CHAPTER TWO LITERATURE REVIEW
From Product Writing to Process Writing
Before the 1970s, product approach dominated much of writing research and classroom practices. Most of writing research focused on the written products that students produced. Realizing that the product approach which focused on the written products did not take into consideration the act of writing itself, many writing teachers and researchers in the late 1970’s and early 1980’s, proposed a new perspective to writing research and education (Flower & Hayes,1977; Sommers, 1980, 1982; Zamel, 1982, 1983, 1985).
Product Approach
In a typical product-oriented writing classroom, the instruction is primarily concerned with linguistic knowledge. Knowledge about the structure of language, including the appropriate use of vocabulary, syntax and cohesive devices, is
emphasized (Badger & White, 2000). In other words, the emphasis of the class was on grammar and usage (Flower & Hayes, 1977).
As for the evaluation of writing, teachers focus mostly on the final product, or the only copy of a writing assignment handed to them. They give comments and grades as an editor and a judge. Multi-draft writing is seldom required. Such practice does not allow students any opportunity to rethink and rework based on feedback.
Without between-draft feedback and the process of revision, students who receive poor grades are unclear about how to make their writing better (Koffolt & Holt,
1997).
Process Approach
In the 1970s, the process approach that looked into the process of writing began to gain power as an opposing reaction to the traditional product-based approach (Gee, 1997). Researchers in the camp of process approach criticized the traditional product approach as one that “fails to intervene at a meaningful stage in the writer’s
performance” (Flower & Hayes, 1977, p. 449). Teachers and researchers worked under the assumption that understanding the writing process must precede the
teaching of writing (Zamel, 1982). They explored writing behaviors, believing that by studying the process of composing, the key to teaching writing could reveal (See for example, Emig, 1971; Pianko, 1979; Zamel, 1983).
Research on the writing process of skilled and unskilled writers indicates that skilled writers view composing as non-linear and exploratory while unskilled writers see writing as linear and formulaic (Sommers, 1980; Zamel, 1982). The concept and act of revision is identified as one major factor that differentiates the composing process of skilled and unskilled writers. Skilled writers are capable of reading from an audience’s perspective and reconstructing whole chunks of discourse while unskilled writers limit their revision in correcting local errors, like words and phrases (Bridwell, 1980; Faigley & Witte; 1981; Monahan, 1984; Raimes, 1985; Sommers, 1980; Taylor, 1981; Yagelski, 1995). Sommers (1982) suggested that through revision, writers could
“approximate more closely and more accurately one’s intended meaning” (p. 197).
In pedagogy, the process approach to writing lays an emphasis on the “generating, formulating and refining one’s ideas” (Zamel, 1982, p.195). Teachers employ
available techniques, such as individual conferences, group sharing, team editing and
peer conferences, to encourage students to think about their work and revise it until it effectively expresses their ideas (Chen, 1997). Students are thus required to write multiple drafts based on the comments from the teacher or their peers in the hope of learning to view their writing more objectively and critically from an audience’s perspective (Koffolt & Holt, 1997; Zamel, 1982). Koffolt & Holt (1997) proposed a teacher-guided process approach to writing assignments, in which the teacher’s role was to help students “develop useful strategies for gathering ideas on a topic, organizing ideas, drafting, revising and editing and polishing” (p. 54). An essential step during the process is to provide feedback to the students through multi-draft writing.
ESL/EFL Writing Research and Pedagogy
While process approach to teaching writing has been pervading L1 writing classrooms for years, it has not yet been widely adopted in L2 writing classrooms.
ESL/EFL writing teachers are usually preoccupied with the concern of promoting their students’ language skills and thus often emphasize form over content.
Conducting a survey of college English composition instructors’ views and methods of teaching English writing in Taiwan, Chen (1997) found that many instructors focused much of their attention on the correct use of grammar and words and thus devoted much effort and time on the improvement of these features. Chen identified three reasons behind such practice: first, grammar and words are important; second, doing grammar exercises helps students improve their grammar; third, correcting students’ grammar and word errors helps them write better (p. 199). The survey finding reflected a typical product-oriented approach to teaching English writing in an
In an L2 writing class, because students are still in the development of their language ability, the language problems can easily become the primary focus and the linguistic correctness can become instructors’ main concern. Zamel (1982) noticed that L2 teachers emphasize correctness and form more than L1 teachers because L2 teachers were concerned with language acquisition and error analysis. Weak writing by a non-native speaker is easily assumed to result from poor English proficiency.
