COVER PAGE
The Effects of L2 Writing Revision Tasks on Common Non Correct Form (error and mistake) Frequencies among Third Year Burapha University Students
Richard Anthony O’Donnell
This research was supported by the Faculty of Humanities and Social Sciences, Burapha University
TITLE PAGE
Burapha University Department of Western Languages
Faculty of Humanities and Social Sciences
The Effects of L2 Writing Revision Tasks on Common Non Correct Forms (error and mistake) Frequencies among Third Year Burapha University Students.
Richard Anthony O’Donnell Submitted March 11th 2013
ABSTRACT
This study aimed to find out if engagement with revision tasks in
class journals would reduce common Non Correct Forms (errors and
mistakes) in student writing. The study followed a quasi-experimental before and after design with a control group and an experimental group. The experimental group engaged with revision tasks in their
class journals. The research questions were: 1 Is there a significant difference in NCFs before the treatment between the two groups? 2 Is there a significant difference in NCFs after the treatment between the two groups? 3.
Is there a significant difference in NCFs the
experimental group before and after the treatment? An error analysis was completed before and after the treatment. An SPSS independent sample t test showed there was no overall significant difference between the control group and the experimental before the treatment. Overall there was a significant difference between the control group and the experimental group after the treatment (p = .001). In other words, the control group produced significantly more NCFs than the experimental group. Moreover, there was a significant difference in the experimental group before and after the treatment (p = .000),
highlighting three areas of improvement: missing verb to be, part of speech, and near synonym. Engagement with revision tasks may have helped students to ‘notice’ their NCFs.
ACKNOWLEDGEMENTS
I would like to offer my sincerest and deep-felt thanks to my beautiful
wife, Laphatrada O’Donnell, for helping me with this research. Without her continual support and encouragement I could never have completed this paper.
Special thanks Dr Somboon Chetchumlong for his support and encouragement, and to all the staff at the Department of Western Languages, Burapha University, Thailand.
CONTENTS
Introduction .............................................................................................................................. 1 Research questions ................................................................................................................. 3
Literature Review .................................................................................................................... 3 Treatment - journals ............................................................................................................... 3 Treatment - Revision .............................................................................................................. 5 Treatment – learner strategies ............................................................................................... 8 Conceptualization - cognitive approaches to writing and revision ..................................... 12 Data – error analysis ........................................................................................................... 17 Procedural difficulties in doing an error analysis ............................................................... 19 Methodology ........................................................................................................................... 20 Work conditions.................................................................................................................... 20 Course-book ......................................................................................................................... 20 Design of the study ............................................................................................................... 21 Independent variables. ......................................................................................................... 21 Dependant variables ............................................................................................................ 23 Extraneous variables ............................................................................................................ 23 Participants .......................................................................................................................... 24 Results ..................................................................................................................................... 24 Research Question one ......................................................................................................... 24 Research Question two ......................................................................................................... 27 Research Question three ...................................................................................................... 30
Summary of the results ......................................................................................................... 34 Evaluation and Discussion .................................................................................................... 34 Suggestions for further study................................................................................................ 38 Pedagogical implications ..................................................................................................... 39 References ............................................................................................................................... 40 Appendix 1 .............................................................................................................................. 47 Appendix 2 .............................................................................................................................. 49
Introduction Writing in a second language (L2) is no easy task. Given a highly motivated learner, the process of becoming a competent L2 writer may take many years of sustained study (Hyland, 2003; Kellogg, 2008). Writing in a second language is perhaps one of the most difficult skills to learn (Ramires, 1983; Hyland, 2003). To illustrate this point, Cumming (2001) states that to write well learners have to grapple with micro factors and macro factors. Micro factors include syntax, the way an English sentence is constructed and the options available in the construction; and morphology, the method of creating lexemes with available suffixes and prefixes. The macro level includes discourse structure, cohesion, planning, revision and
editing. And all of the above are in themselves complex. Therefore, it is no surprise that second language writers will make Non Correct Forms (NCFs) i.e. ‘errors and mistakes’
when they write. What is an NCF? A definition is given by Lennon (1990). An NCF is ‘a linguistic form or combination of forms which, in the same context and under similar
conditions of production, would, in all likelihood, not be produced by the native speaker’s counterparts’ (Lennon, 1990). In this definition, NCFs are seen in linguistic items and not as originating from style or organization.
There may be a difference between errors and mistakes, so Non Correct Forms (NCFs) is being used in this study as terminology to cover both. Some researchers, Corder (1981) and Richards (1980), advocate a crucial distinction between error and mistakes. Errors are a result of a learner not fully comprehending the language system. In this scenario they utilize an inter-language (Richards, 1981). Errors are representative of incomplete knowledge of the target language (Ellis, 1994). Despite the negative semantics of the term ‘error’, they should not be seen in a negative light. Errors are evidence of an intermediary level of comprehension on the path towards command of the target language. Mistakes, on the other hand, are defined by Corder (1981) as ‘failures to utilize a known system correctly’ (Corder, 1981). Mistakes occur when a learner has command, but simply forgets to apply his or her knowledge. Mistakes are slip ups. They might be caused by tiredness, forgetfulness, emotional state, time constraints and conditions under which the writing takes place (Dulay, Burt, & Krashen 1982). In this paper, Non Correct Forms will appear as NCFs to cover both errors and mistakes. By definition, an NCF is an error or mistake recognized as such by an educated English speaker/ teacher.
1
NCFs made by Burapha students are not very much different from those made by students in other universities in Thailand (Sattayatham & Rattanapinyowon, 2008; Hemchua & Schmitt, 2006; Chakorn, 2005). The common problem areas are concerned with nouns: subject verb agreement and nominal quantifiers; verbs: tense usage, missing verbs and trouble with infinitive forms; lexical items such as wrong semantic word choice or part of speech. Most teachers will attest to the fact that students make the same errors again and again, and this is understandable given the complex nature of writing. One possible solution to this problem is to have the students work with language learning strategies as a means of revision. In this paper, the term ‘language learning strategies’ is synonymous with the term ‘language learning task’.
The point of this study is to look at one possible way in which my students at Burapha
University can begin to reduce the number of non correct forms (NCFs), errors or mistakes,
they are making when they write. One possible way is to make use of L2 writing strategies (tasks) as a means of revising. L2 writing strategies can be applied to revision in a process in which the student writes down in his or her class journal the results of completing the writing strategy. The student identifies a non correct form (NCF) in his or her writing and then applies an L2 writing strategy, the results of which are then written down in their class journals. The student’s attention could be drawn to completing a written record of their revision, a semester long written involvement with revision. By keeping a semester long, class revision journal while engaged with L2 writing strategies, might the students reduce their NCFs? One of the main issues for teachers of L2 writing is what students should do with their errors i.e. how should they revise? Given that they have noticed the teacher corrections, what will they actually do with this input as a means of revision? In my class, students were required to write a second draft, so their involvement was a process of looking at my corrections then improving their drafts. Given the effort required to write well in a second language, is it enough to just revise drafts? When dealing with quantifiers in a nominal group, is correcting it one time necessarily the optimum amount of attention needed? Might it not be a good idea to have students write more and pay more attention to their NCFs by doing something more substantial than simply adhering to teacher corrections between drafts?
2
In order to investigate this, a quasi-experimental design study was set up involving an experimental and a control group. The experimental group, my class, was required to take part in a treatment involving L2 writing strategies in a class journal as a way of revising. The control group did not experience this treatment. Before the semester, an error analysis was conducted on both groups, and after the semester another error analysis was conducted on both groups. The experimental group consisted of 22 fourth year Education major students at Burapha University, 18 females and 4 males, who were taking Essay Writing, a required course in the curriculum. The control group consisted of 22 fourth year English major students, 17 females and 5 males, also taking Essay Writing, the same course with the same book within the same time frame of the semester. Both groups were 19-20 years old. Both
classes were taking an essay writing class, a sixteen week required course in the English major programme. Both classes had taken the same writing courses during their first three
years; basic writing, paragraph writing and essay writing.
Research questions In this study, I hope to answer the following questions: 1 Is there a significant difference in NCFs before the treatment between the control group and
the experimental group? 2 Is there a significant difference in NCFs after the treatment between the control group and the experimental group? 3. Is there a significant difference in NCFs in the experimental group before and after the treatment?
Literature Review This literature review is divided into three main issues which surround the core of this study. Related to the treatment is discussion of journals, revision, and L2 writing strategies. Related to the conceptualization of the study are cognitive approaches to writing and revision. Related to the data is the concept of an error analysis, a principled method of categorizing and counting student NCFs.
Treatment - Journals Journal writing can be seen as offering students a break from course-book writing exercises. Journals may serve as a pressure-free space to write in, away from the anxieties of grades and 3
assignments (Elbow, 1999). Casanave (1994) and MacGowan-Gilhooly (1991) also advocate journal writing as a means to reduce anxiety and increase student confidence. Having the opportunity to write in a journal provides the students with an outlet for writing that is not based on the traditional concepts of paragraph structure, like topic sentences, supporting sentences and conclusions, which may be confusing to second language writers. Journals provide an opportunity to deviate from course-book exercises based on analyses of structure, sentence combination and model paragraphs. The journal can also encourage more writing from the students. Free writing is commonly
taken to mean writing which leans towards fluency and less concern for accuracy (Silva, 1990). Topics for free-writing could include keeping a diary, writing a story, writing down
reflective thoughts about learning or observations (Peyton & Staton, 1991). Zamel (1985)
suggests that journals ought to record daily events and personal thoughts. Ediger (2006)
wants to encourage students to become owners of text. By giving a free choice as to what students write in their journals, writing becomes individualized. This is a very different sort of writing from what they would do in class if only course-books were used. One point about free-writing is that it does not have to be checked by the teacher. It could be, but the idea is to let the students develop as writers through personalized writing. Another view might be to say that asking students to write diaries and such like without giving teacher feedback is a waste of time. However, some studies show the opposite. Hillocks (1984) in a meta-analysis of studies, states that free writing can provide slight but positive improvements, and as such is a potentially useful technique in the writing class. In some studies, Hillocks (1984), claims that free writing produces better work than explicit grammar instruction. In another study, Hashemi and Amerian (2011) looked at one class divided into control group and experimental group. The authors utilized a pre-test and posttest design to examine whether free writing in journals would improve grammatical accuracy. The results indicate positive results for free writing in journals. In addition to this, journal writing may increase student confidence in their ability as writers. In a case study, Hwang (2010) suggests that students gain confidence as they write in journals. It is not controversial to suggest that most writing done by L2 students is cloaked in anxiety, whether for grades or performance levels. Based on my own classroom observations and ten years of teaching L2 writing, most low level writers are faced with numerous 4
problems when writing, and when they receive feedback, the proverbial red ink, they may feel slightly discouraged. Interestingly, most teachers would not correct every mistake or error in student speech and yet they do with writing. This is due to the social expectations put on writing. In addition to free writing, it might be possible to add another component to student journal writing. Might it not be possible to have students focus on their common NCFs (errors or mistakes) in their journals by utilizing Oxford’s (2003) writing strategies? By focusing on NCFs through L2 writing strategies, might the students learn to reduce the number of
common NCFs? The experimental group in this study, my class at the time, was being encouraged to match up their free writing with more focused revision writing. Revision
writing means engaging with L2 writing strategies in their journals, with some focus on
freewriting in the form of diary entries.
