (1997) makes some rough estimates for vocabulary acquisition. He reasons that the effects of repeated exposure to a word will only be cumulative if the word is met again before the previous meeting is forgotten. Assuming a retention period of one week, a typical second year student reading books at levels three to four (1,1001,700 basic words, 250 words per page) would need to read around 40 pages a week to get sufficient exposure to any new vocabularywhich typically occur once every 10,00015,000 running words at this level. Accepting Nation's number, second year students would need to increase their reading dramatically from the 23 pages a week targeted here. It should be noted that Nation's figure is based on word frequencies in unsimplified texts; frequencies in simplified texts should be greater, thus requiring less reading for the needed repetitions. This also says nothing about acquisition of many grammatical and rhetorical features which may occur with greater frequency than some individual words.

A promising direction for future work would be to obtain average word frequencies for readers at each level and use these to set appropriate reading goals.

8.2 Exit Questionnaire

At year end, second and third year students were asked to fill out anonymous questionnaires, and responses to relevant items were helpful in evaluating the extensive reading program. Because the questionnaire was offered only in English and had not been put through any piloting procedure, conclusions can only be considered preliminary. Some general observations, however, can be made.

8.2.1 Questionnaire Enjoyment and Helpfulness

Krashen (1993) encourages free reading with minimal accountability and few, if any, extrinsic motivators as the route to maximum enjoyment, motivation, and thus, acquisition. Although, in this program, minimum reading targets were imposed along with mandatory book reports, 79% of 59 respondents reported that the reading program was "enjoyable," with 9% "so-so," 3% "not very enjoyable," and 9% "not enjoyable." Asked why they enjoyed the reading, most simply responded that they had read interesting or enjoyable books, or that the variety of books available had allowed them to select books of interest. Most students responding negatively to the program cited a lack of time.

Eighty-eight percent of 61 respondents reported that the program had been "helpful" ("very helpful"62%, "helpful"26%), with 8% "somewhat helpful," and 6.6% "not very helpful." Asked why it was helpful, most mentioned improvements in reading speed, comprehension, and/or increased vocabulary. Some commented on contentthat they had learned new things about places, cultures, or points of view. Surprisingly, eight students (13%) reported that the book reports were the most useful aspect of the program because they provided good writing practice and/or were thought provoking

These responses indicate that the requirements of the program did not prevent the great majority of students from enjoying their reading or finding it helpful. In fact, it was assumed (but in no way proved) that these requirements were necessary to ensuring that substantial amounts of reading actually took place. One direction for future research would be to test what, if any, effects increasing or doing away with minimum reading targets and post-reading tasks would have on reading quantities and student attitudes.

8.2.2 Questionnaire -- Favorite and Least Favorite Books

Another item on the questionnaire asked students to give the titles of their favorite and least favorite books and reasons for liking or disliking these. For favorite books, 82% of 84 respondents gave content related reasons (e.g., "interesting," "enjoyable," "like the genre," "learned a lot") with 12% offering language- or difficulty-related factors ("easy to read", "short stories enjoyable", "long book challenging"). For least favorite books, conversely, only 28% of 60 respondents gave content related reasons (e.g., "uninteresting," "dislike story/characters"), while 63% noted that books were too difficult to understand or too long, and 8% identified books as too easy or too short.

These responses suggest that, as expected, learners reading at appropriate levels were able to successfully access content and make judgements based on content related issues. On the other hand, books that were too linguistically difficult or not sufficiently controlled for content (too many events packed into each page){6} (Hill and Thomas, 1988) appear to have frequently blocked understanding enough to rule out enjoyment. The positive side of this is, as book reports also revealed, that many students were periodically trying to step up, testing the limits of their competencea process critical to continued progress.

