March 2014
The Role of Reading Aloud in EFL Writing Revision
Theresa Jinling Tseng
Abstract
This study examines what changes EFL writers at three different proficiency levels can make when they read aloud their writing, and how they perceive the effect of reading aloud on their writing revision. Twenty-eight English-major sophomores of a university in Taiwan were given two writing assignments. After students completed their first drafts of each assignment, they read their writing aloud to the instructor while the instructor also marked on another copy of their writing. Then, the students were given questionnaires and interviews. The study found that reading aloud (RA) helped EFL students at various levels to self-detect mostly local problems. The advanced writers benefited from it the most. However, prepositions and word choice, whose correct usage depends much on native-speakers’ competence, appeared to be challenging to students regardless of proficiency levels. In addition, occasionally RA triggered undesirable or unsuccessful corrections, and they happened more frequently when the beginning writers read aloud. Meanwhile, RA offered limited help in detecting global problems. The interview results revealed that most students felt relaxed and not distracted by their pronunciation while reading aloud, and that most of them found it a useful revision strategy. The results suggest that RA is not a panacea, yet EFL writers can still benefit from it, especially in making local changes.
Keywords: EFL writers, revision, reading aloud, L2 competence, L2 proficiency, L2 writing
Assistant Professor, Tunghai University, English Language Center.
䫔ḅ㛇炻枩 221-252 Ḵ˕ᶨ⚃⸜ᶱ㚰
р
рᖂ᠐ዺӧम/ѦᇟቪբঅׯύޑفՅ
මዠ࣓
ᄔा
ҁࣴزϐҞޑӧᔠຎрᖂ᠐ዺૈᔅշΟᅿૈΚભޑम/ѦᇟቪբᏢғϙሶ অׯǴϷдॺჹрᖂ᠐ዺ೭ᅿঅׯמѯޑຑሽǶѠࢌεᏢमЎس28ঁεΒᏢғୖ ᆶࣴزǶӅٿԛǴԛӚቪጇբЎǶԛቪֹ߃ዺࡕǴঁᏢғᆶԴৣय़Ǵ җᏢғ᠐ዺ٠অׯǴԴৣӕਔӧќҽᏢғ߃ዺঅׯǶӧಃΒԛჹ᠐ዺঅׯ ࡕǴঁᏢғ༤ቪୢڔǴ٠ௗڙԴৣೖፋǶࣴز่݀ᡉҢрᖂ᠐ዺૈᔅշ܌Ԗόӕ ૈΚભޑᏢғֽȐlocalȑঅׯǴՠჹѦᇟૈΚനଯޑᏢғനԖਏǶрᖂ᠐ዺ ૈЇଆᏢғޑεጄൎȐglobalȑঅׯߚதϿǴόᆅૈΚଯեӵԜǶӕኬǴ܌ԖᏢ ғǴόᆅૈΚଯեǴჹϟسຒǴᒧӷ೭٤Бय़ޑᒱᇤΨࡐᜤҗᏢғрᖂ᠐ዺ ԾՉวǴӢࣁ҅ዴҔݤததѸᇥमᇟࣁ҆ᇟޑΓޑޔΚǶԜѦ, Ԗਔрᖂ ᠐ዺଽԶᡣԖ٤ᏢғȐЀځૈΚၨեޑȑעჹޑׯԋᒱޑǶೖፋ่݀ᡉҢεϩ ޑᏢғӧрᖂ᠐ዺਔόᆙǴΨόдॺԾρޑวॣυᘋǴԶЪεϩޑᏢғ ளрᖂ᠐ዺࢂԖҔޑբЎঅׯמѯǶᕴԶقϐǴࣴز่݀ᡉҢрᖂ᠐ዺόࢂᡫϏǴ ՠࢂჹम/ѦᇟᏢғբЎঅׯǴЀځࢂֽঅׯǴϝฅԖۓޑӳೀǶ ᜢᗖຒǺम/Ѧᇟቪբޣǵቪբঅׯǵрᖂ᠐ዺǵΒᇟૈΚǵΒᇟዕግำࡋǵΒᇟቪ բǶ 㜙㴟⣏⬠劙婆ᷕ⽫≑䎮㔁㌰ˤ ⇘䧧㖍㛇烉2013 ⸜ 12 㚰 31 㖍ˤIntroduction
Revision has been well-recognized as a crucial component in the development of L1 and L2 writing skills (Anson, 1989; Freedman, 1987; Hedgcock & Lefkowitz, 1992; Hillocks, 1986; Jenkins, 1987; Smith, 1982; Zamel, 1985). One of the most commonly used revision strategies is for students to read their writing aloud during the reviewing process. The value of reading aloud (RA) has been confirmed by many L1 writing experts, stating that when most L1 students read their writing aloud, they will find rhetorical problems and correct in essence all errors of spelling, grammar, punctuation (Bartholomae, 1980; Hartwell, 1997).
In the 1980s, some L2 writing experts suggested that there was much to learn from the research and pedagogy in the teaching of writing to native English speakers (see Zamel, 1976, 1982, 1983, 1985, 1987, 1988; Kroll, 1978; Raimes, 1983a; 1983b; 1986; 1991). Consequently, L1 writing pedagogy and revision strategies, including RA, were introduced to L2 writing instruction. Since then, RA has also been recommended by authors of L2 writing textbooks and L2 writing teacher-educators in training future L2 writing teachers (e.g., Ferris, 2007, 2008). However, learning to write in L2 is not the same as learning to write in L1. While L1 writers learn to write the language they already speak, L2 writers learn to write the language they are not completely familiar with. One question worth asking is whether RA is equally effective in helping L2 writers.
A literature review reveals that an abundant of research has focused on reading aloud as a reading strategy for both L1 and L2 learners (see, e.g., Deng & Huang, 2011; Gibson, 2008; He, 2009; Morrison & Wlodarczyk, 2009; Santoro, Chard, Howard, & Baker, 2008; Xie, 1998, 2000), yet fairly few studies have investigated reading aloud as a revision strategy for L2 writers. Moreover, in these few studies, researchers have different views on the effectiveness of RA when it is applied to L2 writing revision. Although some researchers have provided evidence to show positive effects of RA on student revision, they have not indicated if students at different levels of L2 proficiency can benefit equally from RA. On the other hand, some researchers have speculated that RA may even cause negative consequences, yet evidence is needed to verify their claims (see the Literature Review section for more details). In other words, the effectiveness of RA has not been clear. This study is a preliminary study that explores to what extent RA can help EFL writers become self-sufficient editors, just as it can help L1 writers.
The following sub-questions are designed to find out how RA may help EFL writers in their revising process:
1. Can RA trigger EFL students (L1=Chinese) to make self-initiated revision? If so, what types of self-initiated changes are EFL students more likely to make? 2. What types of problems are difficult for EFL students to self-detect even after
reading their writing aloud?
3. Which level of students can benefit most from RA? Does RA help one level of students more with global revisions and another level more with local?