Treatment - Revision Kietlinska (2006), reports that ‘…existing research on revision patterns of ESL students is rather limited and tends to focus on various forms of feedback’. In addition to Kietlinska’s
(2006) misgiving concerning research into revision, Plonsky (2011) points out that research into revision has been ‘inconsistent’ and few studies have the necessary controls to attain a more qualitative methodology. The issue of teacher feedback has a long and somewhat controversial history. There has been much debate on the usefulness of teacher feedback. Ellis (2009) identifies five basic ways to give feedback. Direct feedback: the teacher provides the correction explicitly to the student. Indirect feedback: the teacher indicates there is a NCF, but does not provide the correct form; indirect feedback often takes the form of underlining. Meta-linguistic feedback: the teacher uses symbols to indicate the NCF. Meta-linguistic feedback can also include explanations of the NCF. Electronic feedback: as it suggests, the teacher can provide hyperlinks to grammar explanations. Lastly, reformulation: this involves the teacher re-writing the entire student text to represent what a native speaker might produce. One of the main advocates for teacher feedback, Ferris (1997a, 2004b), argues that it is essential. However, researchers such as Truscott (1999a, 2006b) argue the opposite and say it does not lead to improvement. Whatever the merits of each position, most students want teacher feedback, institutions require it and teachers mostly feel that giving feedback on 5
NCFs is part of their duty as a writing teachers. Recently, a meta-analysis of feedback studies conducted by Biber, Nekrasova and Horn (2011) indicate that teacher feedback does appear to improve L2 writing, however slightly. This is good news for teachers who spend long hours giving feedback to students. As this paper is not dealing specifically with the issue of teacher feedback, no further discussion will be offered on this issue. Suffice to say, teachers have to be aware of both sides of the argument. Revision suggests something that comes after an event, but with writing it can also occur while the text is being produced. Revision in this study is used to mean anything the student
does to improve the content and form of his or her writing, either while writing or after writing. As the students in this study were engaging with recent teacher feedback, the revision focused on after writing.
Most of the literature surrounding revision is concerned with what L2 writers do when they write. What kind of revision is seen? Low-level writers tend to focus on surface errors. Horning and Robertson (2006) describe Low-level writers as those who have no formal experience of L2 writing, have little exposure to L2 texts or text genres, have numerous punctuation problems, and commit many errors i. e. an abundance of nouns and verb
problems. According to Kietlinska (2006), Low-level L2 writers tend to create text, read, revise, create text, read, and revise in a manner that can be described as choppy. I have certainly seen this in my classes; student writing is slow and it seems a struggle just in terms of actually physically doing some writing. The reason for the more laborious method of choppy editing seen in low level L2 writers may be due to the students trying to ‘construct writing as they go’, in addition to worrying about making errors. It is all very well for teachers to ask students to just write quickly and not worry about grammar at first, but this is not always an easy option for low level writers. On the other hand, higher-level L2 writers tend to write the complete text then read then revise the whole text. Successful writers also monitor what they are revising. They also rewrite extensively (Kobayashi & Rinnet, 2001). In a classic case study, Zamel (1983) observed competent L2 writers completing an assignment. Zamel’s study, which is widely quoted in the literature, argues that advanced L2 writers write in a non linear way i.e. the students in Zamel’s study did not appear to go through a three step process of pre-writing, writing, revision, but rather did all three at once as they wrote. In addition to this, Zamel 6
found that advanced students discarded text without worry, and produced more text. Furthermore, it is also found that advanced writers reread what they have written continuously. What needs to be taken into account, however, is that each individual student carries with her or him different cognitive ability, varying degrees of motivation, separate cultural beliefs about learning, and different overall language learning styles and abilities. Taking this into account, successful L2 writing learners, on the whole, tend to be willing to make mistakes and learn from them, willing to work hard on eradicating and understanding their mistakes,
and willing to read extensively along with writing extensively. They should also be prepared to focus on extensive practicing of the language. In some ways, this should not come as a
surprise. After all, rarely does someone become successful at anything without the above. It
would indeed be surprising if a student was a competent L2 writer but never had to try very
hard. Some researchers, Hyland (2003) and Raimes (1983), state that perhaps the most crucial element is motivation. Many teachers might acknowledge this as being crucial. Students who are not motivated to learn seldom do learn. L2 writing teachers should therefore try and motivate students as best they can and, perhaps, refrain from saying how difficult it is to write. Many researchers, Kobayashi & Rinnet, (2001); Hyland (2003); Kietlinska (2006), have attempted to categorize various types of revision available to students. Various revision strategies are reading aloud, using a dictionary, making changes to sentence structure, and changing the content. In addition, revision can mean peer revision, a process in which classmates look over each other’s writing for NCFs. This method is somewhat controversial in that some studies, Kollar & Fischer, (2010), say it leads to improvement and others report no improvement. One common argument against peer revision is that students sometime feel that classmates who make NCFs themselves are hardly in a position to revise. However peer revision in class can lead to some changes for the better (Villamil & De Guerrero, 1996; McGroarty & Zhu, 1997; Kollar & Fischer, 2010). In an early study, Faigley and Witte (1981) attempted to categorize types of revision available to L2 writers. The authors distinguish between the effects of the revision and the nature of the revision. Their suggestions include: Additions, Deletions, Substitutions, 7
Permutations, Distributions, and Consolidations. Additions emphasize what can be inferred. For example, in this made up example: She gave this ring to me could be revised to say: She gave me this ring for my birthday, adding information which might be inferred; inferring here meaning guessing. A reader could guess, perhaps, but by adding information I make clear what may be inferred. Deletions are the opposite. I revise my text to delete information. Substitutions: revising texts by substituting original words with synonyms; The old Chinese pottery is amazing is revised to say Those ancient Chinese bowls and cup are incredible. Permutations: rearranging clauses and prepositional phrases: Last night I went with my gang to my best friend’s birthday party changes into My best friend’s birthday party was last night and I went with my gang. Distributions occur when the revision takes one unit of information,
a sentence, and stretches it into two or more sentences. Consolidation is the opposite: I saw him. He stood near his car. He had his sports bag; this might be revised to say: I saw him
standing with his sports bag near his car. These suggestions by Faigley and Witte (1981) are a useful way of conceptualizing what form revision can take. It might be worthwhile pointing out to L2 learners that expert L1 writers revise extensively. Professional writers often write multiple revisions on their work. One method of showing this may be to show students pictures of drafts done by famous novelists. What they will see are pages covered with revision notes, often with whole blocks of text marked up for deletion. This might drive home the point that revision is not just something L2 learners do. From my own experience of reading about this topic, all the well known novelists, from Hemmingway to Rowling, revise extensively. Indeed, authors are often revising right up to deadlines, changing this or that, improving this paragraph or that paragraph. The only well known writer who did not revise much was the British philosopher Bertrand Russell. However, he was an exception to the rule.
Treatment – learner strategies There are many definitions of learner strategies in the literature (see O'Malley, Chamot, Stewner‐Manzanares, Kupper & Russo, 1985; Lessard-Clouston, 1997). Rather than going into different definitions, I have decided to stick with Oxford’s (1990) definition. Oxford (1990) refers to LLS (Language Learning Strategies) as follows: actions, steps, techniques and behaviors which foster learner autonomy and enhance student learning. This appears to be a workable definition. The important aspect for Oxford (2003) is that learners need to engage with their learning through actions. In other words, L2 writing strategies are not just 8
something students should simply think about, but something they should put into action if they wish to improve. Anderson (2005) states that since the 1970s, researchers have been looking at taxonomies of language learning strategies (LLS). A consensus is reached with the following strategies: memorization, clarification, communication, monitoring, and prior knowledge. Oxford’s (1990) well known and often used SILL inventory contains these headings in the form of memory, cognition, and comprehension. Oxford (1990) distinguishes between direct strategies and indirect strategies. Direct strategies include those that facilitate memory, cognitive strategies and comprehension strategies.
Memory strategies, as it implies, are tasks which may help students memorize various aspects of their learning. These include applying mental images to areas of writing, or drawing such
things as spider-grams. Cognitive strategies are those tasks which enable learners to manipulate and transform their production of the target language. Cognitive strategies, it can be said, are ‘hands on’ tasks; students change or work with their writing in a focused manner. When students practice their L2 writing, they engage with cognitive strategies. When students use techniques such as repeating i.e. re-writing non correct language as correct
language, they are engaged cognitively with the target language. Compensation strategies are designed to overcome student limitation and include such actions as guessing word meaning from context. Oxford’s (1990) indirect strategies include metacognitive aspects to learning. Metacognitive strategies are less ‘hands on’ and are more concerned with planning, evaluating and coordinating learning. It should be stressed though that there is no clear, delineated line of difference between cognitive and metacognitive strategies; these are terms of convenience only, as Oxford (1990) states in her book. Indirect strategies also include affective aspects of learning, connected with emotional proclivity. It is perhaps easy to see the difference between cognitive and metacognitive when we explicitly look at examples: when students translate they are engaged cognitively, and when they draw up a plan of study and monitor what they are doing they are engaged meta-cognitively. Oxford (1990) claims that both types of strategy are essential for learning and should be used in tandem. The following, adopted from Oxford (1990), is a small example of Memory and Cognitive strategies, and how they can be used in L2 writing revision. It should be noted that Oxford’s 9
(1990) strategies are concerned with all four skills, not just writing. However, in her book, some strategies designed either for writing or for other skills can be transferred to writing. The following were given to the experimental group in this study as revision tasks. Tasks are defined as activities in which students act on and solve problems. MEMORY STRATEGIES 1. Placing corrections in a different context: This could include placing quantifying corrections, for example, in contexts other than the original work. If a student writes ‘many of people’ and sees the correction ‘many of the people’, she or he could practice writing many of the students, many of the boats, many of the teachers in full sentences or just clause form in
their journals. This may help to focus attention on the NCF by applying it to another context. 2. Grouping: This refers to creating meaningful units; words could be grouped according to
word families or part-of-speech; students could group assignment errors according to the type. For example, students can keep all pronoun errors together in the same section of their note-book. 3. Associating/Elaborating: This could include comparing the mother language with the L2
language, comparing what is similar or what is different. When an NCF is made which can be traced to Thai then the student has to make note of this and write down several examples. 4. Semantic Mapping: Drawing mind maps to help remember grammar rules and vocabulary. Semantic synonym maps can help students to group synonyms according to shades of meaning. COGNITIVE STRATEGIES 5. Repeating: This means doing something over and over again. This could involve writing the correction several times. This may appear simple and somewhat redundant, but experts in L2 writing, Kellogg (2010), claim that repeating correct language is essential. 6. Formally Practicing: Students need overt practice. Oxford (1990) suggests copying correct English from a book or magazine or the Internet. This gives students practice in writing correct English and gives them a feel for what it looks like on paper. Again, this may strike some as being simple and very basic, but it is deemed essential by Oxford. Sometimes the things the students need to do are simple and basic, but important. 10
7. Recognizing Patterns: This is Oxford’s (1990) tip of the hat to lexis, fixed phrases, single lexical units and the like. It may include verb plus preposition patterns, which are not governed by any rules. 8. Recombining: Oxford (1990) states that recombining involves making meaningful sentences or longer expressions by reordering parts in new ways. In other words, practice joining simple sentences using conjunctions. This is often a staple of writing textbooks. 9. Deductive Reasoning: This can involve looking at general English rules and applying them in writing. This would suggest the student looks up tutorials on the web about, for example, subject verb agreement, taking the rule, writing it in their journals, and using it to
write.