However, the real problem may not lie in the language-related errors. It is often the way the author presents his/ her ideas that is ineffective or confusing. As Koffolt &
Holt (1997) put it, “It is easier to tolerate a ‘foreign accent’ (grammar errors) in a research paper by an international student if the paper has a clear thesis,
well-developed ideas and an effective organizational structure” (p. 54).
Although process approach has not been widely adopted in L2 writing classrooms, many ESL/EFL researchers and teachers are becoming aware of the paradigm shift in composition teaching (Zamel, 1983). Zamel (1983) studied six advanced ESL student writers and found that their writing process was the same as that of native-speaking writers and thus suggested what had been found in the writing process for native speakers could be applied to non-native speakers. Since then, many researchers have started to engage in empirical studies in an ESL/ EFL contexts. Kim (1996), for example, compared writing produced by three different groups of writers;
native English-speaking Americans, Koreans studying in the United States and Korean students in Korea. He found that the native “advanced” writers revised a little more than the non-native “advanced” writers, who revised much more than the non-native “basic” writers. He also noticed that the non-native advanced writers paid more attention to grammar and spelling than did non-native basic writers. Koffolt &
Holt (1997) found process writing approach advantageous for non-native writers.
They identified four key points in implementing this approach: encouraging
pre-writing, encouraging planning, encouraging conference groups and encouraging multiple drafts. Raimes (1985), on the other hand, studied unskilled ESL students’ act of composing and attempted to propose teaching methods specific for ESL classrooms.
She concluded that unskilled ESL students, compared with unskilled L1 writers and proficient ESL writers, needed not only more instruction and practice in generating, organizing and revising ideas but also more emphasis on editing for linguistic form and style. She warned that, for ESL student writers, “attention to process is necessary but not sufficient” (p. 250).
The two main ideas of process writing approach—emphasizing content over form and the writing process over the writing product—have been accepted by many researchers and educators. However, it may be as harmful going extreme as ignoring the suggestions made by researchers. L2 students, in particular, need a carefully balanced instruction in terms of attention to content and form as well as attention to the writing process and the writing product. More empirical studies are needed to examine the effects of process-oriented writing instruction on the content and form in L2 students’ writing.
Effects of Writing Procedures on Writing
Writing procedures, which include providing instruction, writing, and giving feedback, have been found to be a major factor that determines how students approach the writing process, view feedback and make revisions to their writing.
Focus in Instruction: on Form vs. on Meaning
Wallace and Hayes’ (1991) study demonstrated that even as short as eight
minutes’ instruction on revision could result in significant improvement in students’
revision quality and amount of global revision. It is, therefore, assumed that student writers may be able to fulfill teachers’ expectation of global revision if they have appropriate understanding of what in-depth revision entails. Wallace and Hayes’ study demonstrates the effect of instruction on students’ revision task.
Teachers’ focus on form or on meaning is one of the factors in an instructional context. Teachers’ emphasis on from or meaning can seriously influence students’
concept of revision and writing. Reviewing the connections between the context of a writing activity and the revisions students make in that context, Yagelski (1995) pointed out that students’ concept and practice of revision had much to do with their experience as a writer in schools. Her own 1995 study further supported the
conclusion. Thus, she argued that the instructional context might have reinforced students’ tendencies to focus their revisions on lower-level concern and disputed Bridwell’s suggestions that students’ revision was determined by their “developmental differences”(Bridwell, 1980, p. 219), Sommers (1980) compared the revision process of experienced writers and college student writers. She asserted it was not that the student writers were unwilling to revise beyond sentence level but that they
misunderstood revision process as “a rewording activity” (p. 382). She concluded that these students might have been simply doing “what they have been taught to do in a consistently narrow, predictable way” (p. 383).
In other words, the writing instructor is in charge of what to emphasize in the writing classroom and thus influences how students view the writing task. If
instructors emphasize form over meaning or if their instructions lack global revision strategies, students will see revision as a rewording activity and focus on lower-level concern when writing or rewriting (Hedgcock & Lefkowitz, 1994, 1996; Raimes,
1986, 1987; Sommers, 1980; Wallace & Hayes, 1991; Yagelski, 1995; Zamel, 1983).
One issue of the instructional emphasis in a writing class is the role of grammar correction. The traditional writing class has been criticized to overemphasize correct grammar and words. Conducted a survey of college English composition instructors’
views and methods of teaching English composition in Taiwan, Da-Wu Chen (1997) found that many instructors still used grammar exercises as a tool to improve students’
writing and they invested much time and energy in correcting students’ grammar and words. Such emphasis on grammar correction has prompted some researchers to examine its effectiveness. Truscott (1996), for example, criticized grammar correction as ineffective and suggested a complete abandonment in the writing classrooms.