The teaching of second language writing sits at the junction of two disciplines--composition and rhetoric as well as language learning (Carson, 2001; Kroll, 2003; Silva & Matsuda, 2001). The importance of understanding the relationship between the two disciplines has been evident in the themes of recent second language writing conferences (e.g., 2010 Symposium on Second Language Writing) as well as a gradual increase of studies that focus on the connection between SLA theories and second language writing instruction. For example, Carson (2001) and Williams (2005) have attempted to explain the relationship between SLA theories and L2 writing (or L2 writing instruction). In addition, empirical studies on the interface of the two disciplines have also started to emerge. The majority of them, however, has focused on the relationship between teachers’ written feedback (or instruction) and student writers’ noticing (e.g., Ma, 1997; Qi & Lapkin, 2002; Swain & Lapkin, 2002; Chandler, 2003; Santos, Lopez-Serrano, & Manchon, 2010; Barnawi, 2010). This study represents an initial attempt to investigate the relationship between RA, a revision strategy, and EFL learners’ interlanguage system, which is evidence of their second language learning. More specifically, it provides empirical evidence of EFL writers’ self-initiated revision for the purpose of modifying their interlanguage production. Thus, this study serves to add one more building block to the recent endeavors to strengthen the interface between L2 writing research and second language acquisition.
The findings of the study have pedagogical significance. Firstly, the findings can inform writing teachers what problems EFL students at various proficiency levels are more (or less) likely to self-detect after reading their writing aloud. Secondly, the findings can increase teachers’ awareness of EFL students’ perceptions of the use of RA. With the understanding of the role RA plays in EFL writing revision, writing teachers can make informed decision in designing lessons that best meet EFL writers’ needs.
Literature Review
RA as a Recommended Revision Strategy
Reading aloud is a revising strategy recommended by many writing experts. Murray (1988) argues writing does not exist until it is read, for the reading during writing helps the writer monitor, an act that involves awareness on many levels and includes the opportunity for change. Authors of writing textbooks have also recommended reading aloud. For instance, in his 8th edition of College Writing Skills, Langan (2011) recommends that students read their writing aloud as they revise: “Hearing how your writing sounds will help you pick up problems with meaning as well as with style” (p. 33). Likewise, in her 11th and also an international edition (for EFL writers) of Steps to
Writing Well, Wyrick (2011) argues that “reading aloud is a good technique because it
allows your ears to hear ungrammatical ‘chunk’ or unintended gaps in sense or sound you may otherwise miss. Reading aloud also flags omitted words” (p. 108). In her Developing Composition Skills, Ruetten (1997) advises ESL writers to get a classmate to read their paragraphs aloud and give them feedback. Lange (2011) presents an interview with author
Hannah Pittard, who states that nothing she has ever published has not been read aloud many times by herself, and that her advice to aspiring writers is to read everything aloud. Reading aloud, as Pittard stresses, helps her revise sentences for rhythm. Based on their own writing experiences, scholars such as Hughes (2003), Sousa (2006), and Griffin (2011) testify that reading aloud can be a key to effective revision. As Hughes (2003) states, “reading aloud helps us spot flaws and errors, . . . [find] holes in logic or construction, . . . [and detect] word problems” (p. 18).Thus, RA has been recommended as an effective strategy for revision because it helps writers revise not only surface errors but also rhetorical aspects such as meaning, style, and logic.
Reading aloud is also one of the typical practices found in writing center sessions. The ubiquity of this practice is evident in popular tutor training manuals, which frequently recommend the writer to read aloud: “Have the writer read the draft aloud” (Caposella, 1998, p. 10); “We recommend that you ask the writer to read the paper aloud to you while you take notes” (Gillespie & Lerner, 2004, p.30). One reason why these manuals encourage writers to read their own work concerns writer control. Caposella (1998), for example, argues that having “the writer read the draft aloud” is one of the “ways to encourage a high level of involvement from the writer” (p.10). Ryan and Zimmerelli (2006) also state that “asking writers to read aloud engages them more in the tutoring session” (p.46). Furthermore, RA may prompt self-initiated revision: “Reading his own work, . . . helps the writer hear passages that ‘sound funny’ even though they may look fine on paper” (Caposella, 1998, pp. 10-11). Finally, RA may help the writer become more aware of the audience: “When the writer reads the paper, . . . he [will notice] that the reader is listening . . . to the way the draft is working. . . he’s giving his draft a critical reading in ways that will help him revise” (Gillespie & Lerner, 2004, p. 30). Therefore, writer control, self-initiated revision, and audience awareness are at least three major reasons why RA is recommended by writing tutor training manuals.
Some Theoretical Bases of RA
L1 Competence. One obvious reason for L1 writers to read their writing aloud during the revision process is to check if the writing “sounds right.” The judgment is based on the writers’ competence, or the intuitive knowledge about the system of their native language. According to Chomsky (1965), native speakers’ competence develops naturally, and they can rely on it to judge whether the speech or writing is grammatical. Based on Chomsky’s notion of competence, Hartwell (1997) argues that “native speakers of English, regardless of dialect, show tacit mastery of the conventions of Standard English” (p. 202). This mastery of literacy, as Hartwell continues, “seems to transfer into abstract orthographic knowledge through interaction with print” (p. 202), and regardless of instruction, the mastery is “from the top down, from pragmatic questions of voice, tone, audience, register and rhetorical strategy” to errors of spelling, grammar and. punctuation (p. 202). Hence, Hartwell (1997) suggests that reading aloud allows students to tap their unconscious or intuitive knowledge of their native language to check not only the mechanical but also the rhetorical aspects of their writing.
Comprehensible Output and Noticing. Comprehensible output and its related
notion noticing are two reasons why RA may also help L2 writers. According to Swain’s (1985, 1993) output hypothesis, comprehensible output is a necessary mechanism of language acquisition. The act of writing is a type of output. Based on Swain’s (1985, 1993) hypothesis, output may prompt the learners/writers to consciously recognize some of their linguistic problems. The act of reading aloud is vocalizing the writers’ written output, which may further increase students’ consciousness of what needs to be changed. Schmidt (1990) describes the consciousness raising effect as noticing. In Schmidt’s view, noticing is an essential element that accelerates the learner’s language learning process. The importance of noticing for language learning can be found in assertions such as “those who notice most, learn most” (Schmidt & Frota, 1986, p. 313) and “no noticing, no acquisition” (Ellis, 1995, p. 89). In short, RA may be beneficial to L2 students because it serves as a route to noticing.
Monitoring. Another reason why RA may help L2 writers is that reading aloud allows them the time they need to monitor their language production. Krashen’s monitor hypothesis (1982) predicts that under three circumstances L2 learners may call upon the learned knowledge to correct themselves when they communicate: (1) There must be sufficient time for the learner to apply a learned rule, (2) the learner must be focused on the form of the output, and (3), the learner must know the rules. Time is an essential element for self-correction to take place. Cohen, Larsen-Freeman, & Tarone (1991) also argue that when students are able to spot their errors, they are able to correct most of the errors, but the premise is that the learners need to have time to self-correct. Compared with reading silently, the speed of RA is relatively slower because writers have to vocalize the written words. The slower speed of RA offers the time that L2 writers need to focus on form and recall the rules learned.