10. Analyzing Expressions: This refers to determining meaning by breaking it down into
parts to understand the whole; looking at a corrections and identifying nouns phrases and verbs. 11. Analyzing Contrastively: This can involve comparing the languages; looking for similarities and differences, making note of them in journals, and keeping these in mind while
writing. The obvious question which arises when looking at the issue of L2 strategies, whether it is with writing or another skill such as reading, is this: do they work? Is there evidence to suggest that engagement with L2 strategies actually produce better writers or better readers? O’Malley & Chamot (1985) and Anderson (2005) argue convincingly that strategy instruction benefits learners. Many other studies seem to suggest that strategies can work. Dulger (2011) conducted a 14 week long experiment involving 77 university freshman students to see if the use of Oxford’s (1990) metacognitive writing strategies leads to improvement. Using a
pre-test, post-test experimental group and control group design,
Dulger (2011) reports that the experimental group showed statistically relevant improvement in all areas of L2 writing i.e. content, organization, grammar and vocabulary. In other skill areas, such as vocabulary learning, the literature claims that the evidence is good for the efficacy of LLS. According to other researchers, Nation and Coady (1998); Lawson and Hogben (1996); Miller and Veatch (2010), research in L1 and L2 vocabulary instruction appears to suggest that direct instruction in cotextual clue usage is beneficial to students. 11
Conceptualization - cognitive approaches to writing and revision A common perception of the writing process is one in which the writer goes through distinct stages of pre-writing/planning, writing, and re-writing/revision. This can be seen as a linear pattern, a step by step process. It is common for L2 teachers to advocate this model in the classroom. These stages can be seen as stepping stones towards the final product; picture them as stones across a river. Planning is completed, then the writer writes, then the writer revises what he or she has written. However, according to Flower and Hayes (1981), this three stage model is perhaps a much too simple an explanation as to what writers actually do when they write. Advanced writers do not break down the composing process into distinct stages. Planning, writing and re-writing are cognitive acts which operate in a distinctly non-
linear fashion. In other words, all three processes can be at work at the same time. Anyone who has some skills in writing and who analyses what he/she does when he/she writes can
confirm this to be true. Flower and Hayes suggest that the term ‘stages’ is perhaps misleading. Instead, they suggest that each element in the process be seen as a mental act which can occur at any time. Instead of a linear set of stones across a river, picture many stones scattered in a non-linear manner. The writer moves towards the other side of the river,
the final product, by going forward, then perhaps to the side, then back, then forward again, until eventually they reach the other side and the written product is finalized. In order to get a clearer picture of this dynamic process, Flower and Hayes (1981) call upon protocol analysis, a research tool in which writers think aloud as they compose. The verbalized results of the writing process are recorded and then analyzed. A protocol analysis gives a picture of what happens when writers compose, including false-starts, diversions, questions asked, thoughtfragments and complete sentences. The result of this protocol analysis reveals that writing consists of the task environment, long term memory and the writing process itself. As Flower and Hayes (1981) state: The task environment includes all of those things outside the writer's skin, starting with the rhetorical problem or assignment and eventually including the growing text itself. The second element is the writer's long-term memory in which the writer has stored knowledge, not only of the topic, but of the audience and of various writing plans. The third element in our model contains writing processes themselves, specifically the basic processes of Planning, Translating, and Reviewing, which are under the control of a Monitor. (p 369)
12
What is clear in this description is that the traditional model of ‘stages’ is only one part of a much more complex phenomenon. Flower and Hayes (1981) go on to explain in detail what each element consists of. The task environment includes the problem solving aspect of writing and includes the assignment topic, audience expectations, and the writer’s own goal. This is a crucial element in the writing process because without a clear understanding of these aspects of the task environment the writer is unlikely to be successful. In other words, if a writer cannot clearly understand reader expectation, this will affect planning, writing and revision decisions. To stretch out our crossing the river metaphor, if the first stone is not secure, then moving to other stones will become problematic. Connected to the task environment is also the text as it develops on the page. As text is produced, it influences what can follow it. If a writer is unsure how to proceed with text they have already produced, the writing becomes incoherent. This is the common problem of unconnected sentences seen in
low level writing.
Regarding long term memory, Flower and Hayes (1981) state: The problem with long-term memory is, first of all, getting things out of it-that is, finding the cue that will let you retrieve a network of useful knowledge. The second problem for a writer is usually reorganizing or adapting that information to fit the demands of the rhetorical problem. (p 371)
This may explain, in part, the phenomenon of stop-start-stop-start writing seen in low level writers. They struggle with retrieving the knowledge of English syntax from their long term memory, if indeed they have this knowledge, and so pause half way through a sentence to think about what grammatical structure is needed. More advanced writers do not do this. Again, a small amount of self analysis is relevant here. For example, whatever other problems a writer may have, be it organizational skills or whatever, a more advanced writer does not need to wonder if the next preposition is correct or what tense is correct. These kinds of decisions are made automatically. Unfortunately, with Lower-Level writers, this is not the case. They struggle with decisions such as these.
This leads us to the writing process itself, although it should be clear that it is not a distinct entity. In the cognitive model advocated by Flower and Hayes (1981), the writing process consists of the mental acts of planning, translating and revision, although these are not 13
separate acts which can be easily delineated. Planning includes such elements as generating ideas, organization and goal setting. Translating is the term Flower and Hayes (1981) use for the act of putting words on paper or screen. They use the term ‘translating’ not in the one language to another language sense, but to represent the notion of translating thoughts into words, phrases or sentences. This process is related to short term memory and long term memory. If a writer has to spend time accessing both types of memory on a continual basis, then what is seen is stop-start-stop-start writing. It is as if after moving from one stone to another, the river crossing writer has to remind himself or herself about which direction he/she is moving in. If they have to do this all the time then NCFs are likely to occur.
Moving on, Flower and Hayes (1981) discuss the process of reviewing. …reviewing depends on two sub-processes: evaluating and revising. Reviewing, itself, may
be a conscious process in which writers choose to read what they have written either as a springboard to further translating or with an eye to systematically evaluating and/or revising the text. These periods of planned reviewing frequently lead to new cycles of planning and translating. However, the reviewing process can also occur as an unplanned action triggered by an evaluation of either the text or one's own planning (that is, people revise written as well as unwritten thoughts or statements). The sub-processes of revising and evaluating, along with generating, share the special distinction of being able to interrupt any other process and occur at any time in the act of writing. (p 374)
Last of all, but extremely important, is the notion of monitoring. As the term suggests, this is the element of composing which sees the writer monitor what they have already written and also what they will write. It also influences how much time a writer will spend on other aspects of producing text or how much time will be spent on planning, goal setting or accessing syntactic knowledge.
This is, in a very basic fashion, part of the current cognitive model of writing as expounded by leading experts in the field. Clearly, it is a lot more than just three distinct plan-writerevise stages. The evidence seems to suggest a non-linear process in which, for example, a writer engages in planning throughout the text, not just at the start of it. A writer will revise as they plan as they write as they organize as they revise as they plan, and on and on like this until finally a text is produced. 14
Another important finding from Flower and Hayes (1981) is this: more skilled writers often see their first draft as an exercise in exploration. In other words, the idea of a rough draft means just that, a rough exploration of ideas. This is one of the most crucial things that lower-level writers need to grasp. However, this is easier said than done. Given the difficulty a low level writer has in producing text, to have them finish a paragraph and then suggest that they start again with the process is often seen as highly frustrating. In my experience in the classroom, telling lower level writers that they ought to now discard what they have struggled with for an hour or so and re-write it from the start is often met with looks of incredulity. It is
as if they are saying to me: You cannot be serious! Look how much effort I have put into this already! Of course, students can use what they have already written, I am exaggerating
somewhat, but I am sure L2 writing teachers can recognize this problem. And yet, the truth is that good writing is never produced first time around. If the goal is good writing, then
students need to accept this idea that the first, rough draft is a means of exploration.
It could be argued that all of the above applies to skilled writers and not lower-level writers.
However, research seems to suggest that the same cognitive mental acts are at work in lowlevel writers (Silva, 1993; Sasaki, 2000; Cumming, 2001). Obviously, lower-level writers operate these mental acts with varying degrees of success. Skilled writers handle these mental acts with ease, and rarely become stuck on them, whereas low-level writers do become stuck at various points in the process.
Several researchers have posited a cognitive approach to revision which fits into a general cognitive approach to L2 writing as a problem solving process. Schmidt’s (1990) ideas are based on cognitive psychological research and the theory of conscious awareness in learning. Schmidt (1990) posits three levels of consciousness as it relates to revision: perception, noticing and understanding. 1). Perception: The student can see the problem area. This is where teacher feedback becomes important. 2). Noticing: Certain items which are perceptions are focused on. A choice is made to select. 3). Understanding: The student understands the NCF he/she has made. Schmidt (1990) asserts that L2 writing learners do not subliminally learn how to write. In other words, there is input (the students work, the book, my speech) but without noticing and paying attention to problems in their writing there will not be any intake, no ‘conscious understanding of the target language’ (Schmidt, 1990). The cognitive 15
theory as it pertains to NCFs in writing means to think about them and then apply mental and physical energy on memory and practice. Once a student has chosen something for attention e.g. an error in a paper, they need to move from noticing it to understanding it which means analyzing it, comparing it, separating it, finding out the significance of it, and experimenting with it, all of which takes considerable time and effort. In another cognitive model, Flower and Hayes (1981) suggest that revision is ‘sequenced subprocesses’ by which a writer modifies or plans a text. Firstly, task definition; in terms of the number of errors to be looked at, students must define what they are going to do. Secondly,
evaluation; students read to comprehend, to evaluate, to define text problems at all levels. Thirdly, problem-presentation; this means detection and diagnosis.