However, other researchers have argued for the role of grammar instruction in writing. Ferris (1999) suggests error correction does help students write more correctly and communicate more successfully. Zamel (1982), though advocating a process-oriented writing class that pays more attention to the process of writing, still acknowledges the importance of language instruction in an ESL writing class. Zamel (1984) pointed out “We need to find out if a minimum level of language competence is required before students are able to view writing in a second language as a process of discovering meaning” (p. 198). Lee and Schallert (1998) also suggests that one of the priorities of the EFL teachers is to help students with lower English proficiency to improve their English. Raimes (1985) suggested that, for unskilled L2 writers,
“attention to process is necessary but not sufficient” (p. 250) and that they need not only “more instruction and practice in generating, organizing, and revising ideas” but also “more emphasis on editing for linguistic form and style”(p.250).
Ashwell (2000) gave a clear account of how process writing advocates view grammar correction. He pinpointed that process writing advocates acknowledged the
importance of linguistic accuracy because it affected the communicative effectiveness of a piece of writing and grammar correction was intended to improve the accuracy.
However, linguistic accuracy should not dominate instruction. It is regarded as a secondary concern lower in priority than concern about meaning.
Still others have acknowledged the effectiveness of grammar instruction only under certain conditions. Bamberg (1978), for example, suggested, besides instruction and revision combined with writing practice, functional grammar instruction was another major approach to teaching writing. She also recommended that teachers teach grammar based on student errors and integrated with the total writing process instead of formal grammatical rules apart from writing.
Despite some arguments on the effectiveness of grammar instruction in writing, researchers over these years have generally agreed upon a change of emphasis from the writing product to the writing process and suggested instructors to teach students strategies of generating ideas, dealing with feedback, composing multiple drafts and revising their essays on all levels (Paulus, 1999).
Effect of Revision and Multi-draft Writing on Writing Qaulity and Attitudes
Revision has been emphasized as an integral part of the writing process by advocates of process writing approach. Researchers claim that it can enhance students’ understanding of writing as a discovery of meanings (Zamel, 1983).
Researchers have determined that the requirement of multi-draft writing can influence students’ perception of revision and writing. Koffolt and Holt (1997) suggested that multi-draft writing could encourage students to become more aware of their writing process and to regard planning and drafting as crucial elements of good
writing. Similarly, Chen (1997) studied the effects of two different writing contexts on students’ writing quality and perceptions. She found students in the multi-draft group not only were more positive toward the effects of revision but also were more willing to revise than students writing in the single-draft context.
Another benefit of multi-draft writing is that students will be guided to see content as more important than form. Much research has revealed that experienced writers deal with global issues, such as content and organization, before turn to work on local problems, such as grammar and vocabulary while inexperienced writers often pay too much attention to local problems and ignore global issues (Bridwell, 1980;
Raimes, 1985; Yagelski, 1995; Zamel, 1985). Once students have put much effort on grammar and vocabulary, they become even more unwilling to engage in a whole reconstruction of the content (Sommers, 1982). In a multi-draft writing classroom, teachers comment on content in preliminary drafts before switching to focus on form in later drafts and students will be guided to emphasize meaning-level issues or form-level issues.
Still another benefit of multi-draft writing is that students tend to pay more attention to teacher feedback. Ferris (1995) conducted a survey to investigate ESL students’ reactions to teacher comments in multi-draft classrooms. She found that students paid more attention to teacher feedback provided on preliminary drafts than final drafts of their essays. She indicated that when students were required to revise previous drafts according to the teachers’ advice, they were more likely to pay close attention to the comments and corrections than when they received comments and corrections only to apply to a completely new essay assignment.
Having determined the benefits of multi-draft writing and revision on students’
attitudes toward writing and revision, many researchers tries to further prove their
effect on students’ writing quality through empirical studies (Chen, 1997, Ferris, 1995;
Paulus, 1999). Chen (1997), for example, indicated that multi-draft writing was found more helpful than single-draft writing in developing students’ writing ability and noted that revision provided an opportunity for students to see how a text which was previously thought completed can be further improved. Paulus (1999) found that required revision did significantly improve the essay scores of the class and suggested that teachers make revision, combined with meaningful peer and teacher feedback, an integral part of the writing classroom. Likewise, Ferris (1995) also confirmed the positive effects of revision on students’ writing with the help of teacher feedback.
However, not all researchers agree that revision leads to better writing quality.
