A reciprocal peer review system to support college ... Student... · A reciprocal peer review system to support college students’ writing ... Lundstrom & Baker ... Three learning

A reciprocal peer review system to support collegestudents’ writing_1059 687..700

Yu-Fen Yang

Yu-Fen Yang is a professor of Graduate School of Applied Foreign Languages at National Yunlin University of Scienceand Technology. Address for correspondence: Yu-Fen Yang, National Yunlin University of Science and Technology,123, University Road, Section 3, Douliu, Yunlin 640, Taiwan. Tel.: 886-5-534-2601 Ext. 3136; Fax: 88655312058; email: [email protected]

AbstractAs students’ problem-solving processes in writing are rarely observed in face-to-faceinstruction, they have few opportunities to participate collaboratively in peer review toimprove their texts. This study reports the design of a reciprocal peer review system forstudents to observe and learn from each other when writing. A sample of 95 under-graduate students was recruited to construct texts with the support of web-based recip-rocal peer review in the processes of modelling, coaching, scaffolding, articulation,reflection and exploration. The results of the study revealed that these six processeshelped students externalise and visualise their internal writing processes so that theycould observe and learn from peers in writing as well as support peers in making textrevisions. During their extensive and reciprocal interactions with various peers, studentsaddressed mutual concerns in each other’s text revisions. They constructed collaborativelanguage knowledge for text improvement as local revisions (grammatical corrections)and global revisions (corrections on the development, organization or style of a text)were made in their final texts. The students’ perceptions towards text improvement inthis web-based peer review of modelling, coaching, scaffolding, articulation, reflectionand exploration are also discussed in this study.

IntroductionImproving writing skills is important for college students since ‘professional and academicsuccess in all disciplines depends, at least in part, upon writing skills’ (Cho & Schunn, 2007, p.409). Many studies have emphasised that writing is no longer regarded as an individual task butrather may be supported by feedback from peers to improve the text (eg, Liou & Peng, 2009;Lundstrom & Baker, 2009; Min, 2006; Vass, Littleton, Miell & Jones, 2008). Students can benefitmore from different peers if they have gone through similar writing situations and difficulties inproviding helpful suggestions and revisions (van Leeuwen, Tiesinga, Jochemsen & Post, 2009).Through the reciprocal process of peer review in writing, students are able to develop new ideasand perspectives as well as improve their writing skills (Lundstrom & Baker, 2009).

Different from face-to-face writing instruction, web-based peer review allows students to visualisetheir thinking and behaviors in solving problems of writing, where the thinking process is com-plicated and not easily observed (Liu, 2005; Patel, Kinshuk & Russel, 2003). Through web-basedpeer review, students are able to observe each peer’s writing processes recorded online and engagein interaction to receive peers’ assistance or to provide scaffolding to other peers for text improve-ment. In the pursuit of text improvement, they not only benefit from receiving feedback of peersbut also from providing assistance to them (Cheng & Ku, 2009; Duran & Monereo, 2005).

British Journal of Educational Technology Vol 42 No 4 2011 687–700doi:10.1111/j.1467-8535.2010.01059.x

© 2010 The Author. British Journal of Educational Technology © 2010 Becta. Published by Blackwell Publishing, 9600 Garsington Road, Oxford OX42DQ, UK and 350 Main Street, Malden, MA 02148, USA.

However, previous studies have shown several drawbacks to peer review in writing. First, the rateof adopting peers’ revision is low when peers engage in non-revision-oriented feedback such aschatting and making complimentary remarks, rather than revision-oriented feedback such assuggestions, clarifications and critical evaluations (Liou & Peng, 2009). Second, peer review inwriting is restricted to the lexical level for students often fail to participate in the discussion of textcontent or organisation (Cho & Schunn, 2007; Lundstrom & Baker, 2009; Storch, 2005). Third,students have few opportunities to reflect upon peer review as they cannot compare the differencesbetween the different feedback from various peers (van Leeuwen et al, 2009; Vass et al, 2008; Xiao& Lucking, 2008). Fourth, students will not benefit from peer review until they are asked to rewritetheir texts based on peers’ revision (Braine, 1997; Paulus, 1999). Thus, some important issuesneed to be addressed to ensure the quality and effectiveness of peer review in writing.