Finally, strategy
selection; this includes ignoring the problem, rewriting the text or revising it. In this study
this means using the strategies seen above to engage with the NCFs in their journal.
Izumi (2003) describes stages of comprehension which are similar to other models. Stage 1: Apperception; input is perceived in light of previously held knowledge and current experiences. Stage 2: Comprehended input; in this stage the apperceptive stimulus is analyzed for meaning. It can be analyzed for semantic meaning or syntactic meaning. It has been suggested in the literature that syntactic meaning provides better acquisition opportunities than semantic meaning. This is related to noticing and raising awareness. Stage 3: Intake; this leads to long term memory storage but depends on the level of analysis given to stage 2. Stage 4: Integration; in this stage the information is shuffled, so to speak, between what is already known and what is not known. A process of ‘re-confirmation and rulestrengthening’ takes place Izumi (2003). Stage 5: Output; extensive writing represents an increase in student output. Izumi suggests that output triggers a distinct set of cognitive processes into motion which differ from input. Output helps learners notice syntactic features as they are being created. Input is semantic processing but output is syntactic processing. Swain and Lapkin (1995) argue that output is a conscious raising activity. The crucial element in the above model is Stage 3 and its relationship to Stage 2. To notice something needs to be a process in which the thing noticed is stored in long term memory. In order to do this, attention needs to be given to it. In this study, it is hoped that the attention is given in the journal work.
16
What is clear from the above models of cognitive writing is the importance of revision for successful writing. However, one method of revising that does not appear to be mentioned in the literature is the idea of revision task writing in journals. Tasks are essential for learning to write. In a meta-analysis of 500 L2 writing papers, Horowitz (1986), as cited in Dyer (1996), concludes that traditional grammar teaching minimal effect, but the environmental mode, a combination of task based writing with clear goals and objectives was the most effective. There seems to be a lack revision in journal studies in the literature. Revision task writing consists in identifying an NCF from teacher feedback, and then
applying various L2 writing strategies to the NCF. What this involves is taking the strategy and applying it to an NCF in the student’s journals. For example, if a student makes an NCF
related to part of speech, say, for example they write: The city is very pollution, they would
then take this NCF, as marked by the teacher, and in their journal they would make a list of
the different forms the word ‘pollution’ can take. In essence, what they are being asked to do is make a record of word families. As explained by Yule (1996), a word family consists of inflectional bound morphemes and derivational bound morphemes. Students might also write a separate sentence for each form. This, in itself, is a rather simply idea, but one which is not, as far as I can discern, been covered in the literature regarding revision. Of course, learner strategies themselves have a long history in the literature, as do journals and revising practices, but the combination of strategies and journals as a way of revision is not seen in the literature, as far as I can tell. It is this combination which is at the core of this study. It should be noted that revision in this sense is not about improving drafts as such; students did improve drafts, obviously, as they corrected what the teacher had marked. However, this kind of task revision in journals is more about improving writing over the long term.
Data – error analysis To conclude this literature review, it is pertinent to briefly discuss aspects of error analysis, the method by which teachers make a record of what NCFs students are making. As mentioned in the introduction with regards to terminology, experts in the field of error analysis argue that mistakes and errors are not the same thing. These two terms may not be interchangeable, but are often used in this way, perhaps because of convenience factors. As Corder (1981) explains, errors can be seen as evidence that the learner does not yet have full command of the target language system. Mistakes, on the other hand, occur when a learner does, perhaps, have command, but simply forgets to apply his or her knowledge. Mistakes are 17
slip ups and may be caused by any number of reasons; tiredness, forgetfulness, emotional state and environmental factors such time constraints and conditions under which the writing takes place. Clearly, if at all possible, it would be useful for a teacher to know which are which. Corder and Ellis (1994) suggest one way to find out is to look at student errors across different pieces of writing. What has to be recognized here is that when a teacher says, ‘my students make a lot of mistakes’ it is not the same thing as saying, ‘my students make a lot of errors’. Second language writers make errors and mistakes, sometimes many of them, and however
much this may frustrate teachers it should come as no surprise. They are learning the language, and to learn means to make errors. In the field of language pedagogy and second
language acquisition, the study of learner’s errors and mistakes is called error analysis. This
subject has a long history dating back to the 40’ and 50’ where it was dominated by
behaviorist theories of learning. Within the behaviorist paradigm errors were seen in terms of bad habits influenced by the mother tongue, but Corder (1981) postulated that errors were made when learners formatted hypothesis about the target language. Errors represented a leaner’s ‘transitional competence’ (Corder). In the 60’s, Corder reformed the study of errors by placing more emphasis on the cognitive process of learning. An error analysis can serve two important functions: pedagogical and theoretical. The theoretical side is concerned with making hypothesis about the second language acquisition process of learners, while the pedagogical side is more concerned with classroom teaching implications. This study is concerned with the pedagogical side of error analysis. A pedagogical error analysis can provide teachers with information which may serve remedial purposes related to course content and teaching practices (Corder, 1981; Lennon, 1991; Etherton, 1997; Dulay, Burt & Krashen, 1982). As Etherton (1997) points out, it is the only way a teacher can unequivocally see the frequency of student errors. Etherton suggests that a teacher may develop a vague sense of what students do not know while correcting writing errors, but only through doing an error analysis will he or she gain accurate knowledge. Corder and Richards, (1980) also suggest error analysis results can also assist in syllabus design and remedial work.
18
Procedural difficulties in doing an error analysis One of the fundamental aspects of doing an error analysis is to reconstruct the NCF into a correct from. This will provide a yardstick for the researcher to compare. However, as Corder (1981) points out, this is far from a straightforward exercise. Corder describes two kinds of reconstruction; authoritative and plausible. An authoritative reconstruction is only possible if the learner who made the original NCF is available to ask. A plausible reconstruction is the analyst’s interpretation, which is subjective and open to error. A plausible account, then, is the analyst’s subjective interpretation of what the learner was originally trying to write. In this study, as the learners were not available for asking, the interpretations are plausible only. Please see appendix 1 for a sample of student NCFs.
One thing that becomes apparent when reconstructing NCFs is that global errors are much more difficult to interpret. Global errors are errors which hinder communication of meaning
and are generally concerned with the overall sentence organization. Local errors are errors concerning single elements in a sentence and generally do not cause problems with meaning. Put plainly, global errors make understanding the meaning of the sentence difficult. From the data, a global error such as: 1). Homework is the easiest thing for a good friend can help is clearly a lot more problematic to reconstruct than a sentence with a local error such as: 2). Your friend has many stories he want to tell you. It is not the case that utterances which contain global errors cannot be reconstructed. However, they are a lot more difficult, and the subjective factor becomes larger. The classic framework used to conduct an error analysis is given by Corder (1981). It consists of six steps: 1). Selection of data, 2). Identification of errors, 3). Reconstruction of errors, 4). Classification of errors, 4). Explanation of the psycholinguistic causes of the errors, 5). Evaluation of the error gravity. The 6th step given by Corder is related to second language acquisition, an area not relevant to this study, although obviously important in its own right. In order to study the NCFs (non-correct forms), the researcher must provide a qualitative linguistic classification and a quantitative description of the frequency. Corder (1981) suggests four linguistic headings: Orthography, Morphology, Syntax, and Vocabulary, although these days we might say Lexical instead of Vocabulary. Within these categories more specific items are identified. For example, morphology can be unlocked to produce third person s, past tense, comparative and possessives, while syntax can be unpacked to 19
omission of article, wrong word order, and plurals. Dulay et al., (1982) suggest other methods of classification including error types based on surface strategy. The authors explain that this kind of classification highlights the way surface structures are altered. Briefly, a surface based classification would take into account omission of a necessary item in order to correct the error or addition of a necessary item in order to correct the error. This concludes the literature section of this paper. In this section there has been a discussion of the use of journals in L2 writing classes. The conclusions drawn illustrate that journals can be a useful addition to the L2 writing class. In addition, this section has briefly discussed the
notion of revision within a cognitive framework of L2 composition, briefly talked about the notion of L2 strategies, and briefly mentioned some of the key elements involved in doing an
error analysis. This now leads us onto the methodology section.
Methodology
In this section of the paper, I will briefly discuss my work conditions and the course-book used. The purpose of this is give relevant local context to the study. In addition, the design of the study, details of the participants and data collection procedures are explained.
Work conditions I currently, (2013) teach at Burapha University which is situated in Chonburi province, Thailand. I have been working here for ten years. The university is situated in the rapidly expanding town of Bang Saen. Working at the university is professionally rewarding and conducive to teacher research. We are encouraged to do research in our classes.
Course-book The course-book includes material which illustrate a number of what Hyland (2003) calls the ‘guiding concepts’ of L2 writing. These concepts are current ways of thinking about the best way to teach L2 writing. Hyland describes the concepts under the headings of structure, function, themes, creative expression, composing process, content and genre. These headings represent conceptualizations of writing instruction. Several of these headings seem to be influencing the content of the course book. The book features extensive practice with simple sentences, subject/verb agreement, fragments, compound sentences, coordination conjunctions, subordinators; the grammar components are evident in the book. A hard line structural focus assumes that writing is constructed from a writer’s knowledge of 20
grammatical and lexical elements; writing is result of imitating models Hyland (2003). Clearly the book draws somewhat on this way of looking at writing with its model essays used as stimulus for tasks. The book also includes a functional element with its description of the three parts of an essay. The book also appears to adopt elements of content writing instruction. This approach emphasizes, among other things, writing around certain themes connected with cultural issues. Some of the model essays in the book seem to be chosen with this aspect in mind; a wedding, a special birthday. Furthermore, the book adheres to a process approach to writing. The guidelines in the book include prewriting, planning, revision, rewriting, revision, and final draft. The course book appears to attempt a synthesis of at least some of the guiding principles.
Design of the study
This study utilized a prospective, quasi-experimental before-and-after design, featuring a
control group and an experimental group. This kind of design is useful for measuring the impact of independent variables. As Kumar (1996) explains, the experimental group receives or is exposed to the treatment, whereas the control group is not exposed to the treatment. After the treatment, an after data collection is made. Any difference in the ‘before’ and ‘after’ observations between the groups regarding the dependent variable (s) is tentatively attributed to the treatment (Kumar). In the current study, the control group was taught without attention given to journals and L2 strategies, while the experimental group did pay attention to journal writing and L2 strategies. This before-and-after design is useful for measuring change and the result of an intervention. The change, the dependent variable, is the number of NCFs after the treatment. The two populations, control and experimental, are looked at here in terms of error rates before and after the intervention/ treatment. Before the course began, an error analysis of both groups was made (see appendix 1 for samples). The course continued for sixteen weeks, at the end of which another error analysis was done on both groups. The differences were analyzed using the SPSS statistical package.