An early study done by Hansen (1978), who compared two college classes on measures of “proofreading,” “editing,” and “general composition” skills, found revision did not bring about better writing quality and concluded that “rewriting is a waste of time.” Later, Bridwell (1980) reviewed several revision studies and
concluded that “questions about the relationship between revision and qualitative improvement remain largely unanswered” (p. 199). In her own study, Bridwell (1980) also found that the most extensively revised papers received a range of quality ratings from the top to the bottom of the scale. Moreover, Faigley and Witte (1981) noted that expert writers revise in various ways with some making almost no revisions and concluded that writing skill might be defined in part as the ability to respond to variables, such as the writers’ familiarity with the subject and the reason why the text is being written. Harris (1989) asserted that studies of revision did not show
convincing evidence that we should continue with multi-draft writing.
In conclusion, research has repeatedly proved the benefits of revision and multi-draft writing on students’ attitudes toward revision and writing. The benefits
include that it can help students see writing as rewriting and that it makes students pay more attention to teacher feedback. Yet, the effect of revision and multi-draft writing on students’ writing quality is still a controversial issue.
Students’ Difficulties in Revision
Three possible sources of difficulties in revision for students have been identified by researchers: first, students may misunderstand the revision task as a task of
changing words and sentences instead of modifying the goals or organization; second, students may lack necessary skills for successful revision; third, they are not able to handle teacher’s feedback on both form and content simultaneously.
Teachers and students usually have different interpretations of “revision.” While teachers may assume students to revise globally, students define revision differently from their teachers. Sommer (1980) studied the different revision strategies of student writers and experienced adult writers. She found that her student writers did not use the term revision or rewriting but developed some functional terms, such as “scratch out and do over again”, “Reviewing,” or “Redoing,” in which “the predominant concern is vocabulary” (p.382). Sommer suggested that the students misunderstood the revision process as a rewording activity. Other research has also found
inexperienced writers often revise locally and fail to improve their texts (Wallace &
Hayes, 1991; Faigley & Witte,1981; Bridwell, 1980). In her study of six advanced ESL college student writers, Zamel (1983) noticed that these writers understood that
“composing involves the constant interplay of thinking, writing and rewriting” (P.172).
They were aware that they could return to lexical and syntactic difficulties and that the “exploration of their ideas was of primary importance” (p.175)
Only after students have recognized that writing is a recursive and non-linear
plans … in order to meet their readers’ expectations” (Zamel, 1983, p.166). However, unskilled and beginning writers do not understand this process.
Another reason why revision fails is that students are unable to detect or correct their own errors. Hull (1987) studied student writers’ ability to correct a self-written essay and three essays written by others under two conditions: no feedback and feedback on location of error. He concluded that while the more skilled writers were able to correct more errors than the less skilled on either of the two types of essays, both groups performed better on the standard essays than on the self-written essays and with feedback than without feedback. On the self-written essays, neither of the two groups was able to correct many errors. These findings demonstrated that most students were not able to detect or correct their own errors and even skilled writers needed feedback when making revisions.
Hayes, Flower, Schriver, Stratman and Carey (1987) suggested that college freshmen often lacked the skills necessary for them to detect problems in their texts (as cited in Wallace & Hayes, 1991). Bartlett’s (1981) study with seventh and eighth grade students demonstrated that even when students do detected problems in the text, they still were incapable of fixing them, which necessitated extended training and practice in revision skills (as cited in Wallace & Hayes, 1991).
Another difficulty students encounter in revision is that they do not know how to respond to teacher’s feedback on both form and content at the same time. Zamel (1985) criticized the way feedback was provided: when teachers address both surface errors and content problems in the same version of a text, the suggestions are often contradictory in the sense that students do not know whether to see their text as a finished product and make the small-scale grammatical changes or to see it as
something still developing and make the large-scale changes on the content. Thus, she
advocates the multi-draft writing procedure, in which teachers can lead student writers to revise through the “cycles of revision” (p.95) by commenting on content in
preliminary drafts before switching to focus on form in later drafts.
To conclude, while revision has been proved as an important issue in writing, students do not revise successfully. The multi-draft writing procedure is regarded by some researchers as a good way to improve students’ revision quality and their attitudes toward revision and writing.
Teacher Feedback
Another line of research has focused on the role of teacher’s feedback as it is an important factor in a writing procedure. In both L1 and L2 writing classes, the teacher is traditionally the person who reads students’ writing (Nelson & Carson, 1998).
Teachers comment on students’ writing and provide a reader’s perspectives in the hope of improving students’ present essays or the subsequent ones.