Web-based peer review processesThe development of the web-based peer review processes described in this study is based on thetheory of cognitive apprenticeship where six processes are included for externalising knowledge—modelling, coaching, scaffolding, articulation, reflection and exploration (Collins, Brown &Newman, 1989). As writing is usually combined with complex internal processes, how to trans-form one’s thinking into a visible form is important in web-based peer review. Through theweb-based peer review processes conducted in this study, students were able to observe peers’writing process, provide revisions both at the lexical and discourse level, communicate with peers,compare one’s own texts with those revised by peers and decide whether to accept or reject peers’suggestions.

In the process of modelling, all of students’ texts are presented in the online system for students toobserve how their peers read and revise others’ texts. With modelling, students can learn reviewingskills more effectively without repeated struggles of trial and errors (Wang & Bonk, 2001; Williams,1992). In addition, students have to be more responsible in revising others’ texts because theirreviews could also be models for others (Yoshimura, 2009).

In the process of coaching, students read and revise peers’ texts and provide local revisions(corrections of grammatical errors). ‘Coaching may serve to direct students’ attention to a pre-viously unnoticed aspect of the task or simply to remind the student of some aspect of the taskthat is known but has been temporarily overlooked’ (Collins et al, 1989, p. 481). It is reported thatstudents do not detect their own grammatical errors in writing (Cho & Schunn, 2007; Li, 2006).In this study, peers’ local revisions are considered as coaching as these errors are generallyunnoticed by students and regarded as minor mistakes in writing (Cho & Schunn, 2007; Li,2006).

Different from the process of coaching which refers to peers’ direct corrections on the texts,scaffolding refers to peers’ suggestions for student writers to reorganise their texts by themselves.This support helps students manage complex tasks (Chee, 1995). In this study, global revisions(corrections to the development, organisation or style of a text) are considered as scaffolding asthey enable students to reexamine and reorganise their texts in terms of development, organisa-tion or style. This may result in a student writer’s deletion or addition of a new paragraph or thereorganisation of an entire text. Thus, global revisions are not limited to the lexical level butrather expand upon discourse level. Both local and global revisions are important to students’ textimprovement (Cho & Schunn, 2007; Li, 2006).

Articulation involves any method of encouraging students to express their knowledge, reasoningor problem-solving processes in peer review. For example, students are encouraged not only tomake revisions but also to provide reasons for revisions. Such tasks require students to participatein generating language knowledge and evaluating the writing outcomes. In reflection, studentscompare their own problem-solving processes in writing with those of peers (Liu, 2005). Such

688 British Journal of Educational Technology Vol 42 No 4 2011

© 2010 The Author. British Journal of Educational Technology © 2010 Becta.

comparisons aid students in diagnosing their difficulties and adjusting their revision strategiesuntil they achieve the goal of text improvement. Exploration enables students to become inde-pendent learners as they are encouraged to select appropriate and effective peer review forimproving their own texts. For example, students are encouraged to revise their texts into finaldrafts based on their evaluation of the feedback from peers. As a result, students are expected tobecome more active learners.

Purpose of this studyThe purpose of this study is to engage college students in reciprocal peer review for text improve-ment using an online system. Participants in the system are encouraged to externalise theirthinking throughout the entire process of peer review in writing. Among the reciprocal interac-tions with peers, the focus is placed on the processes of peer review, because the students’decisions for making text revisions can be observed (Hansen & Liu, 2005; Storch, 2005). Thisfocus differs from many studies whose emphasis is on the students’ written products. In this paper,three research questions are addressed: (1) how do the processes of peer review improve students’texts in the online system?, (2) how do students receive support from and provide support to peersin web-based peer review?, (3) how do students perceive their text improvements in web-basedpeer review?