The Effects of RA on L1 Writers
Most L1 writing teachers and researchers have recognized that RA has value in improving the quality of students’ writing. In her case study of two L1 students’ revising strategies, Selfe (1984) found that RA was a revision strategy often adopted by one student, who was not at all apprehensive about writing, but not by the other, who was an apprehensive writer. It was likely that the confidence of the non-apprehensive writer came from his ability to take advantage of reading aloud as a revision strategy.
Evans (1988) studied a remedial L1 writer whose tutor read his compositions back to him during the reviewing sessions. While hearing the sounds of his writing, the writer was able to not only attend to syntactic and morphological defects in his text, but also find the passages which inadequately expressed his intentions. Thus, hearing the sound of one’s writing may cause the writer to re-check if the writing matches his or her internal voice or intended meaning.
Smith and Cosmic (1995) argue that RA may improve writing ability because it helps students learn to “hear” the sounds of the material written. Similarly, Gibson (2008)
indicates that RA has several benefits, two of which are related to writing. First, it can help develop writing skills by using it as oral proofreading. Gibson explains that RA makes students see the graphemic-phonemic correspondences; consequently, errors might be more easily picked up with visual and aural inputs working together. Second, RA may help writers listen to their “internal voice” as they write. While writing has no intonation, stress, or pauses (apart from those that punctuation and italicization can indicate), it is very likely that both readers and writers do assign these features to whatever they are reading and writing (Tench, 1996, cited by Gibson, 2008). Therefore, Gibson (2008) argues, intonation can affect what is written. When students write, they will need to organize the structure clearly using discourse markers, and to reinforce this with the appropriate intonation when they read aloud to themselves or friends so that the message will be understood. This process might involve students in listening to their “internal voice” as they write.
The Effects of RA on L2 Writing Revision
To date the literature on RA as an L2 revision strategy hardly exists. Some L2 writing researchers, however, have confirmed the value of reading aloud as a revision strategy, although RA was not the focus of their studies. While studying advanced ESL students’ composing processes, Zamel (1983) found all of the students reread their drafts aloud in order to evaluate whether the form matched their intent and to discover ideas that were to be recorded. She argued that reading aloud helped the students construct meaning and assess it at the same time. Zamel (1983) also noticed that RA seemed to be able to help students overcome writer’s block. When her students claimed to be “stuck,” after rereading their texts aloud several times, they figured out how to continue. Moreover, RA appeared to trigger global revision, whereby sentences were rewritten to express the writer’s intent, paragraphs or parts of paragraphs were shifted around, and new paragraphs were formed as thoughts were developed and expanded. “Very often these revisions would occur after rereading the text aloud, as if hearing it spoken meant ‘seeing’ it in a new and more removed way,” stated Zamel (1983, p. 174). Zamel (1983) further pointed out that more skilled writers seemed to be more aware of the recursive nature of writing, and they revised by rereading aloud larger parts of their writing, e.g., whole paragraphs rather than one sentence or two. Finally, Zamel’s students expressed that rereading their writing aloud helped them find errors and spelling problems.
Another confirmation of the value of RA to L2 writers comes from Hedgcock and Lefkowitz’s (1992) study of collaborative oral/aural revision in foreign language writing instruction (L1 = English). They found that students in the experimental group, who received feedback after reading aloud their texts in peer review sessions, produced better final drafts than the students in the control group, who received the teacher’s written feedback only. Hedgcock and Lefkowitz noticed that reading aloud to peers produces in students an awareness of the rhetorical structure of their own writing and an ability to self-correct surface errors. Hedgcock and Lefkowitz further indicated that reading aloud one’s text in a group of attentive peers fosters an awareness of audience, which
consequently enables writers “to identify and correct inconsistencies between what they have written and what they actually intended to write” (Hedgcock & Lefkowitz, 1992, p. 258).
Hedgcock and Lefkowitz (1992) have argued that a read-aloud revision procedure may be a particularly beneficial technique for L2 writers for two reasons: (1) thinking about language use as well as the ideas writers attempt to convey is central in L2 composing (Cumming, 1990), and (2) the vocalization aspect of the RA may “facilitate revision to an even greater extent, inducing less-proficient writers to capitalize on the ‘running accompaniment’ described by Vygotsky (1962, 1978)” (Hedgcock & Lefkowitz, 1992, p. 259).
Some Reservations about RA
Although Zamel, Lefkowitz and Hedgcock have attested to the usefulness of RA to L2 writers, some other researchers have cast doubt on the efficacy of this revision strategy. Powers (1993) pointed out that a couple of presumptions of adopting RA in L2 revision are that the writers hear the language correctly and are more familiar and comfortable with the oral than the written word. Yet Powers (1993) argued that those presumptions are not valid for ESL writers for two reasons. First, few beginning second language writers “hear” the language correctly; they do not have the “inner editor to prompt them to stop and raise questions about their writing” (Powers, 1993, p.43). Second, many ESL writers are more familiar with written than spoken English, as their L2 learning experience in their countries is usually more focused on reading and writing rather than speaking. Hence, some L2 writers may not feel more comfortable with reading their writing aloud than reading it silently.
Some writing researchers (e.g. Capossela, 1998; Cogie, 1999; Harris & Silva, 1993; Matsuda & Cox, 2009; Reid, 1997; Silva, 1993/2001; Williams, 2005; Williams & Severino, 2004) also believe that RA may not work well for ESL writers because of their less developed sense of what “sounds right.” Silva (1993/2001) noticed that L2 writers revised with more difficulty (than L1 writers) and that there was less “revising by ear” or making changes on the basis of what “sounds” good (p. 195). Matsuda & Cox (2009) argue that L2 writers who have not developed that intuitive sense of the English language, or those who have not internalized certain rules of the English grammar, may not benefit from reading aloud for revision. Reading aloud, in Matsuda and Cox’s view (2009), may even shift the writers’ attention to the pronunciation of their writing; as a result, the writers may miss the purpose of reading aloud for revision.
Some writing center tutor educators also advise tutors to be more reserved in applying RA when the clients are ESL writers. Gillespie & Lerner (2004) stated that some ESL writers will have great difficulty reading aloud because
reading aloud takes so much energy and concentration on the basic skills of decoding that they (the writers) don’t have anything left to listen to the words they are saying . . . . If writers feel hesitant, don’t force the issue (quoted in Gillespie, 2002, p. 45).
Williams & Severino (2004) suggest that RA may have different effects on different L2 writers: International L2 writers may be less able to catch errors than immigrant L2 writers when reading aloud, because the international L2 writers are sometimes described as eye-based learners; whereas immigrant L2 writers are sometimes described as ear-based English learners (Reid, 1997). That is, international L2 students and EFL learners have mostly learned their L2 or English through reading the language rather than picking up the language from listening to others’ talking, as most immigrant students would do. Thus, international and EFL writers may have less sense about what sounds right or wrong than immigrant students and therefore may not benefit from RA as much as immigrant L2 writers would.