The Effect of Teacher Feedback on Students’ Writing Quality
Researchers have investigated the effect of teacher’s feedback on students’
revision quality (Beach, 1989; Harris, 1989; Taylor, 1981; Zamel, 1982). It was assumed that only with the teacher’s specific requirement and direction would the students engage in major reconstruction. Sommers (1982) gives a vivid description about the importance of feedback:
Without comments from their teachers or from their peers, student writers will revise in a consistently narrow and predictable way. Without comments from readers, students assume that their writing has communicated their meaning and perceive no need for revising the substance of their text. (p. 149)
without feedback. Beach (1979) found that high school students who received between-draft teacher evaluation revised better than those who used guided self-evaluation forms or those who revised without either teacher or guided
assessment. Moreover, Fathman & Whalley (1990) demonstrated in their study that students’ revisions improved in overall quality and in linguistic accuracy when they received comments and/or corrections on both the content and form of their essays.
Inexperienced non-native writers need teacher feedback, more than advanced non-native writers or native writers, to improve their content, language and style (Koffolt & Holt, 1997). Ferris’ (1995) university ESL students found teacher feedback helpful in improving their writing. Taylor (1981) emphasized the importance of teacher feedback for L2 students. He suggested showing students through feedback on their own writing where their arguments were weak or where their logic broke down was even more effective than in-class instruction and analyses of writing examples.
On the other hand, some researchers argue that teacher feedback has little impact on student writing (Frantzen,1995; Hillocks, 1986; Makino, 1993; Polio, Fleck &
Leder, 1998). Polio, Fleck and Leder (1998) noted that when given additional time, learners could do self-correction and make improvements in their grammar and lexical choice even without feedback. They found that students who received additional editing instruction and feedback did not perform any better than those who did not on measures of sentence-level linguistic accuracy.
However, in her review of studies on teacher response, Ferris (1995) found that the above conclusion was questionable. She emphasizes that teachers’ responses to student compositions are most effective when given on preliminary or intermediate drafts rather than final drafts of student essays. In addition, Freedman (1987) and Krashen (1984) argue that “students will not pay as much attention to teacher
feedback when they are not required to do revision than when they are required to think over teacher feedback and revise” (Ferris, 1995, p. 34).
Teacher Feedback on Grammar
Many writing instructors are not sure of the best type of feedback for their students (Leki, 1990; Paulus, 1999). One of the most important questions asked is:
Should teacher’s feedback focus on form or on content? Proponents to process-writing approach have been advocating that teachers should respond as an interested reader and their feedback should focus on content rather than aim at error correction (Sommers, 1982). Some empirical evidence has even found teachers’ comments on grammar correction ineffective or even “useless” (Truscott, 1996, 1999) on the linguistic accuracy on L2 student writing over time.
Other researchers, on the contrary, have recognized the importance of grammar feedback for students’ writing. Ashwell (2000) claimed accuracy was “an important aspect of the final product that, if not attended to, will negatively affect
communication” (p.227). Fathman and Whalley (1990) demonstrated that while students were able to significantly improve their content simply by rewriting even without feedback, they needed teachers’ feedback on grammar errors to enable them to make significant improvement in grammatical accuracy. More recently, Yates and Kenkel (2002) reviewed articles concerning teachers’ comments and error correction in L2 writing and rebutted Zamel’s suggestion of focusing more on meaning and less on language:
We think this prescription is misleading: L2 writing instruction cannot be divorced from L2 language instruction because it is the L2 students’ lack of knowledge about the language to achieve their writing purposes which makes responding to actual L2 writing so difficult, yet so important. (p. 46)
In conclusion, most advocates of process writing approach seemed to agree that teachers’ comments must attend to both content and form while insisting the priority of instruction must be given to content rather than form (Ashwell, 2000; Sommers, 1982; Zamel, 1985). Zamel (1985), for example, suggested teachers should help students understand that “meaning-level issues are to be addressed first” (p.96).
Teacher Feedback in Multi-draft Writing
Zamel (1985) and Sommers (1982) advocate the multi-draft writing practice, in which students are required to write at least three drafts for an essay and teachers to comment on different issues of problems in different drafts. It is, however, not clear whether such content-then-form practice is indeed more effective in improving students’ writing quality than other patterns. Fathman and Whalley (1990) studied four types of teacher feedback (no feedback, content, form, content and form) and concluded that giving content and form feedback simultaneously is just as effective as giving content feedback or form feedback separately. They claim, therefore, that multi-draft practice may not be necessary because content issues and form issues can be dealt with at the same time. Similarly, Ashwell (2000) studied four types of feedback in a Japanese university: (1) the recommended pattern of content feedback followed by form feedback (2) the reverse pattern (3) a pattern of mixed form and content feedback (4) control group: no feedback at either stage. He found that the recommended pattern of content feedback followed by form feedback was not
superior to the reverse pattern or to a pattern of mixed form and content feedback. The mixed pattern even exhibited an advantage over the other two patterns, if only in terms of simple mean gains in accuracy ratings and content scores. Ashwell’s study seems to offer a new perspective to the order of teacher response and revision. The finding challenges Zamel’s recommendation to give content feedback on early drafts
and form feedback on later drafts. It also challenges the idea that the two types of feedback should not be mixed.