8.3 Free Response to Reading

Probably the most intimate look at response to the program was gained through the free response section of Instant Book Reports. Students frequently offered fascinating pictures of the pleasures and pains they experienced through extended readingshowing that more than simply studying language, many had been living, for hours each week, in what one student described as "an only English world." Comprehensible input was, indeed, being comprehended, a hopeful indication for acquisition.

Most of the following examples were drawn from the work of second year students.

9. Conclusion

Research indicates that large amounts of comprehensible input are necessary for significant language acquisition and that extensive reading with graded readers can be an enjoyable, motivating source of such input, contributing to a general competence that supports both spoken and written performance. As a source of plentiful out-of-class input and a resource for interaction in the classroom, an extensive reading program can be a useful component of a course focusing on oral communication. The large majority of students in the author's first through third year English conversation and communication courses met or exceeded reading targets (although word frequency data indicate that targets may need to be raised considerably), reported enjoying the program and finding it helpful to their progress in English, and participated enthusiastically in reading related, in-class activities. Although too-difficult selections sometimes hindered readers' understanding and enjoyment, this indicated that they were continuing to explore the upper limits of their competence, a process necessary to progress. Finally, free written responses to reading revealed a level of engagement likely to be favorable for acquisition.

Future research possibilities in association with the reading program include (a) determining what effect varying levels of accountability (reading targets, book reports) have on student attitudes and performance, (b) more accurately determining word frequencies in graded readers and using these to establish appropriate reading goals, and (c) testing the effects of varying reading goals on vocabulary acquisition.

Acknowledgments

The author would like to thank Marc Helgesen, Julian Bamford, and David Hill for their advice and encouragement, and most of all, his students with whom this work was done.

References

Bamford, J. "Beyond grammar translation: Teaching students to really read." A Handbook for Teaching English at Japanese Colleges and Universities. Ed. P. Wadden. New York: Oxford University Press, 1993, 63-72.

________. "Extensive reading by means of graded readers." Reading in a Foreign Language 2.2 (1984): 218-260.

Brown, H.D. Teaching by Principles: An Interactive Approach to Language Pedagogy. Englewood Cliffs, NH: Prentice Hall Regents, 1994.

Candlin, C. "Discourses of speech and writing: Interdiscursivity and interculturality." The JACET 36th Annual Convention. Tokyo: JACET, 1997.

Dawson, N, ed. The Longman Guide to Graded Reading. Essex, England: Longman, 1992.

Day, R., ed. New Ways in Teaching Reading. Alexandria, VA: TESOL, 1993.

Day, R., and J. Bamford. Extensive Reading in the Second Language Classroom. Cambridge: Cambridge University Press, in press.

Dubravcic, J. "Telling the truth about extensive reading." The Language Teacher 20.12 (1996): 52.

Elley, W. "Acquiring literacy in a second language: The effect of book-based programs." Language Learning 41 (1991): 375-411.

Ellis, N. "Vocabulary acquisition: Psychological perspectives and pedagogical implications." The Language Teacher 19.2 (1995): 12-16.

Ellis, R. Instructed Second Language Acquisition. Oxford: Basil Blackwell Ltd., 1990.

Gass, S., and Selinker, L. Second Language Acquisition: An Introductory Course. Hillsdale, NJ: Lawrence Erlbaum Associates, 1993.

Gradman, H., and Hanania, E. "Language learning background factors and ESL proficiency." Modern Language Journal 75.1 (1991): 39-51.

Hafiz, F., and Tudor, I. "Extensive reading and the development of language skills." ELT Journal 43.1 (1989): 413.

________. "Graded readers as an input medium in L2 learning." System 18 (1990): 31-42.

Haverson, W. "Adult literacy training." Teaching English as a Second or Foreign Language. Ed. M. Celce-Murcia. New York: Newbury House, 1991, 185-194.

Helgesen, M. "Bringing those books back to the classroom: Tasks for extensive reading." The Language Teacher 21.5 (1997): 53-54.

Hill, D. "Setting up an extensive reading program: Practical tips." The Language Teacher 21.5 (1997): 17-20.