Reading aloud as a proofreading strategy can be difficult to do well. Harris (1987) noted that reading aloud as a proofreading strategy involves a deliberate effort to counteract the “normal” process of reading, which includes predicting. If readers predict, they will see or read what they mean rather than what the text really is. However, while applying reading aloud, writers should pay close attention to the text and suppress their tendency to predict as they would normally do in reading. In addition, reading aloud as a proofreading strategy can be challenging for less proficient writers due to the constraint of humans’ short term memory. Citing cognitive psychologists’ findings, Harris (1987) indicated that the number of discrete items that can be held in short-term memory is seven plus or minus two. Thus, when the subject and verb of a sentence are separated by seven or more words, the writer may forget to make the verb agree with the subject due to the writer’s short memory constraint. Harris (1987) explains that because less proficient writers are unfamiliar with many of the sentence patterns used in academic writing, it is difficult for them to hold linguistic features in mind. Gibson (2008) also points out that reading aloud can be challenging even for native speakers in areas such as pronunciation, intonation, pace, and naturalness. These difficulties can be demoralizing for some L2 students who are self-conscious about their oral reading.
To sum up, the literature review shows that RA has been recognized as a useful strategy for L1 writing revision, yet its value for L2 writers has been inconclusive. While RA could be useful for L1 writers because it is based on their intuitions, some writing researchers have argued that RA may not be equally useful for L2 writers because whether L2 learners can develop the intuitions similar to those of the L1 speakers is still questionable. Although empirical studies by Zamel (1983) and Hedgcock & Lefkowitz (1992) have suggested that RA is beneficial to L2 writers, neither one of them has identified what kinds of self-initiated changes students were more likely to make intuitively, and how the changes may reflect the students’ L2 proficiency. Furthermore, neither one of them has considered students’ perceptions of the usefulness of RA to their revision. Although some researchers have argued that L2 students may be distracted by the pronunciation while reading their writing aloud, evidence is needed to verify this assumption. Likewise, since RA can be cognitively demanding for some L2 writers, whether they feel comfortable reading their own writing aloud to an audience is a question yet to be answered. Finally, even though Hedgcock & Lefkowitz’s (1992) study has
illustrated that reading aloud coupled with peer interaction was more helpful than simply receiving teacher’s written feedback, the gain ought to be attributed to a combined effect of both peer collaboration and reading aloud. It is not clear if the student writers they investigated could still gain as much if they just read their writing aloud without interacting with their peers. That is, we still do not know if (and to what extent) RA can help L2 writers to be self-sufficient in revising and editing. We equally do not know what self-initiated changes students are more likely to make, and how the changes may reflect their L2 proficiency. Therefore, this study intends to fill the gap in our knowledge of how useful RA can be in helping L2 writing revision.
Method
This preliminary study was conducted by the author, who is also the teacher/researcher, with an attempt to better understand how a pedagogical practice, RA, may influence EFL learners’ writing revision. To achieve this goal, an action research approach was adopted. Defined by Mills (2011), action research is “any systematic inquiry conducted by teacher researchers, . . . , in the teaching/learning environment to gather information about . . . how they teach, and how well their students learn” (p. 20). Like other types of research, action research also engages the researcher in identifying the research focus, (in this case, the effectiveness of RA in EFL writing revision), collecting data, and analyzing as well as interpreting the data. The data of this study came from the teacher/researcher’s students. The study involved 28 college-level English-major students (L1 = Chinese), who were enrolled in a sophomore English writing course at a university in Taiwan.
The following procedure was designed by the author to increase the validity of the findings. In the preliminary stage of this study, two steps were taken in order to find out the suitable participants. In the first step, each participant filled out a “Writer’s Profile” form (see Appendix A), which was designed to identify students’ proficiency levels and their revision habits. The form includes questions about the participants’ experience of learning English; their performance on major English proficiency exams, if any; their grades in Freshman English writing; their habits or methods of revision, if any; and if they feel comfortable with reading aloud to an audience. In the second step, the students were given a diagnostic writing assignment. Based on students’ proficiency test results, their grades in Freshman English, and their performance on the diagnostic writing, the students were separated into advanced (10 students), intermediate (9 students), and basic levels (9 students) according to their English proficiency.
In order to identify global and local errors, Zamel’s (1983, pp. 165-187) descriptions of global revision were adopted as criteria to identify global revision. That is, global revision may include one or some of the following changes:
1. Sentences are deleted and added to clarify ideas and make them more concrete. 2. Sentences are rewritten until they expressed the writer’s intention.
3. Paragraphs or parts of paragraphs are shifted around.
Students revise globally when a change in one part of their drafts drives changes in other parts of the drafts. Global changes may result in improved unity, support, coherence, logic, and overall effectiveness of the essay. Local changes, on the other hand, focuses more on sentence-level revision: changing words so that a sentence is clearer, correcting grammatical or spelling errors, changing mechanical mistakes, etc. Local changes to a text affect only the one or two sentences that the writer is currently working on.
In the study, the students were given two essay assignments (See Appendix B, Appendix C), and they were required to meet with the instructor for a read-aloud review after they wrote their first drafts of the assignment. The review was carried out on a one-on-one basis, during which each student read his or her writing aloud twice to the instructor. Students read their own texts rather than their peers’ to avoid decoding problems because reading an unfamiliar text may result in “barking at print” without much comprehension (Broughton, Brumfit, Flavell, Hill, and Pincas, 1980). For the first reading, they were advised to focus on meaning, and the second time, on form. Students were instructed to feel free to pause any time they wished to mark changes. While a student was reading his or her writing aloud to the instructor, the instructor also marked on another copy of the student’s writing.
After students finished reading their writing aloud, the instructor asked the students to explain how they wished to change the marked items, so that the instructor could know their intent. Then, their marked copies were collected. In total, 56 student-marked drafts and 56 teacher-marked drafts were collected. Student-marked drafts and teacher-marked drafts were photocopied, making one set for students to facilitate their revision, and another set for the teacher/researcher for the sake of this study. The markings that students initiated and the teacher initiated were further categorized by the teacher/researcher into global and local changes that students and the teacher/researcher had marked respectively, and all the changes were tallied in order to find if any patterns existed.
The second reading aloud of the second assignment was followed by a survey as well as an interview on students’ perceptions of applying RA in their revision process (See Appendix D, My Perception of Reading Aloud). The questions in the questionnaire and interview are based on claims about the potential benefits and problems of RA stated in the literature.
Results
The findings show that RA did help all the EFL students at various levels to self-detect some local as well as global problems, although it could not help EFL students correct “all the spelling, grammar, and punctuation errors” as L1 writing experts have claimed it could do to help most L1 writers (see Hartwell, 1980, p. 201). Below is a table of the global and local changes students made as a result of reading aloud.
Table 1
Students’ Self-initiated Global and Local Changes
Number of Ss Global Local Total
Advanced 10 6 (11%) 47 (89%) 53
Intermediate 9 5 (14%) 32 (86%) 37
Basic 9 12 (29%) 29 (71%) 41
This Table demonstrates that among all the changes students made, regardless of levels, all made more local than global changes. However, the rate of the global to local changes differs at each level. Basic level students had the highest rate of global changes, yet advanced level students had the highest rate of local changes. That is, basic level students made global changes much more frequently than the advanced and intermediate level students.