Direct and Indirect Teacher Feedback on Errors
Another concern in feedback research is how explicit should the comments on errors be. In Ferris & Roberts’ (2001) review, they defined two types of feedback:
“direct feedback” and “indirect feedback.”
Direct feedback is given when the teacher provides the correct form for the student writer; if the student revises the text, s/he needs only to transcribe the correction into the final version. Indirect feedback occurs when the teacher indicates in some way that an error exists but does not provide the correction, thus letting the writer know that there is a problem but leaving it to the student to solve it (p. 163).
They found that indirect feedback was more preferable because it required students to reflect and thus “foster[s] long-term acquisition” (p. 164). They also suggested that indirect feedback “helps students to make progress in accuracy over time more than direct feedback does” (p. 164). In their study of three types of indirect feedback on errors: (underlined and coded, underlined but not coded and no error markings), they found that students whose errors were marked could edit more successfully than those who received no error markings, but students who received more explicit (coded) indirect feedback on errors did not outperform those whose errors were simply underlined, not coded. They recommend that if teachers ask students to self-edit their papers with the help from teacher feedback, they may consider, for some students’ compositions, to locate errors but not label them by error type. In a similar study, Makino (1993) also suggests that detailed cues, not correction, can improve learners’ abilities to self-correct grammatical errors.
Effective and Ineffective Teacher Feedback
Given the importance of teacher feedback, how can it be accepted by students and applied effectively in their revisions or subsequent compositions? Some
researches have found students not attentive to teacher comments due to vagueness, the negative quality of the comments, or the undue emphasis on error correction (Cohen & Cavalcanti, 1990; Leki, 1990; Hillocks, 1982). Conrad and Goldstein’s (1999) study explored three students’ revisions after written comments during a university-level ESL writing course. They analyzed the comments and the students’
subsequent revisions of texts and found that students generally preferred “longer comments, especially those that explain specific problems and make specific
suggestions” (p. 148). Ferris (1997) found that ESL students revised most successfully in response to comments that requested specific information or provided summary grammatical evaluation. They revised, on the other hand, less successfully when the teacher’s comments were given in question form, asked students to deal with
problems in logic or argument or asked questions for further information (Conrad &
Goldstein, 1999). Ferris concluded the lack of success in revision might have resulted from the fact that ESL students were still in the process of language development and thus need more specific and explicit suggestions.
Another factor that makes teachers’ comments ineffective is the negative tone in the comments. Although Sommers (1982) did not specifically state what caused the ineffectiveness of teacher comment, she described the majority of teachers’ comments as “hostile and mean-spirit” (p. 149) in her comparison of computer comments and teachers’ comments:
The sharp contrast between the teachers’ comments and those of the computer
highlighted how arbitrary and idiosyncratic most of our teachers’ comments are. Besides, the calm, reasonable language of the computer provided quite a contrast to the hostility
and mean-spiritedness of most of the teachers’ comments. (p. 149)
Cardelle and Corno (1981) studied four types of feedback and discovered the factors that affect the choices students made as they revised. They concluded that giving a combination of criticism with praise was more effective than giving criticism alone. Nelson and Carson (1998) in a more recent study, however, found that students actually preferred negative comments that showed them where their problems were.
To sum up, researchers have generally agreed that grammar should be addressed but not overemphasized in teacher feedback and most research has found that indirect feedback on errors is more effective than direct feedback and a mixed feedback of criticism and praises is more helpful to students’ writing than either criticism or praises alone. Yet, whether multi-draft feedback is more effective than single-draft feedback is more effective in improving students’ writing quality is still to be determined.
Peer Review
Besides teachers, peers can also provide audience perspectives in composition classes. Peer review, in which students read and respond to drafts of their fellow students’ essays, has been an alternative form of feedback to teacher feedback in a process-oriented curriculum in many L1 writing researches over the last decades and in L2 writing researches in recent years (Min, 2003; Mangelsdorf & Schlumberger, 1992; Nelson & Carson; 1998; Paulus, 1999; Tsui & Ng, 2000).