Hill, D., and Reid Thomas, H. "Survey review: Graded readers (part 1)." ELT Journal 42.1 (1988): 44-52.

Janopoulos, M. "The relationship of pleasure reading and second language writing proficiency." TESOL Quarterly 20.4 (1986): 763-768.

Krashen, S. The Power of Reading: Insights from the Research. Englewood, CO: Libraries Unlimited, 1993.

________. "The power of reading: Plenary session 10." TESOL'94 Baltimore: TESOL, 1994.

________. "Providing input for acquisition." Making it Happen: Interaction in the Second Language Classroom from Theory to Practice. Ed. Patricia A. Richard-Amato. White Plains, N.Y: Longman, 1988, 330-342.

________. "We acquire vocabulary and spelling by reading: Additional evidence for the Input Hypothesis." The Modern Language Journal 73.4 (1989): 440-464.

Krashen, S., and Terrell, T. The Natural Approach. Oxford: Pergamon/Alemany, 1983.

Lightbown, P., and Spada, N. How Languages are Learned. Oxford: Oxford University Press, 1993.

Long, M. "Group work and communicative competence in the ESOL classroom." Developing Communicative Competence in a Second Language. Eds. R. Scarcella, E. Anderson and S. Krashen. New York: Newbury House, 1990, 303-315.

Mason, B., and Krashen, K. "Extensive reading in English as a foreign language." System 25.1 (1996): 91-102.

Nation, P. "The language learning benefits of extensive reading." The Language Teacher 21.5 (1997): 13-16.

Paran, A. "Reading in EFL: Facts and fictions." ELT Journal 50.1 (1996): 25-34.

Pica, T. "Research on negotiation: What does it reveal about second-language learning conditions, processes, and outcomes?" Language Learning 44.3 (1994): 493-527.

Pica, T., Young, R., and Doughty, C. "The impact of interaction on comprehension." TESOL Quarterly 21.4 (1987): 737-758.

Robb, T., and Susser, B. "Extensive reading vs. skills building in an EFL context." Reading in a Foreign Language 5.2 (1989): 239-251.

Schmitt, N. "A brief perspective on vocabulary." The Language Teacher 19.2 (1995): 34-35.

Schweers, C."Negotiated interaction, transfer, and the second language classroom." TESL Reporter 28.1 (1995): 1-14.

Stevick, E. Memory, Meaning & Method: A View of Language Teaching. 2nd ed. Boston: Heinle & Heinle, 1996.

Swain, M. "Communicative competence: Some roles of comprehensible input and comprehensible output in its development." Input in Second Language Acquisition. Eds. S. Gass and C. Madden. Rowley, MA: Newbury House, 1985, 235-253.

Tsang, W. "Comparing the effects of reading and writing on writing performance." Applied Linguistics 17.2 (1996): 210-233.

Ur, P. Discussions that Work. Cambridge: Cambridge University Press, 1981.

Ying, H. "What sort of input is needed for intake?" IRAL 33.3 (1995): 175-194.

Notes

{1} An alternate route to materials selection is through the Edinburgh Project for Extensive Reading (EPER) (Hill, 1997). They provide placement/progress tests, and based on the results, can suggest appropriate levels and quantities of readers and/or provide a list of recommended titles. For information, contact David HillFax: 44-131-667-5927, e-mail: .

{2} True reading lengths could be simply determined if publishers included a word count with each book. Requests to major publishers for word counts have met with little success, but rough word counts for many titles are available from EPER (see note1).

{3} Program startup delays held reading targets approximately 100 points lower than they would have otherwise been.

{4} These ratings were also entered inside the back cover of each book, providing a convenient reference for students.

{5} One third year evening class was not included. Many students worked full time, and their average of 390 reading points was not consistent with the two day sessions.

{6} In future questionnaires a clearer distinction between linguistic difficulty and content control as barriers to understanding will be made.