Findings from the Students’ Texts
Student-detected Global Problems. Findings of this study show that reading aloud did help students at all three levels make global revision, yet the global problems they detected were limited. Below is a table of each type of global problems students detected. Table 2
Global Problems Detected by the Students
Level Quantity Distribution of Global Problems
Detected by Students
Advanced 6 Unity (50%), Coherence (30%), Support (20%)
Intermediate 5 Coherence (60%), Unity (20%), Support (20%)
Basic 12 Support (73%), Unity (18%), Coherence (9%)
The global changes students made can be categorized into three areas: unity, support, and coherence. The most frequently detected problems for each level are different. The advanced students detected more global problems in unity (50%); the intermediate level students, in coherence (60%); and the basic level students, in support (73%). The actions students took to correct the global problems are similar at all three levels: Deleting irrelevant information, adding a thesis statement, increasing the clarity of a thesis statement, or modifying the support so that it backed up the main point (a thesis statement or a topic sentence). To improve the coherence of writing, students changed the title, added transitions, or provided concluding remarks or conclusions. To improve the support of writing, students often added background information or explanations in the form of sentences. Students also revised the support by rewording, rewriting sentences, deleting sentences, or shifting sentences around. Occasionally, however, the revision was futile, for the added information was not needed.
The global changes students made at all three level were very limited and at sentence-level only. No paragraphs were shifted around, and no new paragraphs were formed. Most of the changes made their writing more effective, although none of the changes was at paragraph-level.
Student-Detected Local Problems. Reading aloud proved to be beneficial to students in helping them detect local problems. In the advanced, intermediate, and basic level students’ texts, the students self-corrected 45, 32, and 30 local problems respectively. In economy of space, listed below are only the top four kinds of problems, which take up over 50% of total local problems detected by students at each level (see Appendix E for the complete results).
Table 3
Local Problems Detected by Students Most Frequently
Level Total # of
Changes
Top FourLocal Problems Detected by Students Advanced 45 Ineffectiveness (33%), Word choice (11%), Tense (11%),
Prepositions (9%)
Intermediate 32 Ineffectiveness (22%), Word choice (13%), Prepositions (9%), Part of Speech (9%)
Basic 30 Ineffectiveness (60%), Word choice (7%), Prepositions (7%), Tense (7%)
Students at all three levels self-detected many types of local problems in their texts. Among them, ineffectiveness, word choice, and prepositions were the most frequently detected problems shared by students of all three levels. Table 5 shows examples of frequent problems in student essays that students self-detected most often:
Table 4
Types of Most-Frequent Student-Detected and Self-Corrected Local Problems
Type of Problem
Example Change
Ineffectiveness I felt frustrated. After this winter, . . .
I felt frustrated completely. After this winter vacation, . . . Word Choice We were required to *remember
one article every week.
We were required to memorize one article every week.
Preposition We felt ashamed *to ourselves. We felt ashamed of ourselves.
Tense Our human beings *didn’t
change much from past till now.
Our human beings haven’t changed much from past till now.
Part of Speech In today’s social, . . . In today’s society, . . .
What is worth noting is that ineffectiveness, which ranked as the top among all the self-detected problems, is not a grammatical or mechanical error, but spots where students decided to modify. The results show that 33% of the advanced, 22% of the intermediate, and 60% of the basic level students’ self-detected problems belong to the category of ineffectiveness. To deal with the ineffectiveness, students added details/words such as adjective clauses, prepositional phrases, or adverbs, and they also reworded a sentence, deleted words, replaced the vague words with specific ones, or added modals to show possibility. Their modification did increase clarity, and their writing became easier for the
audience to understand.
In addition to ineffectiveness, two other types of the most often self-corrected local problems shared by students at three levels are word choice and prepositions. Word choice took up 11%, 13%, and 6% of the total local problems self-detected by advanced, intermediate, and basic level students respectively. Moreover, problems in prepositions detected by students of advanced and intermediate levels were 9% each, which is a little bit higher than the basic level students’ 7%. Thus; students at higher proficiency levels were more able to self-correct word choice and preposition problems than basic-level students.
Tense errors were also among students’ frequently corrected local problems. 11% of the advanced, 6% of the intermediate, and 7% of the basic level students’ self-corrected local problems are tense errors. The rest of the grammatical and mechanical errors that were self-corrected by students at all three levels include problems in punctuation, articles, conjunctions, verb forms, sentence structures (missing a verb, an object, or a relative pronoun), idiomatic expressions, pronoun references, run-ons, and fragments. However, each of these errors was self-corrected only once or twice in each level students’ texts.
Problems in Students’ Texts Detected by the Teacher
The reading aloud sessions also allowed the teacher to mark the global and local problems that she found in students’ texts. To obtain a clearer picture of the effectiveness of RA on students’ revision, the numbers of problems that the students’ detected and that the teacher detected were compared.
Teacher-detected Global Problems. The global problems identified by the teacher can be categorized into four major areas: unity, support, coherence, and logic. The last area, logic, is an area that only the teacher detected. The global problems detected by the teacher are 24, 20, and 42 at the advanced, intermediate, and basic level students’ texts respectively. Below is a table of the most frequent problems that the teacher found in the students’ texts of each level.
Table 5
Global Problems Detected by the Teacher
Level Quantity Global Problems Detected by the Teacher
Advanced 24 Support (29%), Logic (29%), Unity (25%), Coherence (17%)
Intermediate 20 Unity (47%), Support (29%), Coherence (18%), Logic (6%) Basic 42 Support (64%), Logic (17%), Coherence (12%),
Unity (7%)
An examination of the distribution of global problems reveals that advanced level students had a more even distribution in all four types of problems than students of two other levels. The most noticeable result is that the problems of support occupy 64% of all the problems the basic level students have. The problems of support are mostly caused by underdevelopment, such as incomplete meaning, incomplete explanations, or lack of
continuity between one idea and another. Other problems of support include redundancy, un-specific evidence, or unbalanced support (e.g., too much information about the background yet not enough about the topic).
Logic problems are also very frequent in the advanced and basic level students’ texts. In fact, for the advanced students, the logic of writing is as challenging as the support: Each one of them occupies 29% of their global problems detected by the teacher. Oftentimes the logic problem is in the relationship between cause and effect: The cause does not necessarily lead to the effect; or the effect is exaggerated, forming over-generalization. Other times students were not aware that the information they provided in different places was in conflict. Yet other times, the students were not aware that there was a gap between the lead and the thesis statement.
Unity is the area that intermediate level students have most global problems. It is also the source of the third and the fourth most frequent global problems found in advanced and basic level students’ texts. Some frequent problems include (1) the main point is not mentioned or not clear, (2) the focus of the essay is not right, and (3) the information provided is irrelevant to the topic.
Coherence is yet another area that causes global problems for students at all three levels, although the frequency of coherence problems is the lowest among the four areas of global problems. Coherence problems take various forms. For example, mismatches between the title and thesis statement, between the introduction and body, and between the body and conclusion. In addition, using inappropriate transitions and presenting new information in the conclusion were also common.
Teacher-detected Local Problems. In the advanced, intermediate, and basic level students’ texts, the teacher found 53, 63, and 147 local problems respectively. It is clear that the lower the proficiency level, the more the local problems students have. Below is a chart that records the most frequent local problems detected by the teacher.