Advantages of Peer Review
Many researchers’ empirical evidence has been encouraging writing teachers to incorporate peer review into both L1 and L2 writing classes to complement teacher
feedback. Five major beneficial effects of peer review have been outlined by a number of researchers in L1 and L2 writing.
To begin with, peer review is found to enhance audience awareness and enable students to see egocentrism in their own writing (Min, 2003; Mangelsdorf, 1992;
Mittan, 1989; Paulus, 1999; Tsui & Ng, 2000; Zamel, 1982). When compared with teachers, peers were more likely to be interested readers (Munice, 2000) and thus peer review may bring a genuine sense of audience into the writing classroom (Paulus, 1999). This consideration to audience perspectives can ultimately develop in students the crucial ability of “re-viewing their writing with the eyes of another” (Zamel, 1982, p.206).
The activity of peer review can also encourage collaborative learning (Tsui & Ng, 2000). With the same register, students can communicate more easily with their peer reviewers than with the teacher. Min (2003) and Tsui and Ng (2000) both suggested that the assistance of supportive peers could not only improve learners’ attitudes towards writing but also alleviate their apprehension.
In addition, peer review can help develop students’ critical reading and analysis skills (Paulus, 1999). Through discussing and questioning of each other’s drafts, students learn to analyze different texts and sharpen their judgment of what makes writing successful (Paulus, 1999). As one student in Mangelsdorf’s (1992) study reflected, “It was … very valuable to review other students’ papers. It helped me get a better overlook on how to write successfully—in organization and content” (p. 278).
As students critically read others’ writing, they also learn to read their own writing critically and thus become more active readers and writers (Mittan, 1989).
Finally, peer review exposes students to different perspectives and a variety of ideas (Koffolt & Holt, 1997; Mangelsdorf, 1992; Min, 2003). Mangelsdorf (1992)
studying 40 advanced ESL university freshmen’s attitudes toward peer feedback, found that content and organization were the main areas that peer reviews improved.
Disadvantages of Peer Review
Despite repetitive reports of the beneficial effects associated with peer review, some researchers argue peer review is not always effective and practical. The potential for peer review to become “an unproductive experience” can discourage writing teachers from adopting it as a regular activity in their class (Paulus, 1999, p. 268).
Some researchers have criticized that peer comments fail to be incorporated into students’ subsequent drafts. Min (2003) pinpointed that studies focusing on measuring the outcomes of revision based on peer review did not reveal encouraging results despite the advantages of peer review. Then, why do students not incorporate peer comments into their revision? Nelson and Carson (1998) investigated Chinese and Spanish-speaking students’ attitudes toward their interactions in peer response groups in an ESL advanced composition class and both groups commented that the feedback was not always helpful. Mangelsdorf (1992) investigated ESL students’ perceptions of peer review and found that even students who had some positive reactions to peer reviews preferred teacher feedback as one of her students commented:
The problem with peer reviews is that most of the time the reviewer isn’t critical enough as the instructor would be, so not much help can be obtained from the reviewer. (p. 280)
Research has found that students’ advice was not as comprehensive as teachers’
advice. Some studies indicate that peer review only focuses on linguistic errors while others indicates that students can give advice only on meaning-level problems. Leki (1990) noticed that students’ advice often failed to facilitate revision because it tended to focus on surface errors, not semantic or textual ones. Connor and Asenavage (1994)
The study revealed that in the total revisions only 5% were influenced by peer
comments while 35% were influenced by the teacher’s comments and that 70% of the peer-influenced changes and 22% of the teacher-influenced changes were concerned with meaning-level problems. It may suggest that most students could only comment on meaning level problems because they were not able to correct linguistic errors.
Another reason for the failure to incorporate peer’s comments is students’
difficulties in deciding on the validity of their peers’ comments. Tsui and Ng (2000) pointed out this problem was especially acute in L2 context. L2 students may not trust their peers’ suggestions because they are not native speakers; they are still language learners themselves (Paulus, 1999; Lockhart & Ng, 1993). Mangelsdorf (1992) discovered that almost all of the students that held either mixed or negative attitudes toward the peer-review task mentioned students’ ignorance and incapability of giving reliable advice. Nelson and Carson (1998) pointed out peer comments might not be as effective in L2 context as in L1 context. While native speakers, who have greater knowledge of English and more confidence in their language abilities, can devote their attention to develop their writing skills, nonnative speakers have to attend to both language and writing. Mangelsdorf (1992) further suggested that students from some cultures might resist such a student-centered task. She found that almost all of her students with totally negative views came from cultures that preferred
teacher-centered environment, such as Asia. Tsui and Ng (2000) studied the roles of teacher and peer comments in revisions in writing among secondary L2 learners in Hong Kong. They found that students from cultures that regarded the teacher as the only source of authority may consider their peers not qualified to make comments.