MethodTo support college students in reciprocal peer review, an online system was designed for them toexperience the processes of modelling, coaching, scaffolding, articulation, reflection and explo-ration. In the system, students followed each process in peer review with a common goal of textimprovement.

ParticipantsThree learning English as a Foreign Language (EFL) writing classes were randomly selected froma university of science and technology in central Taiwan. In these three writing classes, therewere 95 students with a common background in two respects: (1) all of them had passed theintermediate level of General English Proficiency Test, a nationwide screening test, administered bythe university in the selection of students who wish to major in English; and (2) they had takenthe same writing class for 2 years in this university and were in the third year of their studies. Theobjective of these three writing classes was to develop students’ writing skills through the supportof peer review. Students were randomly assigned a user identifier (ie, P1–P95) in order to beanonymous in the online system. They were encouraged to interact with their peers by revisingpeers’ texts and evaluating peers’ suggestions before rewriting their first drafts.

System developmentDesigning an extensive and reciprocal learning environment is important to promote peer reviewin writing (Patel, Kinshuk & Russel, 2003; Woolley & Jarvis, 2007). Six processes—modelling,coaching, scaffolding, articulation, reflection and exploration—were implemented in the systemto encourage students’ thinking and problem-solving processes in writing. Each process isdescribed as follows.

1. Modelling: Some students might have difficulties in writing a text that their teacher hasassigned. One possible way to overcome their difficulties is to read peers’ texts before postingtheir own texts in the system. As shown in Figure 1, the number of clicks made whenreading each individual student’s text is presented in the system; students are also allowed touse the function key ‘My Collection’ to collect their favorite peers’ texts for modelling. That is,peers’ texts can serve as models for them to develop and organise their own texts on the sameline.

A reciprocal peer review system 689


2. Coaching: A list of error types in local revision is provided by the system for students to selectwhen checking peers’ texts (Figure 2). The definitions and examples for each error type arealso provided in the system for students to look up. After the students’ selection of errortypes, their local revisions are automatically highlighted with a number and a triangle icon(Figure 3). With peers’ coaching in local revisions, students can pay attention to the gram-matical errors that they have not noticed.

3. Scaffolding: A list of error types in global revision is provided by the system for students torefer to (Figure 4). A definition and example of each error type is also presented for studentsto look up.

4. Articulation: In the system, students are allowed to discuss each revision in a dialogue box.In the dialogue box (Figures 5 and 6), peers’ selected error type and explanations for a localor global revision are presented. Student writers then evaluate the revision and give feedbackto peers’ comments. Through articulation, students externalise their knowledge and think-ing process in making text revisions.

5. Reflection: Students can use the Diff Engine embedded in the system to compare their textswith the ones revised by peers. As shown in Figure 7, the differences between the twoselected texts are detected and highlighted by boldface and crossing-out. While reflecting onpeers’ revision, students decide whether to accept it or not. With reflection, students canacquire multiple perspectives from peers on the same topic.

6. Exploration: In order to become independent writers, students are encouraged to revisetheir first drafts into their final drafts based on their evaluation of peers’ comments.

Title Author Date & Time No. of Clicks in reading the text

My Collection

Literacy (Literacy in the United States)

P16 2008-11-23 21:17:15 59 6

HW 14: The importance of literacy (Literacy in the United States)

P03 2008-11-23 20:30:17 57 5

HW14: Unclear definition of literacy (Literacy in the United States)

P01 2008-11-23 20:10:34 82 7

Figure 1: Peers’ texts serve as model texts in the system

Singular and plural error Spelling error

Erroneous part of speech …

Run-on sentence Fragment

Others

Run-on sentence The night was cold we forgot to bring our coats. (X)

The night was cold, and we forgot to bring our coats. (O).

Figure 2: Error types in local revision

Everyone wants to improve [1] hishim or her knowledge.

Figure 3: A local revision highlighted in the system



Different versions of a revised text are shown in the system. These enable students toraise their awareness to take steps in improving their texts. After posting their final drafts,students were also asked to evaluate their revisions for the final drafts by reporting theirreasons for revision. With evaluation, students are responsible for their text revision andimprovement.