Table 6
Local Problems Detected by the Teacher Most Frequently
Level Quantity Top Four Local Problems Detected by the Teacher Advanced 53 Word choice (25%), Prepositions (13%), Idiomatic (9%),
Tense (9%)
Intermediate 63 Word choice (11%), Prepositions (10%), Idiomatic (10%), Run-ons (8%)
Basic 147 Word choice (21%), Prepositions (10%), Idiomatic (10%), Discourse Markers (7%)
The top three most frequently found local problems in students’ texts at the three different levels are the same: Word choice, preposition, and idiomatic expressions.
Word choice appears to be the most challenging item for students at all three levels. Word choice problems may result from students’ unfamiliarity with usages of certain words. It took up 25%, 11%, and 21% of the total local problems found in advanced, intermediate and basic level students’ texts. In the advanced and basic level students’ texts, the frequency of word choice problems is almost twice that of prepositions, the second
most frequent problems.
Looking at the second and third most frequent problems, we can find that the rates of both prepositions or idiomatic expressions are the same 10% for either intermediate or basic level. Therefore, prepositions and idiomatic expressions are equally difficult for students of these two levels.
In addition to the top three most frequent local problems, students at all three levels also share the following types of problems: tense, articles, capitalization, run-ons, and wordiness. Yet the frequencies of these problems are much lower.
Efficacy of RA on EFL Writers’ Revision
Two questions in this study are (1) which level of students benefited most from RA, and (2) in what area they benefited most. The results indicate that RA helped the basic level students more than students of two other levels in detecting global problems. The ratios of student-self-detected to teacher-detected global problem at advanced, intermediate and basic levels are 25%, 25%, and 29% respectively. Although the basic level students’ ratio of self-detected global problems is 4% higher than the ratios of two other levels, the difference is not significant because the total numbers of global problems self-detected by all three level students are not high. The global problems detected by students and teacher most often at each level do not completely match. While advanced students self-detected most global problems in unity, the teacher found they had most problems in support. The intermediate level students self-detected most problems in coherence, yet the teacher found they had most problems in unity. The basic level students self-detected most problems in support, and this is the only group that matched what the teacher detected. However, no students at any level detected any logic problems, which may suggest that they had little experience in dealing with logic problems prior to this study.
While RA is most beneficial for the basic level students to detect global problems, it is most beneficial for the advanced level students to detect local problems. Comparing the ratios of student-detected local problems to teacher-detected local problems, the advanced level students’ is the highest (89%), much greater than the intermediate level (51%), and basic level (20%). That is, when advanced level students read aloud, they can find more local problems than intermediate and basic level students can do.
An examination of the local problems self-detected by students reveals that ineffectiveness was what all the students self-detected most often. As mentioned earlier, ineffectiveness is not a grammatical or mechanical problem, but the spot that students felt they wanted to change so that their meaning could become more effectively expressed. The basic level students had the highest rate of detecting ineffectiveness. In addition to ineffectiveness, grammatical or mechanical problems were the local problems that students could self-detect with higher rates of success than other types of local problems: 67% of the local problems detected by the advanced level students, 78% of the total local problems detected by the intermediate level students, and 40% of the total local problems detected by the basic level students were grammatical or mechanical problems. These
findings suggest that after reading aloud, basic level students are not as good as intermediate and advanced level students in detecting grammatical or mechanical problems.
Ineffectiveness is not an item identified by the teacher. It is because the problems of ineffectiveness perceived by the students sometimes were not considered to be ineffective by theteacher. Students were the ones who could decide whether they needed to add a prepositional phrase here or a few words there. Oftentimes, without the addition, the sentences were still correct or clear enough to understand.
In addition to ineffectiveness, word choice and prepositions are two local problems that both students and the teacher detected most frequently. However, a closer look at the results in Appendix E reveals that students of advanced and intermediate levels detected just about 50% of the word choice or preposition problems detected by the teacher, and basic-level students detected only 6% of the word choice and 13% of the preposition problems that were detected by the teacher. In other words, there were still a great number of word choice and preposition problems that students were not able to detect.
Furthermore, the results also show that the teacher found that local problems in idiomatic expression were the third most frequent local problems in the students’ texts at all three levels, but they were rarely detected by students. In fact, there was only one case of idiomatic expression correction made by an advanced level student. Therefore, it is not exaggerating to say that problems in idiomatic expressions can be extremely challenging for EFL students to detect.
Moreover, run-ons and discourse markers are two other types of problems that teacher often detected but students did not. Run-ons are the fourth most frequent problems that the teacher found in intermediate level students’ texts. The results show that there was only one case of run-on corrected by an advanced student. Problems in discourse markers are also the fourth most frequent problems in basic level students’ texts yet rarely detected by the students. Both run-ons and discourse markers are related to the junction of two thoughts. Students’ lack of ability to deal with them may suggest that intermediate and basic level students have more problems in identifying the boundary of thoughts and the logic between one thought and another.
Over all, the modification students initiated during RA mostly resulted in improvement. Occasionally, however, RA also caused the students to make unsuccessful or undesirable changes, which resulted in grammatical errors, unconventional wording or less effective sentences. Below are some examples from the students’ texts:
1. My father *who likes reading and also likes buying me books. (“who” added) 2. I didn’t know what the words in the books mean *meaning.
3. I had a passion for *of reading.
4. I wanted nothing *for me but my mother’s approval (“for me” added).
The undesirable changes triggered by RA were found in students’ local changes only, and the problems tended to happen more frequently when beginning level students read their writing aloud. Although the undesirable changes did not happen very often, it is clear that RA is not definitely helpful for EFL writers, especially for those who have less developed sense of what sounds right.
So far we have covered the most frequent problems found in the students’ texts. Now we turn to the second major category of the research data—the interview results.
The Results of the Interview
The results of interview show that 94% of students have the habit of revising their writing, and 60% of them usually revise their writing by reading aloud. At the advanced level, 50% of the students prefer to read their drafts aloud; at the intermediate level, 75% of the students prefer to read aloud; at the basic level, 60% of the students prefer to read aloud. Thus, the intermediate level students have strongest preference in reading aloud.
In the interview, 80% of all students, even including some of those who prefer reading their drafts silently, agreed that RA did help them. Among them, 60% felt RA was very helpful, and 20% of them felt RA helped, but not much. In the teacher/ researcher’s observation, 95% of students made some changes as they were reading their writing aloud to the teacher. In answering the question how RA helped them, 89% of the students felt that RA helped them become more aware if the language they used was accurate or clear enough to express their meaning, 71% of them felt RA helped them understand the audience’s needs, and 57% of them felt RA could help them overcome writer’s block.
In terms of their affective level, 78% of students felt relaxed when they read their writing aloud to the teacher. Some positive comments include: “I know the purpose of RA is to help me improve my composition,” “I know exactly what I want to express,” and “I feel like reading to a friend who is concerned about my writing.” However there were also some students who were more reserved. One student felt “nervous” because she was afraid the teacher could not understand her writing. Another student felt “RA is interpersonal, and I’m not used to reading my thought out loud to teachers, especially.” Yet another student expressed: “At first I felt nervous because I thought my writing was poor. But later, after reading my writing aloud and talking with the teacher, I became relaxed because I knew I could still do something to improve it.”