Independent variables. The independent variable is commonly described as the cause supposed to be responsible for the change in a situation. The independent variables in this study were the journals and the student’s engagement with their NCFs through Oxford’s (1990) strategies. The treatment in this study consisted of asking the students in the experimental group to engage with a selected number (see appendix 2) of strategies taken from Oxford’s seminal work. Students in 21
the experimental group were required to engage with their NCFs, as indicated by teacher feedback, in their class journals. They were expected to do this outside of class. In this respect, students in the experimental group would be applying the strategy in their journals as homework. All students in the experimental group were given ten strategies adopted from Oxford (1990) (see appendix). All students in the experimental group had the same kind of journal; a commonly available notebook bought from the university bookstore. Each journal was physically the same size. There was 10 percent grade given on these journals, as it was felt that some extra motivation might ensure full cooperation with the strategies. In the journal they were asked to engage with the L2 tasks and try to improve their writing. They were instructed to pick five tasks out of ten each week. They were asked to look at their NCFs, and asked to apply the strategies to them in their journals. An example of what this entailed was shown to the experimental group in class. The journals were checked every
week in class by the teacher/researcher. After about the first two weeks, the class appeared to engage with the tasks in their journals for the rest of the semester, 14 weeks. It was clear that they were following the instructions given, and they were using the strategies in their journals to engage and think about their NCFs. No specific training in using Oxford’s (1990)
strategies was given to the experimental group, as they were already familiar with the idea from previous classes in which I had taught them. In addition, another independent variable was the number of NCFs made before and after the treatment. Before the treatment, at the start of the semester, students in the control and experimental groups were asked to write a short in class essay on their hometown. No time limit was given for both groups for these in-class essays, in order to ease anxiety. The word counts from both groups with regards to these before the treatment essays were similar. No grades were given for this assignment. It was indicated that this assignment was a warm up essay in order to give the instructors some kind of perspective on their writing. All students from both groups completed this essay. From these data sets, the number of NCFs was extracted. The NCFs were categorized into three main areas: nominal, verbal, and lexical. Three native speaker colleagues checked these results and there appeared to be widespread agreement as to the accuracy. After the treatment, at the end of the semester, another error analysis was conducted. For this data set, the student’s final exam essays were used.
22
Dependant variables The dependant variable is commonly described as being the outcome of the change brought about by the independent variable. The dependent variable in this study was the number of NCFs after the treatment. The after treatment data were final exam essays featuring an opinion essay. Both groups wrote an opinion essay in their final exam from which the number of NCFs was extracted. The criteria used to categorize the NCFs were kept relatively simple. NCFs were categorized into three main areas: nominal, verbal, and lexical. The nominal and verbal groups were straightforward. In the nominal category, NCFs with of, pronouns and plural quantifiers were counted. In the verbal group, NCFs related to missing verb to be, addition of verb to be, the verb to have, tense, and infinitive were extracted.
The lexical category perhaps needs further elucidation. The lexical categories used were a simplified version of a detailed study done on lexical errors by Hemchua and Schmitt (2006).
The authors suggest that ‘inappropriate lexical choices’ are frequent in language learning, and are the kind of errors which can cause problems with the content and ‘the quality of academic writing’ (Hemchua & Schmitt, 2006). The authors include morphological errors, limited to derivational affix errors, (suffixes and prefixes). In these cases, the student shows that he or she knows the word but not the correct derivation. Included in this lexical category were, wrong semantic word choice, verb + preposition combinations, part of speech, and near synonyms. MacMillan (2007) categorizes synonyms into two kinds: simple synonym, where a lexical item can be replaced by another, in context, and complex synonym, where the lexical item is not the same word class. For this study, either of these types was classified as a NCF related to the general heading of synonym. Please refer to appendix for examples of student NCFs in all three categories.
Extraneous variables These are variables apart from the independent variables which may have an effect on the dependent variable. In this study such variables may include how much motivation each student has or how much additional writing each student did apart from what was required in the course and the treatment. In addition, the extraneous variable may include general intelligence level and/or attitudes towards writing. Despite an attempt to keep these extraneous variables to a minimum, these kinds of variables are not so easy to measure in an objective manner, hence the quasi-experimental title in the design. Any or all of these extraneous variables may or may not have an effect on the dependent variable. 23
Participants The two population groups each contained 22 students who were third year Education and English major students at Burapha University. Both groups were on average 19 years old at the time of the study. Both groups at the time of the study were taking Essay Writing, a required course in their programme. Both had taken paragraph writing in their second year of study and sentence level writing in their first year of study. In addition to this, both classes had been taught syntax and grammatical structure in other classes. Both groups used the same course-book and followed the same syllabus. Both groups adopted
the process approach to writing which means they underwent the process of draft –correction- draft-correction-final draft.
Results
Research Question one Is there a significant difference in NCFs before the treatment between the control group and the experimental group?
Table 1: Overview comparison of the control group and the experimental group on NCFs before the treatment (control group n=22, experimental group n=22) Variable
M
SD
NCFs Control
.69
.23
Experimental
.68
.17
p
t
df
p
.11
42
.911
.05
Table 1 shows that the control group and the experiment group were not significantly different on NCFs before the treatment. Inspection of the two groups means indicated that the average NCFs for the control group (M = .69) is higher than the experimental group (M = .68). The effect size d is approximately .05, which is small (Cohen, 1988). The difference between the mean is .01. The control group did not differ significantly from the experimental group on NCFs (p = .911), but the control group did have more NCFs. 24
Table 1.1: Specific areas comparison of the control group and the experimental group on NCFs before the treatment (control group n=22, experimental group n=22) Variable
M
SD
missbe control
.82
.79
experimental
1.05
.89
addbe
t
df
p
-.88
42
.379
-2.02
42
.050*
.96
42
.343
.99
42
.326
.00
42
1.000
48
42
.629
.48
42
.633
control
.36
.58
experimental
.77
.75
have control
1.00
1.38
experimental
.68
.71
tense control
1.05
1.25
experimental
.73
.82
infinitive control
.36
.65
experimental
.36
.58
word choice control
.41
.66
experimental
.32
.56
verb prep.
25
control
.64
.58
experimental
.55
.67
part of speech control
1.73
1.12
experimental
1.64
.90
near synonym
.29
42
.768
-.63
42
.529
-.47
42
.637
.00
42
1.000
- .61
42
.541
.45
42
.654
control
.27
.45
experimental
.36
.49
spelling control
.45
.67
experimental
.55
.59
with of control
.77
.68
experimental
.77
.61
pronouns control
.36
.72
experimental
.50
.74
plu. quantifier
p
control
.82
1.14
experimental
.68
.83
.05 26
Table 1.1 shows that the control group and the experiment group were significantly different on addbe before the treatment. Inspection of the two groups means indicated that the average addbe for control group (M = .36) is lower than the experimental group (M = .77). The effect size d is approximately .62, which is medium (Cohen, 1988). The difference between the mean is .41. The control group differed significantly from the experimental group on addbe (p = .050), but the control group made less addbe NCFs.
Research Question two Is there a significant difference after the treatment between the control group and the
experimental group?
Table 2: Overview comparison of the control group and the experimental group on NCFs after the treatment (control group n=22, experimental group n=22)
Variable
M
SD
NCFs Control
.73
t
df
p
6.38
29.20
.000
.22
Experimental p
.39
.10
.05
Table 2 shows that the control group and the experiment group were significantly different on NCFs after the treatment. Inspection of the two groups means indicated that the average NCFs for control group (M = .73) is higher than the experimental group (M = .39). The effect size d is approximately .50, which is medium (Cohen, 1988). The difference between the mean is .34. The control group differed significantly from the experimental group on NCFs (p = .001), and the control group made more NCFs.
27
Table 2.1: Specific areas comparison of the control group and the experimental group on NCFs after the treatment (control group n=22, experimental group n=22) Variable
M
SD
missbe control
1.14
.71
experimental
.50
.59
addbe
t
df
p
3.21
42
.003*
1.08
42
.286
1.75
34.63
.088
1.61
42
.113
1.31
42
.197
.00
42
1.000
-.23
42
.814
control
.64
.72
experimental
.41
.66
have control
.73
.93
experimental
.32
.56
tense control
.68
.71
experimental
.36
.58
infinitive control
.50
.59
experimental
.27
.55
word choice control
.27
.45
experimental
.27
.45
verb prep.
28
control
.50
.67
experimental
.55
.59
part of speech control
1.59
.79
experimental
.45
.59
near synonym
5.35
42
.000*
1.88
25.51
.072
1.83
42
.073
.46
42
.642
1.04
42
.302
2.34
42
.024*
control
.32
.64
experimental
.05
.21
spelling control
.86
.64
experimental
.50
.67
with of control
.73
.55
experimental
.64
.72
pronouns control
.73
.70
experimental
.50
.74
plu. quantifier
p
control
.91
.86
experimental
.36
.65
.05 29
Table 2.1 shows that the control group and the experiment group were significantly different on three specific areas with reference to NCFs after the treatment; 1) missbe, 2) part of speech and 3) plural quantifier. Inspection of the two groups means indicated that the average 1) missbe for control group (M = 1.14) is higher than the experimental group (M = .50). The effect size d is approximately .98, which is large (Cohen, 1988). The difference between the mean is .64. The control group differed significantly from the experimental group on missbe (p = .003), and the control group has a higher number of incorrect use of missbe NCFs. Table 2.1 also provides descriptive statistics and statistical tests to indicate that the
experiment group was also significantly different from the control group on making part of speech NCFs after the treatment. Inspection of the two groups means indicated that the
average mean of part of speech NCFs for the control group (M = 1.59) is higher than that of
the experimental group (M = .45). The effect size d is approximately .60, which is medium
(Cohen, 1988). The difference between the mean is 1.14. The control group differed significantly from the experimental group on part of speech (p = .001), and the control group made more part of speech NCFs. Table 2.1 reveals the last significantly different area which is 3) plural quantifier. Inspection
of the two groups means indicated that the average 3) plural quantifier for control group (M = .91) is higher than that of the experimental group (M = .36). The effect size d is approximately .73, which is medium (Cohen, 1988). The difference between the mean is .55. The control group differed significantly from the experimental group on plural quantifier (p = .024), and the control group made more plural quantifier NCFs.