Students’ actions in each of the six processes of peer review are recorded in the trace result of thesystem. This is available not only for the teacher to monitor students’ actions in the system, butalso for students to visualise their writing processes. From the trace result, the process of peerreview for text improvement can be examined.

Perspective Article Construction

Organization Clarity of Purpose

Development of Purpose Inter-paragraph Transition

Correctness in the Standard Conventions Others

Clarity of purpose The main ideas delivered were clear and precise.

Figure 4: Error types in global revision

Everyone wants to improve [1] hishim or her knowledge.

[1] Case Misuse

It should be “improve one’s knowledge.”

Evaluation:

Feedback: Thank you for your reminder.

Figure 5: Articulation for local revision

Suggestion

[Organization] The organization is not good. You need a topic sentence at the beginning of a paragraph and a concluding sentence at the end.

Evaluation:

Feedback

Thank you for your suggestion. I will revise it.

Figure 6: Articulation for global revision



Procedures of data collectionThe present study was conducted between September 15th, 2008 and June 15th, 2009. Ninety-five undergraduate students were encouraged to write and rewrite their texts by themselves inand after class. They were recommended to finish the following tasks in each writing cycle:

1. write and post the first drafts of a reaction essay;2. revise peers’ first drafts by providing local and global revisions;3. rewrite their own first drafts into final drafts after evaluating peers’ local and global revisions;

and4. write a reflective journal to express their perceptions towards web-based peer review in

writing.

Each task took 3–4 days so that the participants could complete each writing cycle in approxi-mately 3 weeks.

Procedures of data analysisData collected in this study, including students’ first drafts, final drafts and reflective journals,were analysed by content analysis. This procedure helped the researcher discover and describe thefocus of individual, group, institutional or social attention, and allowed the researcher to makeinferences (Patton, 1990; Weber, 1990). Four stages of content analysis were conducted in thisstudy: coding, categorization, description and interpretation. First, the researcher and theresearch assistant coded meaningful statements with highlights while reading students’ first/final drafts and reflective journals. Next, the meaningful statements were assigned into categoriesin terms of local and global revisions for each participant, and the students’ text improvementwas assessed in terms of local and global revisions in their final drafts. Then, the researcherdescribed the statements by presenting and summarising the main ideas. Finally, the researcherinterpreted the main ideas by offering explanation, drawing conclusions and making inferences.The inter-rater reliabilities of students’ first draft, final drafts and reflective journals were 0.78,0.80 and 0.83, respectively. Any disagreement between two raters (the researcher and theresearch assistance) was resolved by discussion. Data interpretation driven by these researchmethods is further explained in the following sections.

ResultsThe results of this study are described in three sections: (1) implementation of peer reviewprocesses for text improvement, (2) students’ reciprocal interactions in web-based peer review, (3)

Version Choice1

Choice2

Author Date and Time No. of Clicks

1 P16 2008-11-23 21:17:15 62

2 P07 2008-11-25 21:36:06 12

3 P04 2008-11-27 01:31:16 8

Show differences

…I believe the literacy rate in U.S. is right because it has a good education.education and the immense achievement of U.S. The potential is there.

Compare the chosen versions

Figure 7: Comparison between one’s own text and the one revised by a peer



students’ perceptions towards text improvement in web-based peer review. One student, studentA, was randomly selected as an example to show the influence of peer review processes on her textimprovement. Her reciprocal interactions with peers to support and to be assisted by peers arealso presented. In addition, a sample of 30 students’ reflective journals, 10 from each writingclass, was randomly selected to examine students’ perceptions in general towards text improve-ment in web-based peer review.

The influence of peer review processes on student A’s text improvementFrom the trace result, it was found that student A read peers’ texts before writing and posting hertext in the system. As shown in Figure 8, she read 11 peers’ texts before posting her own text;seven of them were reviewed more than twice. Among the 11 peers, student A read two peers’texts (P24 and P42) for four times. In repeatedly reading peers’ texts, student A was engaged inmodelling to observe peers’ writing processes and obtain some ideas for her own writing.