In one question, the students were asked if they would be distracted by pronunciation when reading their writing aloud. Although 53% of students answered “no,” 47% expressed that they would. Among those who would, one student stated, “I’m always aware how I sound when I’m reading aloud,” another student worried about her “poor” pronunciation, yet another student said, “I felt I had to continue reading and I had no time to think about the meaning.”
In the question regarding whether students can see their writing from the audience’s point of view when they read their writing aloud, 44% affirmed that they could, 11% felt not sure, and 56% of students stated they could not. A student who answered “No” explained, “I would think in my own logic, and I felt satisfied with the content when I finished writing. I knew the background of my content, without thinking about what others might not know.”
In the last question, the students were asked when they re-read their writing to revise, if they re-read a small portion (words or sentences) or a larger portion (a whole paragraph
or paragraphs) before and after the place where they wished to change. The majority (78%) of the students would re-read a larger portion, even though that their proficiency levels were different. One student noted, “I usually do so because I can know if the paragraph is fluent or not.” Only 22% of students would read a small portion. In this 22%, there were both advanced and basic level students.
Summary and Discussion
In this section, the key findings are first listed and then discussed. Below are the key findings.
1. RA did trigger EFL students to make self-initiated revision.
2. Regardless of students’ proficiency levels, most changes students made are local changes. Among them, modification of the ineffectiveness of sentences ranks the top. Other local changes are fairly few.
3. Although RA helped students most in making local changes, problems of word choice, prepositions, and idiomatic expressions are still difficult for them to self-detect.
4. RA was not quite effective in helping EFL writers revise globally. Less than 30% of the revisions students at all three levels made are global revisions.
5. The higher the students’ proficiency level, the more they gain from RA for local revision. RA appears to be most beneficial for the advanced level students to make local changes.
6. RA occasionally triggered students, especially of the basic level, to make unsuccessful changes.
7. Most students perceived RA as a useful revision strategy.
The findings show that RA did help students at all three levels revise their texts. It was particularly helpful for students to make local revisions: Over 70% of all the changes students made are local or at sentence level. There are two major categories of local problems students can self-detect as a result of reading aloud: One is the ineffectiveness in writing, and the other one is grammatical/mechanical errors. The ineffectiveness students could self-detect was mostly at sentence level, and they dealt with the problems by adding words or prepositional phrases to improve the clarity of the information. Students at all three levels revised ineffectiveness most often when they read their drafts aloud. In fact, RA proves to be most effective in helping the basic level students detect ineffectiveness, as 60% of their self-initiated revision is about modifying the ineffectiveness of their writing. This shows RA can help students, even students at basic level, engage in rhetorical concern. In addition, vocalization during revision when students think about the language and content may facilitate revision to an even greater extent; in this study, the basic level students did capitalize on RA as the “running accompaniment” described by Vygotsky (1962, 1978)” (Hedgcock & Lefkowitz, 1992, p. 259).
RA also helped students detect various types of grammatical problems, yet there were also other local problems that RA offered little help. The ratios of the of students’ self-initiated to the teacher’s recommended changes reveals that there were still a great number of problems in word choice, prepositions and idiomatic expressions un-detected
by students. Problems in these areas appeared to be very difficult for EFL learners to self-detect and self-correct because the correct usage depends much on the native speaker’s competence. In other words, rules printed in grammar books cannot possibly provide everything L2 writers need to know in order to avoid errors. On the other hand, RA occasionally triggered unsuccessful corrections, making the original correct sentences wrong. This happened more frequently in beginning level students’ revision while they were reading aloud. These findings confirmed what Matsuda and Cox (2009) have speculated: RA may not work well for the EFL or ESL students who have not developed that intuitive sense of the English language.
Furthermore, RA is not effective in helping them detect global problems. Students at all three levels in general detected about just one quarter of the global problems that the teacher detected. This lack of efficacy in detecting global problems may be attributed to that reading aloud may present a cognitive challenge to some students whose second language is not proficient enough for them to stay focused on many things at once. When reading their writing aloud, L2 writers’ attention is split among several different demands: form, meaning, and even pronunciation. Based on Skehan and Foster (2007), humans have limited information processing capacity and must therefore prioritize where they allocate their attention. Learners cannot pay attention to language forms without a loss of attention to language content (VanPaten, 1990). When the learners have an even more pressing concern--to vocalize their writing, some of them may not be able to make use of everything they know to examine the form and meaning of their writing. That is, when their proficiency level in the English language is not high enough for them to stay focused on several things at once, some of them may shift their attention to pronunciation while reading aloud. The findings reveal that 47% of them were distracted by their pronunciation; consequently, it was difficult for them to pay attention to the global problems. Furthermore, some of them did not feel comfortable reading aloud to a teacher. When reading to the teacher becomes stressful, it was difficult for them to think about any local and not to mention global matters.
Although the changes triggered by RA were limited, the participants of this study generally had positive attitudes toward the use of RA as revision strategy. The majority of them expressed that RA helped them become more aware if the language they use is accurate or clear enough to express their meaning. Some of them expressed that RA helped them overcome writer’s block; others stated RA helped them see what changes they need to make so that the audience can understand them. Furthermore, most students agreed that RA helped them see what they need to add to make the writing more concrete to the audience.
The interview results also indicate that the majority of the students could concentrate on revising their drafts, yet close to half (47%) of the students felt they were distracted by their pronunciation. Matsuda and Cox (2009) have pointed out that reading aloud may shift the writer’s attention to pronunciation; consequently, students may miss the purpose of reading aloud for revision. Although reading their writing aloud did trigger noticing, which prompted the majority of students to consciously recognize some of their writing
problems, RA also triggered some students to notice their pronunciation. For instance, one of them worried about her poor pronunciation, another paid more attention to how she sounded, yet another one felt she was obligated to keeping the reading (vocalizing) going until it was concluded. Results like these suggest that RA may not be an appropriate revision strategy for all EFL students.
Hedgcock and Lefkowitz (1992) have argued that reading one’s text aloud to attentive peers may foster an awareness of audience. In the interview, students were asked if they could see their writing from the perspective of an audience. Only 44% of the students said they could, and 56% of them said no. However, the results also show that when students read aloud, the most frequent action they took in their reviewing process was to modify the ineffective sentences by adding more information. This modification may indicate that they were aware of the audience, but they might not completely understand the question.
In the affective level, most students expressed they felt relaxed when they read aloud. Only 22% of students felt nervous due to a lack of confidence in writing (e.g., “I am a poor writer”), feeling uncomfortable to read personal thoughts to a teacher, or feeling their writing or pronunciation was not understandable to the teacher. Yet one student felt nervous first but relaxed later as the interaction between the teacher and her made her believe there was still hope to improve her writing. Apparently, the rapport between the student and the teacher seemed to have influenced the student’s perception of reading aloud. For most students, RA did not make them nervous. They explained that they felt they were reading their drafts to a concerned friend, they had confidence in their writing, or they trust that the teacher is asking them to do something that will help them improve. In sum, affective factors do influence students’ attitude toward the use of RA.