Research Question three Is there a significant difference in NCFs in the experimental group before and after the treatment? Table 3: Overview comparison of the experimental group on NCFs before and after the treatment (experimental group n=22) Variable
M
SD
NCFs before
.68
.17 30
t
df
p
6.80
42
.000*
after p
.39
.10
.05
Table 3 shows that in the experimental group, there is a significant difference of NCFs between before treatment and after treatment. Inspection of the experimental group before the treatment means indicated that the average (M = .68) is higher than after the treatment (M = .39). The effect size d is approximately .45, which is small (Cohen, 1988). The difference between the mean is .29. The NCFs before and after the treatment differed significantly (p = .001), and the experimental group made fewer NCFs after the treatment.
Table 3.1: Specific areas comparison of the experimental group on NCFs before and after the treatment (experimental group n=22)
Variable
M
SD
missbe before
1.05
.89
after
.50
.59
t
df
p
2.37
42
.022*
1.69
42
.097
1.86
42
.069
1.68
42
.099
.53
42
.597
addbe before
.77
.75
after
.41
.66
have before
.68
.71
after
.32
.56
tense before
.73
.82
after
.36
.58
infinitive 31
before
.36
.58
after
.27
.55
word choice before
.32
.56
after
.27
.45
verb prep.
.29
42
.771
.00
42
1.000
5.12
36.39
.000*
2.78
28.60
.009*
.23
42
.814
.673
42
.504
.00
42
1.000
before
.55
.67
after
.55
.59
part of speech
before
1.64
.90
after
.45
.59
near synonym before
.36
.49
after
.05
.21
spelling before
.55
.59
after
.50
.67
with of before
.77
.61
after
.64
.72
pronouns before
.50
.74 32
after
.50
.74
plu. quantifier
p
39.74
before
.68
.83
after
.36
.65
39.74
.169
.05
Table 3.1 shows that the experiment group was significantly different on three specific areas before and after the treatment; 1) missbe, 2) part of speech and 3) near synonym. Inspection
of the experimental group means of missbe NCFs indicated that before the treatment (M = 1.05) is higher than after the treatment (M = .50). The effect size d is approximately .74, which is medium (Cohen, 1988). The difference between the mean is 1.00. The experimental group before treatment differed significantly from after treatment on missbe (p = .022), and the experimental group made less missbe NCFs after the treatment.
The statistical tests from Table 3.1 also indicate the significant difference before and after the treatment on part of speech. Inspection of the experimental group means indicated that before the treatment (M = 1.64) is higher than after the treatment (M = .45). The effect size d is approximately .62, which is medium (Cohen, 1988). The difference between the mean is 1.19 The experimental group before treatment differed significantly from after the treatment on part of speech (p = .001), and the experimental group made less part of speech NCFs after the treatment. Near synonym was the last specific area of NCFs which is significantly different before and after the treatment in the experimental group. Inspection of the experimental group means indicates that before the treatment (M = .36) is higher than after the treatment (M = .05). The effect size d is approximately .88, which is large (Cohen, 1988). The difference between the mean is .31. The experimental group before treatment differs significantly from after the treatment on part of speech (p = .009), and the experimental group made less synonym NCFs after the treatment.
33
Summary of the results 1 Is there a significant difference before the treatment between the control group and the experimental group? According to SPSS independent sample t test, there is no overall significant difference between the control group and the experimental group before the treatment. p=.91. However, there is one significant difference in ‘addbe’ category. p = .050. The experimental group has more NCFs in this category than the control group. 2 Is there a significant difference after the treatment between the control group and the experimental group?
Overall there is a significant difference between the control group and the experimental group after the treatment. This shows that the control group has more NCFs on average than the
experimental group. 3. Is there a significant difference in the experimental group before and after the treatment? The statistical analysis indicates a significant difference of NCFs in the experimental group before and after the treatment in these three areas: ‘missbe’, ‘part of speech’, ‘near synonym’. In sum, the experimental group improved in these categories.
Evaluation and Discussion In this section of the paper, an evaluation of the study is made which outlines some weaknesses in the study. Following this, a discussion is made as to the possible reasons for the results. Then, some suggestions are made for further studies. Finally, some pedagogical implications will be discussed. Clearly, it would be rash to assign a straightforward cause and effect relationship between the treatment and the apparent positive results to research question 2: Is there a significant difference in NCFs after the treatment between the control group and the experimental group, and research question 3: Is there a significant difference in NCFs in the experimental group before and after the treatment? There are many factors that could be involved in these results. One possible reason for the results could be what Kumar (1996) refers to as the maturation effect. Simply put, as the students in the experimental group were taught English in other classes, such as reading, for example, they became better and gained more knowledge, which 34
in turn had an effect on the number of NCFs they made. Another possible explanation could be related to the extraneous variables. There is no way of knowing for sure if the students in the experimental group were not perhaps taking private lessons or perhaps writing a lot more than what was witnessed in class or in their journals. In addition, it has to be acknowledged that the experimental group was working through the course-book as per usual during the course of the study, and perhaps this regular classroom work had an effect on the results. All of these factors need to be taken into account. Therefore, caution needs to be applied in stating a direct cause and effect relationship.
As to the design of the study, perhaps it would have been better if the second set of data, the after treatment data, had been similar to the first set of data i.e. an in-class, no time limit
essay. As it was, the second set of data was collected from the final exam essay. In addition
to this, another weakness of this study was the fact that there was no data triangulation; the
study would have been more robust if interviews had been set up with participants to discuss the use of the journals. Moreover, questionnaires could have been administered to investigate the impact of following the L2 writing strategies in journals. However, taking into account all of the above it still appears as if the results indicate a positive effect of using journals in conjunction with the strategies on the number of NCFs made by the experimental group, although this is, perhaps, not really surprising or unexpected. What possible explanation can be put forward for this phenomenon? It could be that the journal work helped the students in the experimental group to notice their NCFs. It may be that engaging with the strategies in the journals raised their awareness of their common NCFs and thus helped them to reduce NCFs over the time period of the study. Noticing language is about focusing mental effort on the language. Barnawi (2001) describes noticing as the purposeful allocation of resources to linguistic input and output. Noticing is often synonymous with terms awareness or attention. Schmidt (1990), one of the leading advocates for the cognitive view of writing, argues that noticing is essential to second language acquisition. Kellogg (2010) also backs up this claim. By extension, it could be said that noticing is essential to reducing NCFs in writing. In the L2 writing class, teacher feedback in whatever form is supposed to contribute to the process of focusing student attention on some of the problems or errors and mistakes in their drafts. As Ellis (2009) suggests, a cycle of feedback on drafts produces a mechanism where 35
students improve by comparing what they wrote to what they should have wrote. In theory, this sounds fine, but is this really enough? It may be that, in addition to the rewrites, the journal work helped them notice NCFs in a more salient manner. As stated, Schmidt (1990) and others (see Izumi, 2002; Hanaoka, 2007; Flower & Hayes, 1981; Mackey, 2006) argue that learners need to notice or become aware of the language before learning takes place. Indeed, researchers argue that promoting awareness is essential because intake depends on it. This appears to be connected to a theory of consciousness which advocates that memory training facilitates learning. In this theory, memory can be seen
as consisting of compartments; short term and long term. A perception is put in the short term memory. What happens to the perception in the short term memory will decide if the
perception goes into the long term memory. Short term memory is essential for movement
towards long term storage. Perceptions that demand long term memory storage need to be
further processed by our memory faculties in helping retention (Qi & Lapkin, 2001). Writing text requires not just the language system, argues Kellogg (2010), but also includes memory and thinking. Writers put their knowledge in long term memory, they store it away there, and they can use it widely if they can retrieve it rapidly from long term memory or keep it in working memory, ‘actively maintaining it’, as Kellogg (2010) puts it. Writing is problem solving, Kellogg (2008) argues. What to say and how to say it consumes the writer’s short term memory. Kellogg (2008) says ‘cognitive apprenticeship and more practice’ is what we need for interventions that train and instruct. It could be that the use of journals and strategies helps free up short term memory. Students notice what they are doing more, and therefore appear to make fewer of the common NCFs. The actual process of noticing is not fully understood, but some researchers have indicated what might be involved. As already mentioned in the literature review, Schmidt’s (1990) ideas are based on cognitive psychological research and the theory of conscious awareness in learning. Schmidt (1990) puts forward three levels of consciousness: perception, noticing and understanding. 1). Perception; sight, 2). Noticing; certain items are perceptions and focused on. A choice is made to select. 3. Understanding. Schmidt (1990) asserts that without noticing and paying attention to problems in their writing there will not be any intake, no ‘conscious understanding of the target language ‘(Schmidt, 1990). Taking this into account, it might be the case that doing the journal work helped the students to notice what they were doing with their writing. 36
Moving on to another explanation for the results, one of the strategies put forward to the students in this study is the notion of comparing English and Thai. Without going into great detail, some very basic distinctions can be made between the two languages (see Sattayatham & Rattanapinyowon, 2008; Phoocharoensil, 2011; Pongsiriwet, 2001; Bootchuy, 2008). It may not be too farfetched to say that asking students to make note of these distinctions helped them notice them, raised their awareness of them, and thus helped them reduce these common types of NCFs. Very briefly, studies do appear to suggest that one of the main reasons for certain kinds of NCFs is L2 transfer. As an example, words that are synonymous in Thai may not be in English. Moreover, subject verb agreement, closely followed by tense formation, singular/plural nouns, and word form and verb omission are an issue for Thai writers. Thai
and English share subject-verb-object structure but plural nouns are a different syntax in Thai. In English, an s’ is needed but in Thai it is not. In English, pronouns change form
depending on their position as a subject or an object; however, in Thai this does not occur. Thai has no inflection to mark tense, unlike English. Students clearly need to be aware of this. They need to explicitly notice this when then they write in English. In Thai subject and verb do not have to agree; in English they do. English has irregular verb tenses, while Thai does not. Bootchuy (2008) also states that omission of verbs is an issue in Thai student writing. Bootchuy (2008) also explains the pro-drop nature of Thai. In Thai, the subject can be dropped if the context is inferable. In English, sentences require a subject in the relative clause. Thai has adjectival verbs, whereas English does not. In Thai, an adjective can follow a subject and act as a verb. Furthermore, in Thai, the preposition ‘with’ can mean ‘to’ ‘at’ or ‘with’. In addition, Thai does not have articles. In Thai, quantifiers are ‘much’ ‘many’ ‘a few’ and ‘a little’, but they do not have to agree with the noun. In addition, in Thai, these words carry a wide range of quantity such as ‘plenty of’ ‘more’ ‘much’ ‘several’. All of these basic differences need to be noticed by Thai students. It is not the case that they do not know about them, but they perhaps need to explicitly notice them, and this is where the journal work may, perhaps, help. Obviously, knowledge of the basic distinctions is also important for native speaker teachers who may not have much skill in Thai. Another avenue worth exploring is this: perhaps the journals increased the student awareness of English part of speech formations. This is an area of English called derivation. Along with inflection, derivation is a component of morphology. Morphology is a description of what 37
happens when bound morphemes combine with free morphemes. A bound morpheme cannot stand as a word, but a free morpheme can. For example, ‘enjoy’ is a free morpheme and ‘ful’ is a bound morpheme. When they combine, they change the word class. As mentioned by Jackson and Zé Amvelu (2006), inflectional suffixes (always suffixes in English, no prefixes) do not change the word class. When an inflectional suffix is added, the word does not change word class, that is, the noun stays a noun and the like, explain Yule (1996); Jackson and Zé Amvelu (2006). However, derivational affixes often do change the word class, but not always. A word can change class without derivational affixes. Different phonological stress can change word class (Jackson & Zé Amvelu, 2006; Yule, 1996). It would appear as if some of my students have some misunderstandings concerning derivational suffixes. In English there about sixty derivational affixes (prefixes and suffixes). In a seminal study, Bauer and Nation (1993) tabulated common English derivative affixes according to their
frequency and utility into different levels.