In addition to reading the model texts of 11 peers, student A also interacted with three peers—P24,P19 and P38—who contributed 1, 11 and 5 local revisions, respectively, and suggested globalrevisions. After posting her first draft, student A read peers’ local revision (coaching), read peers’global revision (scaffolding), gave feedback to peers’ comments (articulation) and used the Diffengine to compare her text with the one revised by peers (reflection). The frequencies of theseinteractions with the three peers are presented in Figure 9. It was found that each peer madedifferent contributions to student A’s peer review in writing. For example, in coaching, student Aread local revisions from P19 for 49 times while she read P24’s for three times and P38’s for 31times.Thus, student A read P19’s local revisions much more than those of the other two peers.Thismight have resulted from P19’s better language proficiency in providing more local revisions.

With the support of modelling, coaching, scaffolding, articulation and reflection from peers,student A finally engaged in exploration to revise her first draft into a final draft. Figure 10 showsa comparison between her first and final draft by the Diff Engine in the system. Each revision isindicated in boldface for newly added words or crossing-out for deleted words. Student Aimproved her text in terms of local and global revision, eg, by revising ‘label’ into ‘labeled’ in thesecond paragraph and adding a new sentence in the concluding paragraph.

Student A and her peers’ thinking process and language knowledge construction for each revi-sion could be visualised through the trace result in the system. Table 1 shows student A’s peerreview processes in one of her local revisions. In coaching, P19 revised students A’s text ‘enoughwell’ into ‘well enough’. In articulation, P19 gave the comments and student A provided feedbackto P19: ‘I often misuse “enough”. Thank you for reminding me the mistake. I will keep this usagein mind’. In exploration, student A made a local revision in the final draft and evaluated that ‘well

Figure 8: The frequency of student A read peers’ texts before posting her first draft



enough’ was the correct usage. From student A’s and her peers’ reflective dialogues in the system,student A’s and P19’s thinking process and language knowledge construction for peer review inwriting could be identified.

Table 2 shows student A’s peer review processes for global revision in her final draft. In scaffold-ing, the three peers provided global revision. In articulation, student A agreed with the threepeers’ suggestions to write more of her own opinions towards the reading text provided by theteacher. They thought that because the assignment is a reaction essay, student A should expressher opinions more instead of just summarising the ideas in the reading text. In reflection, studentA compared her own text with those of peers to learn different perspectives. In exploration,student A added a sentence in the conclusion in her final draft. She thought that her text could beimproved by expressing more opinions to support the main idea in her final draft.

With the six processes of peer review in the system, student A and her peers’ thinking process foreach text revision can be externalised. Their learning experiences become more meaningful tothemselves.

Student A’s extensive and reciprocal interaction with peersFrom student A’s trace result, it was found that she not only received support in peer review fromthe 11 peers shown in Figure 8 and the three peers shown in Figure 9, but also provided supportin peer review to the other three peers—P21, P04 and P14—as shown in Figure 11. The inter-action between student A and her peers is presented to show the extensive and reciprocal

1

49

12

22

9

31

53

68

3 30

5

10

15

20

25

30

35

40

45

50

Coaching: readlocal revision

Scaffolding: readglobal revision

Articulation: givefeedback

Reflection: use Diffengine

Action

Frequency

P24

P19

P38

Figure 9: Student A’s actions after receiving peers’ revised texts



interaction possible in web-based peer review. It was found that through web-based peer review,students were able to interact with various peers to receive support from and provide help to peersfor text improvement.

Among the extensive and reciprocal interactions, collaborative language knowledge was con-structed. According to the trace result (Table 3), student A revised two peers’ texts (actions #14and #17) before receiving peers’ revision. After reading P19’s revisions (actions #18 and #26),she started to revise P14’s text (action #34). During student A’s support to and from peers inwriting, mutual concerns about local and global revision were addressed. In local revision, theynoticed that past tense should be used to describe events in the past (actions #14, 26 and 34); inglobal revision, more supporting ideas were recommended (actions #17, 18 and 36). Thus, thelinguistic knowledge they shared and constructed was externalised in the system and could beobserved by peers if they did not know how to support each other.