Conclusion
This study was inspired by Hartwell’s (1997) statement that L1 writers can correct in essence all errors of spelling, grammar and punctuation by reading their writing aloud. The researcher wondered to what extent reading aloud could help L2 writers. The results of this study show that RA can trigger EFL students at various levels to make self-initiated revision, although not as comprehensive as what L1 writers could do after reading aloud their writing. The global problems students could self-detect were fairly limited. Most of the self-initiated revision is local. Even so, RA has proved to be helpful for students to self-detect the ineffectiveness in writing, albeit at sentence level. Overall, students at higher proficiency can benefit more from RA than students at lower proficiency in detecting local problems.
The results of this study also show that problems in word choice, prepositions, idiomatic expressions, and tense are most frequent yet very difficult problems for students to self-detect, even if they have read their writing aloud many times. One common characteristic of these problems is that they are all competence-related. When EFL writers have not developed their L2 competence, in this case their English sense, similar to that of native speakers’, it is difficult for them to detect and correct these problems by reading
their texts aloud.
Furthermore, reading aloud may help students notice the language they produced, both written and pronounced. During RA, most students in this study could concentrate on the texts they wrote, yet some students noticed their pronunciation more; as a result, they missed the purpose of applying RA as a revision strategy. Moreover, interview results reveal that students’ confidence in writing and/or pronunciation, and their trust in their teacher may influence their attitude toward RA.
Zamel (1983) has argued that proficient ESL writers have been found to revise substantially after reading their texts aloud. Yet in this study, even the advanced writers did not revise “substantially.” In fact, students at all three levels did very little global revision. One possible explanation is that in this study the proficiency level of the advanced students is not high enough. Meanwhile, some of the students’ anxiety levels might be raised due to their pronunciation or their worries about facing a teacher as their audience. Consequently, they did not make much global revision.
Limitations and Implications
One limitation of the current study is the low number of participants. Although 28 students participated in this study and each of them produced two drafts, each level has only 9 or 10 participants. Also, there was only one rater. Researchers of future studies can increase the numbers of both raters and participants in order to improve the reliability of the results.
The research design of a future study can also be improved by recruiting two groups of students whose English proficiency levels are sharply different, e.g., one equivalent to the superior level of GEPT, and the other one, intermediate level of GEPT. By doing so, the findings may reveal even more about the interface between the participants’ L2 proficiency and their revision via RA. Nonetheless, this study has some important implications for teaching and research.
Implications for Teaching
In terms of teaching, we have learned that RA may help, but it is not a panacea, and it may not be suitable for every student. Also, teachers may want to establish a good rapport with students before they ask students to read aloud to them, for some students may not feel comfortable reading their writing aloud to teachers. Furthermore, since some accurate language usage may depend on the native speakers’ competence, teachers may want to explicitly point out the problems related to students’ competence and tell students to accept or even memorize how certain words, prepositions or idiomatic expressions are used by native speakers of English. Lastly, to help students benefit most from RA, teachers may need to emphasize that (1) students should read in a way that they can understand the meaning of the texts (e.g., with more intonation), and (2) students should feel free to stop any time to mark the parts that they wish to revise.
Implications for Research
This study can be seen as a stepping stone for some future studies on reading for revision. A future study may compare the effect of reading aloud with reading silently, as this study shows that some students did not feel comfortable with reading their drafts aloud. Another future study may compare the effect of students who read aloud twice with students who did a combination of reading aloud once and reading silently once, because some writers do use both in their revision. Yet another future study can compare the effects of students reading aloud with the teacher reading aloud on students’ revision, so that we can know which method can help students more.
Despite its limitations, this study has shed light on what RA can do to help EFL writers revise. The results have informed writing teachers the problems that students can or cannot self-detect by EFL students at various proficiency levels after reading their writing aloud. The results of this study have further informed teachers how EFL students perceive the use of RA in their revision. With this understanding, writing teachers can design appropriate lessons or syllabi to help students revise more efficiently and effectively.
References
Anson, C. M. (1989). Writing and response: Theory, practice, and research. Urbana, IL: National Council of Teachers of English.
Barnawi, O. (2010). Promoting noticing through collaborative feedback tasks in EFL college writing classrooms. International journal of teaching and learning in higher
education, 22(2), 209-217.
Bartholomae, D. (1980). The study of error. College Composition and Communication, 31, 253-269.
Broughton, G., C. Brumfit, R. Flavell, P. Hill, A. Pincas (1980). Teaching English as a
foreign language (2nd ed). London: Routledge and Kegan Paul.
Capossela, T. –L. (1998). The Harcourt Brace guide to peer tutoring. Fort Worth: Harcourt Brace.
Carson, J. G. (2001). Second language writing and second language acquisition. In Silva, T., & P. Matsuda (Eds.), On Second Language Writing (pp. 191-200). Mahwah, NJ: Erlbaum.
Chandler, J. (2003). The efficacy of various kinds of error feedback for improvement in the accuracy and fluency of L2 student writing. Journal of Second Language Writing,
12(3), 267-296.
Chomsky, N. (1965). Aspects of the theory of syntax. Cambridge, MA: MIT Press. Cogie, J. (1999). Avoiding the proofreading trap: The value of the error correcton process.
The Writing Center Journal, 19(2), 7-29.
Cohen, A. D., Larsen-Freeman, D., & Tarone, E. (1991). The contribution of SLA theories and research to teaching language. ED 337047.
Cumming, A. (1990). Metalinguistic and ideational thinking in second language composing. Written Communication, 7, 482-511.
Deng, H., & Huang, L. (惏ㄏ⏃, & 湫列䚇) (2011). A study of reading aloud to EFL elementary school students. Studies in English Language Literature, 28, 25-36. Ellis, R. (1995). Interpretation tasks for grammar teaching. TESOL Quarterly 26, 87-105. Evans, F. B. (1988). Vocalizationas a prompt for self-correction of written text: A research
proposal. Unpublished manuscript, University of Southern California, School of
education, Los Angeles.
Ferris, D. (2007). Preparing teachers to respond to student writing. Journal of Second
Language Writing, 16, 165-193.
Ferris, D. (2008). Myth 5: Students must learn to correct all their writing errors. In J. Reid,
Writing Myths (pp. 90-114). Ann Arbor, MI: University of Michigan Press.
Freedman, S. W. (1987). Response to student writing. Urbana, IL: National Council of Teachers of English.
Gibson, S. (2008). Reading aloud: A useful learning tool? ELT Journal, 62(1), 29-36. Gillespie, P. (2002). Beyond the house of lore: Writing center as research site. In P.
Gillespie, A. Gilliam, L. Brown, & B. Stay (Eds.), Writing center research:
Extending the conversation (pp. 39-52). Mahwah, NJ: Erlbaum.
Gillespie, P., & Lerner, J. (2004). The tutoring process. In The Allyn and Bacon Guide to
peer tutoring (pp. 23-36). Boston: Allyn and Bacon.
Griffin, L. (2011). Take the cure for revision resistance: Here are 5 steps for improving the condition of your novel. Writer, 124(7). http://
web.ebscohost.com/ehost/detail?sid=45aa8cfc-f16f-487f-8d1d-c8854c0722f%4. Downloaded August 11, 2011.
Harris, J. (1987). Proofreading: A reading/writing skill. College Composition and