Hemchua and Schmitt (2006) suggest that many EFL students need greater awareness of word families, and the same might be said here in this study. A word family, as explained by Yule (1996), consists of inflectional bound morphemes and derivational bound morphemes. The only way to know about this is to study word families and determine which words in the family are nouns, verbs and adjectives.
Suggestions for further study What needs to happen next is an attempt to replicate this study. The results of any future replication may yield different results if done with different sets of population. Hyland (2003), one of the leading experts in L2 writing, bemoans the fact that very rarely is there any attempt to replicate studies showing positive or negative results. Simply put, several studies that indicate negative results using the same methods and design as this study would be invaluable. Of course, studies that replicated this one and found similar results would also be extremely useful to L2 writing pedagogy and research. It is perhaps not surprising that studies are not replicated as most researchers have their heart set on doing something original, something which they can call their own. It seems to be a lot of work to just try and copy someone else’s research. However, Hyland (2003) argues that this is what is needed in L2 writing research. In the hard sciences, such as chemistry and physics, studies are often replicated and results published which either support or undermine conclusions. 38
It might also be worthwhile to focus on just one area of student writing, for example, tense formation, to see if journal work can help in this regard. It might also be useful to use a longer time scale. Perhaps more than just two groups could be used in further studies.
Pedagogical implications Journals might be introduced into the L2 writing classroom. They can provide additional practice for students, in addition to in-class writing and course-book work. As mentioned earlier, they can provide a stress-free environment in which students can write down daily thoughts or practice paying attention to various aspects of their writing. There is also the fun
aspect, an important element where Thai students are concerned. Some students in my class copied down conversations they had on-line and found it amusing to re-read what they had
been saying to friends and strangers.
Journals can also provide students with an opportunity to write without worrying about
grades or comments. It might be said that constant negative teacher feedback is discouraging to students. If the results of this study prove to be robust, then engaging with strategies in journals may help reduce the number of common NCFs made by students, which can only be
encouraging and a good thing for them.
Learner strategies need to be introduced into the L2 writing classroom. The evidence for the efficacy of strategy use is overwhelming (Hong Nam & Leavell, 2005; Wong & Nunan, 2011; Sasaki, 2000; Baker & Boonkit, 2004). In other words, strategies do work. For those classes whose students are not aware of strategies, certain training models can be implemented. One approach to strategies instruction suggested by Anderson (2005) is Styles and Strategies Based Instruction (SSBI). This methodology puts the learning of language learning strategies (LLS) at the heart of its philosophy. In these scenarios, students engage with the LLS on a regular basis as part of everyday classroom instruction. Classes embracing SSBI are explicitly taught LLS in class, and use them on a regular basis for their learning. Perhaps the most important pedagogical implication of this study is acknowledgment of just how complex it is to write in a second language. It is easy to complain that students cannot write simple sentences correctly; however, taking into account the difficulties involved, this should not be surprising. Recall that experts claim that it takes many years of sustained study to become a competent writer. Indeed, in this respect, four years of university study is highly 39
unlikely to produce advanced writers. This is not intended to let teachers ‘off the hook’ so to speak, but simply to acknowledge how difficult it is to write, and teach English writing. As an example of the complexity of writing in English, teachers who can be heard complaining that their students keep forgetting to add an ‘s’ on verbs need to sit back and think about what is involved in this from a cognitive perspective. Pinker (1995) explains that to correctly place a third person ‘s’ on a verb a learner would need to have command of several English language conventions. Is the subject in the third person? They have to remember that ‘he’ ‘she’ and ‘it’ are followed by a different inflectional form of the verb
from ‘we’ and ‘I’ and ‘they’. In addition to this, adding ‘s’ is not the only way to make third person verbs. As Alexander (1988) explains, there are spelling considerations. For example,
some verbs take ‘-es’ and some take ‘-ies’. They have to remember this. They also have to
cognitively process the aspect of what they are saying. Is the action a regular occurrence? Is it
a habitual action? Does it require a continuous tense rather than present tense? Moreover, they have to distinguish the ‘s’ from other verb and noun endings. This is the domain of inflection in English. In order to remember the ‘s’, learners need understanding of inflectional categories. Teachers of L2 writing should, perhaps, from time to time, remind themselves of these complexities. Teachers should try to understand the students’ problems and difficulties, and find ways to help them. Instead of complaining and chastising students about the number of NCFs, teachers may try to point out the NCFs, and advise the student how to deal with these problems, especially the most frequent and major NCFs. The purpose of this study was to help students become better L2 writers by asking them to focus on their NCFs in their journals. The study was designed with a control group and an experimental group. The results of the study indicated that students who worked with their NCFs in journals made fewer NCFs, specifically in the areas of missing verb to be, parts of speech and near synonyms. It may be that by focusing on NCFs through writing tasks, students noticed more areas in which they were making NCFs and thus improved. It is hoped that this study will help other teachers to teach L2 writing, and perhaps inspire them to use journals and tasks as a means of improving student writing. If students do not revise and practice their NCFs in a systematic manner, it is doubtful that improvement will be seen. In other words, students need to do more than just look at corrections. They need to engage with them 40
References Alexander, L. G. (1988). Longman English Grammar. Longman UK Limited. Anderson, N. J. (2005). L2 strategy research. Handbook in second language teaching and learning, 757-775. Baker, W., & Boonkit, K. (2004). Learning strategies in reading and writing: EAP contexts. RELC Journal, 35(3), 299-328.
Barnawi, O. Z. (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.
Bauer, L., & Nation, P. (1993). Word families. International Journal of Lexicography, 6, 254-279. Biber, D., Nekrasova, T., & Horn, B. (2011). The Effectiveness of Feedback for L1-English and L2-Writing Development: A Meta-Analysis. ETS Research Report RR-11-05). Princeton,
NJ: ETS. Bootchuy, T. (2008). An Analysis of Errors in Academic Writing by a Group of First Year Thai
Graduates
Majoring
in
English.
Downloaded
from:
research.rdi.ku.ac.th/world/cache/57/TiptidaBOOAll.pdf Casanave, C. P. (1994). Language development in students' journals. Journal of Second Language Writing, 3(3), 179-201. Cohen, J. (1988). Statistical power analysis for the behavioral sciences. Lawrence Erlbaum. Corder, S. P. 1981. Error Analysis and Interlanguage. Oxford University Press. Cumming, A. (2001). Learning to write in a second language: Two decades of research. IJES, International Journal of English Studies, 1(2), 1-23. Daller, H., Milton, M., & Treffers-Daller, J. (2007). Editor’s introduction: Conventions, terminology and an overview of the book. In H, Daller., J, Milton., & J, TreffersDaller (Eds), Modelling and assessing vocabulary knowledge. Cambridge University Press. Dulay, H., Burt, M., & Krashen, S. (1982). Language Two. Oxford University Press. 41
Dulger, O. (2011). Meta-cognitive strategies in developing EFL writing Skills. Contemporary Online Language Education Journal, 2011, 1(2), 82-100. Dyer, B. (1996). L1 and L2 composition theories: Hillocks' ‘environmental mode ‘and taskbased language teaching. ELT Journal, 50(4), 312-317. Ediger, M. (2006). Writing in the mathematics curriculum. Journal of Instructional Psychology, 33(2), 120. Elbow, P. (1999). Options for responding to student writing. In R. Straub (Ed.) Ellis, R. (1994). The study of second language acquisition. Oxford University Press, USA.
Ellis, R. (2009). A Typology of Written Corrective Feedback Types. ELT Journal, Vol, 63 Etherton, A. R. B. (1997). Error Analysis: Problems and Procedures. ELT Journal, Vol, 32 67-78.
Faigley, L., & Witte, S. (1981). Analyzing Revision. College Composition and Communication, 32(4), 400-414. Ferris, D. R. (1997a). The influence of teacher commentary on student revision. Tesol Quarterly, 31(2), 315-339.
Ferris, D. R. (2004b). The Grammar Correction Debate in L2 Writing: Where are we, and where do we go from here?(and what do we do in the meantime…?). Journal of Second Language Writing, 13(1), 49-62. Flower, L., & Hayes, J. R. (1981). A cognitive process theory of writing. College composition and communication, 32(4), 365-387.
Hampton Press.
Jackson, H., & Ze Amvelu. (2006). Words, meaning and vocabulary: An introduction to modern English lexicology. Continuum. Hanaoka, O. (2007). Output, noticing, and learning: An investigation into the role of spontaneous attention to form in a four-stage writing task. Language Teaching Research, 11(4), 459-479. Hashemi, Z., & Amerian, M. (2011). The Impact of Journal Writing on the Correct Use of Grammar among EFL Students of Rafsanjan Vali-E-Asr University. The English Teacher, Vol, 45. Hemchua, S., & Schmitt, N. (2006). An Analysis of Lexical Errors in The English Compositions of Thai Learners. Prospect, Vol, 21. 42
Hillocks, G. (1984). What works in teaching composition: A meta-analysis of experimental treatment studies. American Journal of Education, 93(1), 133-170. Hong-Nam, K., & Leavell, A. G. (2006). Language learning strategy use of ESL students in an intensive English learning context. System, 34(3), 399-415. Horning, A. & Robertson, J. (2006). Basic writers and revision. In A. Horning & A. Becker (Eds.), Revision: History, theory, and practice (pp. 50-62). Indiana: Parlor Press. Hwang, A. J. (2010). A Case Study of the Influence of freewriting on the Writing Fluency and Confidence of EFL College-Level Students. Second Language Studies, Vol, 28.
Hyland, K. (2003). Second Language Writing. Oxford University Press. Izumi, S. (2002). Output, input enhancement, and the noticing hypothesis. Studies in Second Language Acquisition, 24(4), 541-577.