Students’ perceptions towards text improvement in web-based peer reviewFrom the students’ reflective journals, their perceptions towards the six processes of peer reviewcould be investigated (Table 4). In modelling, students liked to read and collect peers’ texts before

Show differences

According to the article "Literacy in the United States" (author and date were not provided), it indicated that although the literacy rate of the United States was estimated to be 99.7%, actually it still had some arguments about the survey and differentthe definitionsdefinition of the literacy.

First, governments labeledlabel individuals who can read a couple of thousand simple words they learned by sight in the first four grades in school as literate. However, in "The World Factbook" the CIA claimed the controversial definition of literacy as being able to read and write when a person is 15 years old or older.older, Government'sgovernment's definition was only marginally literate. Moreover, the comprehensive study argued that such adults were functionally illiterate that they can't read enough well enough to hold a job. The study showed that 21% to 23% adult Americans were not able to locate information in text, and earned a yearly average of 2,105 that less than the poverty level threshold for an individual. Jonathan Kozol in his book "Illiterate America" explained that the Census Bureau reported literacy rates of 99% were not fully correct, because of the object of studycould not can't representrepresented the whole population and the method of determining literacy was certain to underestimate illiteracy.

In my opinion, the definition of literacy should be established firmly and the method of determining should be improved. If someone is considered to be illiterate, I think government can encourage him to improve his reading and listening skill or offer him more learning resources so that he can have more opportunity of finding a good job. No matter in the work place or even the whole society, I think literacy was an essential ability to support people's life.

Global revision

Figure 10: Comparison of student A’s first and final draft

Table 1: Student A’s peer review processes for one of the local revisions

Peer review processes Content

Coaching P19 revised ‘enough well’ into ‘well enough’ in student A’s textArticulation P19’s comment: ‘Well enough’ is the correct usage.

Student A’s feedback to P19: I often misuse ‘enough’. Thank you for remindingme the mistake. I will keep this usage in mind.

Exploration Student A made a local revision by revising ‘enough well’ into ‘well enough’.Student A’s evaluation of P19’s revision:‘Well enough’ is the correct usage.



posting their own texts in the system. In coaching, students received a lot of local revisions(grammatical corrections) from peers that they had not noticed before. In scaffolding, studentsregarded global revisions as useful suggestions for them to improve their text organisation. Inarticulation, students liked to discuss local revision and global revision with peers. In reflection,students compared their first drafts with the one revised by peers to have different perspectives. Inexploration, students expressed more confidence on final drafts because they had evaluatedrevisions in the final draft.

Most students appreciated peer review in writing because they thought that peer support washelpful for text improvement (Table 5). They indicated that peers’ texts served as models for themto plan for their writing. Although the peers were not experts, they were able to provide relevantand timely assistance and encouragement for text improvement, as they encountered similardifficulties in their own writing tasks.

Students not only benefited from peer review but also contributed to peers’ text improvement(Table 6). They made every effort to revise their peers’ texts and they encouraged them to make

Table 2: Student A’s peer review processes for the global revision

Peer reviewprocesses Example

Scaffolding P24’s global revision: You could write more about your opinions in the conclusion.P19’s global revision: Your summary for the article is good, but you should present your

reflection to the article.P38’s global revision: You could express more about your opinions by pointing out how

to improve the literacy in the United States.Articulation Student A’s feedback to P24: I will write more about my opinions in the conclusion to

support my main points. Thank you for your suggestion.Student A’s feedback to P19: My reflection to the article is not enough. It is necessary

for me to improve it. Thank you for your suggestion.Student A’s feedback to P38: It’s a great idea. Thank you.

Reflection After student A compared her text with those of her peers, she developed differentperspectives.