Nelson and Murphy (1993) also saw students’ cultural belief as a major reason why students didn’t accept peers’ comments. They found that for students with a strong
belief in authority figures, peer comments were not reliable.
Not only do students lack trust in their peers as reviewers, but the peer reviewers themselves also lack confidence in themselves as competent readers (Tsui & Ng, 2000). In fact, L2 students of lower proficiency do have difficulty in giving sensible comments. They may fail to comprehend their fellow students’ text and thus give inappropriate suggestions (Mangelsdorf, 1992) or vague suggestions (Mangelsdorf, 1992; Min, 2003). On the other hand, when students do have the capacity to give suggestions, they may also deliberately give vague comments “for fear of hurting the writer’s feelings” (Min, 2003, p.98).
In addition to the problems concerning the effect of peer review on students’
writing, some researchers also question the practicality of putting peer review into practice in a school context. Mangelsdorf (1992) studied the perspectives of 40 advanced ESL writing students. She admitted that reading sample drafts, practicing how to make suggestions for revision and discussing the peer feedback students received all take a great deal of class time. Nelson and Carson (1998), who
investigated ESL students’ perceptions of the effectiveness in peer response groups, also noticed that both the Chinese and Spanish students felt that too much time was
“spent talking about unimportant issues” (p. 125) and one student even described such situation as a waste of time. Another problem of peer review that Mangelsdorf (1992) identified was the difficulty in deciding how to group students:
For the most part, I try to group students who are at similar ability levels and who are writing on closely-related topics. When students vary a great deal in ability, usually the better students give good feedback to the weaker students but get little feedback in return. Similarly, when students know next to nothing about a fellow student’s topic, they tend to offer few ideas about the content. (p. 282)
Whether peer review helped out, according to one of Mangelsdorf’s students,
depends on who the reviewer was. Although still hopeful about peer review, Mangelsdorf, in her conclusion, remarked, “every class reacts differently to peer reviews, some eagerly, some reluctantly” (p. 282). Nelson and Murphy (1993) had similar results in their study. They suggested that although students did incorporate their peers’ suggestions into subsequent drafts, the amount of incorporation greatly depended on a cooperative environment among the group members. Moreover, researchers have indicated that peer review process was extremely complex and, to make it successful, teachers need to train students necessary review techniques and contruct a collaborative environment (Mangelsdorf, 1992; Paulus, 1999).
Despite the above disadvantages, peer review is still accepted as a valuable activity in a writing classroom. To reduce potential drawbacks, teachers should carry out the activity with care. They may also need to practice it with flexibility to suit different students’ age and language proficiency as well as various class atmosphere.
So far, researches on L2 peer review, whether focused on the process or the outcome, mostly involved ESL university students (see examples in Mangelsdorf, 1992; Nelson & Carson, 1998; Paulus, 1999) or EFL university students (see Min, 2003). Most of the ESL subjects are students with at least an intermediate level of language skills and the EFL subjects are mostly university students majoring in
English. Very few studies have investigated low-proficient or less mature EFL English learners.
Comparing Peer Feedback With Teacher Feedback
Both teacher feedback and peer feedback have their advantages and are
recommended by many researchers. However, when comparing the effectiveness of these two types of feedback, researchers are more reserved in the use of peer feedback, especially in L2 context. Nelson and Carson (1998) confirmed the effectiveness of
teacher review in multi-draft settings but expressed doubts about the value of peer review. Many studies have shown that students preferred teacher feedback to peer feedback and they incorporated more teacher comments in their revisions than peer comments (Neslson & Carson, 1998; Tsui & Ng, 2000; Zhang, 1995). Paulus (1999) suggested teachers make peer feedback and teacher feedback an integral part of the multi-draft writing classroom. However, he also showed that while both teacher and peer feedback were used by students, teacher feedback was clearly adopted more often than peer feedback.
Summary of the Reviewed Literature
In this chapter I first cover discussion of research on the transition of two writing approaches, from product approach to process approach. Then, I review the effects of different features of instructional context on writing, including the instructional emphasis, teacher feedback and peer feedback. Next, I review the effects of revision on writing, discussing conflicting results in research on the effect of revision on students’ attitudes and writing quality. Finally, I review the reasons for students’
difficulties in revision.