Kellogg, R. T., Mertz, H. K., & Morgan, M. (2010). Do gains in working memory capacity explain the written self-disclosure effect? Cognition and Emotion, 24(1), 86-93. Kellogg, R.T. (2008). Training writing skills: A cognitive developmental perspective. Journal of writing research, 1(1), 1-26
Kietlinska, K. (2006). Revision and ESL students. In Horning, A. & Becker, A. (Eds.), Revision: History, Theory, and Practice (pp.63-87). USA: Parlor Press. Kollar, I., & Fischer, F. (2010). Peer assessment as collaborative learning: A cognitive perspective. Learning and Instruction, 20(4), 344-348. Kumar, R. (1996). Research Methodology: A Step-by-Step Guide for Beginners. London: SAGE Publications. Lawson, M. J., & Hogben, D. (1996). The Vocabulary‐Learning Strategies of Foreign‐Language Students. Language learning, 46(1), 101-135. Lennon, P. (1990). Error: Some Problems of Definition, Identification, and Distinction. Applied Linguistics, Vol, 12 180-196. Lessard-Clouston, M. (1997). Language learning strategies: An overview for L2 teachers. The Internet TESL Journal, 3(12), 69-80. MacGowan-Gilhooly, A. (1991). Fluency first: Reversing the traditional ESL sequence. Journal of Basic Writing, 10(1), 73-87. Mackey, A. (2006). Feedback, noticing and instructed second language learning. Applied Linguistics, 27(3), 405-430. 43
MacMillan, F. (2007). The Role of Lexical Cohesion in The Assessment of EFL Reading Proficiency. Arizona Working Papers in SLA & Teaching, Vol, 14 75-93. McGroarty, M. E., & Zhu, W. (1997). Triangulation in classroom research: A study of peer revision. Language Learning, 47(1), 1-43. Miller, M., & Veatch, N. (2010). Teaching literacy in context: Choosing and using instructional strategies. The reading teacher, 64(3), 154-165. Nation, P., & Coady, J. (1988). Vocabulary and Reading. Vocabulary and language teaching, 97, 110. O'Malley, J. M., Chamot, A. U., Stewner‐Manzanares, G., Kupper, L., & Russo, R. P. (1985). Learning strategies used by beginning and intermediate ESL students. Language learning,
35(1), 21-46.
Oxford, R. L. (1990). Language learning strategies: What every teacher should know.
Boston: Heinle & Heinle.
Oxford, R. L. (2003). Language learning styles and strategies: Concepts and relationships. IRAL, 41(4), 271-278. Peyton, J. K., & Staton, J. (1991). Writing Our Lives: Reflections on Dialogue Journal
Writing with Adults Learning English. Language in Education Series, No. 77. Phoocharoensil, S. (2011). Collocational Errors in EFL Learners Interlanguage. Journal of Education and Practice, Vol, 2. Pinker, S. (1995). The Language Instinct. Penguin Group. Plonsky, L. (2011). Study Quality in SLA: A Cumulative and Developmental Assessment of Designs, Analyses, Reporting Practices, and Outcomes in Quantitative L2 Research. ProQuest LLC. Pongsiriwet, C. (2001). Relationships among Grammatical Accuracy, Discourse Features, and The Quality of Second Language Writing: The Case of Thai EFL Learners. Downloaded from: https://eidr.wvu.edu/eidr/documentdata.eIDR?documentid=2024 Qi, D. S., & Lapkin, S. (2001). Exploring the role of noticing in a three-stage second language writing task. Journal of Second Language Writing, 10(4), 277-303. Raimes, A. (1983). Techniques in Teaching Writing. Oxford University Press. Richards, C. J. (1980). Second Language Acquisition: Error Analysis. Annual Review of Applied Linguistics, Vol, 1 91-107. 44
Rinnert, C. and Kobayashi, H. (2001), Differing Perceptions of EFL Writing among Readers in Japan. The Modern Language Journal, 85: 189–209. doi: 10.1111/00267902.00104 Sasaki, M. (2000). Toward an empirical model of EFL writing processes: An exploratory study. Journal of Second Language Writing, 9(3), 259-291. Sattayatham, A., & Rattanapinyowon, P. (2008). Analysis of Errors in Paragraph Writing by First Year Medical Students from the Four Medical Schools at Mahidol University. Silpakorn International Journal, Vol 8 17-38
Schmidt, R. (1993). Awareness and second language acquisition. Annual review of applied linguistics, 13(1), 129-158.
Silva, T. (1990). Second language composition instruction: Developments, issues, and directions in ESL. Second language writing: Research insights for the classroom, 11-
23. Silva, T. (1993). Toward an understanding of the distinct nature of L2 writing: The ESL research and its implications. Tesol Quarterly, 27(4), 657-677.
Swain, M., & Lapkin, S. (1995). Problems in output and the cognitive processes they generate: A step towards second language learning. Applied linguistics, 16(3), 371391. Truscott, J. (1998). Noticing in second language acquisition: A critical review. Second Language Research, 14(2), 103-135. Truscott, J. (1999a). The case for “The case against grammar correction in L2 writing classes”: A response to Ferris. Journal of Second Language Writing, 8(2), 111-122. Truscott, J. (2006b). The case against grammar correction in L2 writing classes. Language learning, 46 (2), 327-369. Villamil, O. S., & De Guerrero, M. (1996). Peer revision in the L2 classroom: Socialcognitive activities, mediating strategies, and aspects of social behavior. Journal of second language writing, 5(1), 51-75. Wong, L. L., & Nunan, D. (2011). The learning styles and strategies of effective language learners. System, 39 (2), 144-163. Yule, G. (1996). The study of language. Cambridge University Press.
45
Zamel, V. (1983). The Composing Processes of Advanced ESL Students: Six Case Studies. TESOL Quarterly, Vol, 17 6-28. Bootchuy, T. (2008). An Analysis of Errors in Academic Writing by a Group of First Year Thai
Graduates
Majoring
in
English.
Downloaded
from:
research.rdi.ku.ac.th/world/cache/57/TiptidaBOOAll.pdf Sattayatham, A., & Rattanapinyowon, P. (2008). Analysis of Errors in Paragraph Writing by First Year Medical Students from the Four Medical Schools at Mahidol University. Silpakorn International Journal, Vol 8 17-38
Ellis, R. (2009). A Typology of Written Corrective Feedback Types. ELT Journal, Vol, 63 Jackson, H., & Ze Amvelu. (2006). Words, meaning and vocabulary: An introduction to
modern English lexicology. Continuum. Horning, A. & Robertson, J. (2006). Basic writers and revision. In A. Horning & A. Becker (Eds.), Revision: History, theory, and practice (pp. 50-62). Indiana: Parlor Press. Kellogg, R.T. (2008). Training writing skills: A cognitive developmental perspective. Journal of writing research, 1(1), 1-26 Phoocharoensil, S. (2011). Collocational Errors in EFL Learners Interlanguage. Journal of Education and Practice, Vol, 2. Pongsiriwet, C. (2001). Relationships among Grammatical Accuracy, Discourse Features, and The Quality of Second Language Writing: The Case of Thai EFL Learners. Downloaded from: https://eidr.wvu.edu/eidr/documentdata.eIDR?documentid=2024 Yule, G. (1996). The study of language. Cambridge University Press.
46
Appendix 1 A sample of student NCFs with reconstructions in bold. Nominal: About a few step (About a few steps) a lot of tourist (a lot of tourists)
It is a waste time...
(It was a waste of time) In front the door (In front of the door)
The police warned ours about vehicles… (The police warned us about our vehicles)
I put on helmet and drove away. (I put on my helmet and drove away) Verbal: …because it move fast, (…because it moves fast) My town have only one market. (My town has only one market) I love the food in the market because it cheap and delicious. (I love the food in the market because it is cheap and delicious.) Moreover, it’s shows good student actions (Moreover, it shows good student actions.) If you saw the temple, you will thinking it is great. (If you saw the temple you will think it is great.) 47
At the market I love bargain with seller. (At the market I love to bargain with the seller.) Lexical People in Issan and Laos can understandable each other. (People in Issan and Laos can understand each other.) It is more convenience to live there (It is more convenient to live there)
People buy stricky rice. (People buy sticky rice.)
In my view, the resalt is bad.
(In my view, the result is bad.) The water in my river is very pure. (The water in my river is very clean)
It makes you understand the problems (It helps you understand the problems.) The best thing I like on it is… (The best thing I like about it is…)
48
Appendix 2 The following, taken from Oxford (1990), were presented to the students as tasks for revision in their journals. This constituted the treatment. MEMORY STRATEGIES 1. Placing corrections in a different context: This could include placing quantifying corrections, for example, in contexts other than the original work. If a student writes ‘many of people’ and sees the correction ‘many of the people’, she or he could practice writing many of the students, many of the boats, many of the teachers in full sentences or just clause form in
their journal. This may help to focus attention on the NCF by applying it to another context.
2. Grouping means creating meaningful units; words could be grouped according to families; words can be grouped according to part-of speech; students could group assignment errors
according to the type. For example, students can keep all pronoun errors together in the same section of their note-book. 3. Associating/Elaborating: This could include comparing the mother language with the L2
language, comparing what is similar or what is different. When a NCF is made which can be traced to Thai then the student has to make note of this and write down several examples. 4. Semantic Mapping: Drawing mind maps to help remember grammar rules and vocabulary. Semantic synonym maps can help students to group synonyms according to shades of meaning. COGNITIVE STRATEGIES 5. Repeating: This means doing something over and over again. This could involve writing the correction several times. This may appear simple and somewhat redundant, but experts in L2 writing, Kellogg (2010), claim that repeating correct language is essential. 6. Formally Practicing: Students need overt practice. Oxford (1990) suggests copying correct English from a book or magazine or the Internet. This gives students practice in writing correct English and gives them a feel for what it looks like on paper. Again, this may strike some as being simple and very basic, but it is deemed essential by the experts. Sometimes the things the students need to do are simple and basic, but important. 49
7. Recognizing Patterns: This is Oxford’s (1990) tip of the hat to lexis, fixed phrases, single lexical units and the like. It may include verb plus preposition patterns, which are not governed by any rules. 8. Recombining: Oxford (1990) states: ‘The strategy of recombining involves constructing meaningful sentence or longer expression by putting together known elements in new ways’. In other words, practice joining simple sentences using conjunctions. This is often a staple of writing textbooks. 9. Deductive reasoning: This can involve looking at general English rules and applying them in writing. This would suggest the student looks up tutorials on the web about, for example, subject verb agreement, taking the rule, writing it in their journals, and using it to write. 10. Analyzing Expressions: This refers to determining meaning by breaking it down into
parts to understand the whole; looking at a corrections and identifying nouns phrases and verbs. 11. Analyzing contrastively: This can involve comparing the languages; looking for similarities and differences, making note of them in journals, and keeping these in mind while
writing.
50