Exploration Student A added a sentence in the conclusion to express her opinions.Student A’s evaluation: My text was improved by expressing more about my opinions to

support my main idea.

Student

P39

P34

P25P42

P15

P08

P09

P04P38

P21P19P14P24

P02

Figure 11: The extensive and reciprocal interactions among student A and her peers



further improvements. The motivation for students to support each other could be fostered bycollaborative language knowledge construction from peer review. In reviewing peers’ texts, theycould learn from and reflect on the strengths and weaknesses of peers’ texts. Furthermore, duringthe interaction with peers, they realised that they had the ability to improve peers’ texts. Thisawareness might encourage them to be more confident in their text revisions and facilitate moreinteractions with peers.

Students were extensively engaged in peer review processes supported by the system (Table 7). Itwas found that the system functioned well for students to observe peers’ writing processes,identify local and global revisions, communicate with peers, reflect on revised versions of a textand facilitate text improvement. Without going through the six processes of peer review sup-ported by the system, students could neither externalise their linguistic knowledge and thinkingprocess in writing nor could they further co-construct new knowledge in terms of local and globalrevisions with peers. Besides, without the externalised knowledge and thinking process presented

Table 3: Constructing collaborative knowledge among student A and her peers

Date/Time Action Content

2008-11-27 17:13:48 #14 Student A providedlocal revision to P21

Student A asked P21 to use past tense todescribe events in the past.

2008-11-27 17:32:12 #17 Student A providedglobal revision to P04

‘Your opinions are seldom expressed in thetext. You should write the introduction first,point out the main ideas, and express yourideas in the end. The organization of the textcould be further improved.’

2008-11-27 17:41:28 #18 Student A read globalrevision from P19

P19 encouraged student A to express more ofher opinions.

2008-11-27 17:51:08 #26 Student A read localrevision from P19

P19 suggested student A to use past tense todescribe things in the past.

2008-11-27 18:08:31 #34 Student A providedlocal revision to P14

Student A asked P14 to use past tense todescribe events in the past.

2008-11-29 15:47:29 #36 Student A read globalrevision from P24

P24 encouraged student A to express more ofher opinions.

Table 4: Students’ perceptions towards peer review processes as revealed in their reflective journals (n = 30)

Statement Participants

A In modeling, I would like to collect good texts toimitate.

P03, P09, P22, P35, P56, P84

B In coaching, I learned a lot from peers’ localrevisions that I had never noticed before.

P02, P05, P18, P22, P29, P33, P47, P53, P60,P67, P73, P86, P93

C In scaffolding, the most helpful support from peerswas their global revisions which improved mytext organization.

P09, P15, P17, P33, P42, P51, P69, P83

D In articulation, I enjoyed the discussion withpeers.

P15, P18, P24, P51, P53, P75, P95

E In reflection, comparing my first draft with theone revised by peers, I could understand theweaknesses of my writing and avoid making thesame mistakes next time.

P22, P24, P51, P60

F In exploration, I believed my final draft was betterthan the first one since I had verified eachrevision.

P09, P18, P42, P53, P63, P95



in the system, they would not be able to observe peers’ writing processes and have opportunitiesto improve their own texts. Thus, the system served as an extensive and reciprocal learningenvironment for students to demonstrate their writing process and improve their texts by prac-tising peer review.

DiscussionFrom the results of this study, various aspects of web-based peer review in writing can beobserved. First of all, this study discusses three aspects of peer review: interaction, impact andattitude. These correspond to the findings of Ferris’s study in 2003 that peer review studiesgenerally fall into three categories—the interactions during peer review, the impact of peerreview on students’ text improvement and students’ attitudes towards peer review. With the sixprocesses of peer review conducted in the online system, these categories can be extended toprovide more detailed information for teachers to design follow-up curriculum.

Moreover, peer review in the online system was reciprocal as students received help from andprovided support to peers at the same time. In order to reach the common goal of text improve-ment, they were engaged in interaction to exchange information and language knowledge. Intheir extensive interactions with various peers, some students would go through the sameproblem-solving activities to construct collaborative language knowledge as common local andglobal revisions were found in their final drafts. However, if the understanding is not mutualbetween a student and a peer, it would be hard for them to continue interactions in text revisions.Teacher intervention then became important to keep the interaction going on until studentscould improve their texts.

Table 5: The impact of peer review on text improvement as revealed in students’ reflective journals (n = 30)


A I read peers’ essays before composing my own. P03, P09, P22, P29, P35, P37, P53, P58, P67,P84, P93

B I would imitate my classmates’ excellent writing. P02, P15, P56, P86C I received a lot of comments and feedback from

peers.P02, P03, P05, P15, P18, P, 22, P29, P33, P42,

P51, P53, P58, P60, P63, P69, P75, P83,P84, P86, P93, P95

D I could learn different perspectives from readingpeers’ essays.

P03, P15, P35, P42, P53, P58, P63, P73, P84

E If I did not understand peers’ comments on mytexts, I would ask them.

P05, P22, P37, P51

F Although peers were not experts in writing, theyknew how to improve my texts because wewere in the similar learning situation.

P02, P09, P17, P24, P29, P33, P37, P47, P56,P67, P69, P75, P86, P95

G With peers’ encouragement and assistance, Icould make better improvement to my texts.

P05, P18, P29, P33, P42, P63, P75, P83

Table 6: Students’ contribution to peers’ text improvement as revealed in their reflective journals (n = 30)


A I did my best to edit peers’ texts and provide suggestions. P02, P05, P22, P69B I was happy to help peers revise their texts. P18, P47, P51, P67, P73, P83C I noticed peers’ strengths and weaknesses in writing. P09, P33, P37, P56, P63, P93D I would compliment peers on their writing. P03, P35, P67E I encouraged my peers to improve their texts by revising. P24, P29, P58F I was surprised that I could help correct peers’ texts. P17, P53, P83, P84, P93, P95G I learned a lot by correcting others’ texts. P18, P29, P37, P58, P63, P86



Finally, the online system with the six processes of peer review effectively supported students’ textimprovement by providing an external model of internal writing processes. The system was anextensive and reciprocal learning environment for students to improve their writing because theentire writing processes were presented in the system for students and teachers to observe. Inthese processes, the diversified thinking processes and language knowledge from peers wereexternalised for students to examine, integrate and transfer into their writing. The visible processof decision making in text revision shown in the system also fostered students’ collaborativewriting and enhanced interpersonal relationship when mutual concerns were addressed. As aresult, students are willing to dedicate their time and effort to improve peers’ texts.

Although students’ peer review is supported by the system described in this paper, there are somelimitations in this study. First, knowledge transfer in peer review is not discussed in this study.There is evidence that some students pass on the knowledge from peers to other students (Baird,2004; Kay, 2007). Future studies could be carried out over a longer time period to observestudents’ knowledge transfer in peer review. Second, in addition to the support of peer review,teachers’ interventions might also facilitate students’ writing processes. Jarvela (1995) suggeststhat teachers can provide strategies for problem-solving activities and gradually transfer respon-sibility to students. With externalised and visible writing processes presented in the system, futurestudies could probe further into the timing and fading of scaffolding in teacher intervention.

AcknowledgementsMy heartfelt thanks go to the two anonymous reviewers for their helpful comments on the earlierdrafts of this paper. This paper was supported by the National Science Council in the Republic ofChina, Taiwan (NSC 97-2410-H-224-016-MY2). The research grant made the continuation ofthis study possible.

ReferencesBaird, C. (2004). Transforming knowledge through mentor-supported cognitive apprenticeship learning methods.

Retrieved April, 17, 2009, from http://www.herdsa.org.au/wp-content/uploads/conference/2004/PDF/P069-jt.pdf

Table 7: Students’ perceptions of the online system as revealed in their reflective journals (n = 30)


A The system allowed me to view peers’ essays. This helpedme plan for my texts.

P02, P33, P51, P60

B It was convenient for me to observe the local and globalrevisions provided by peers in the system. These revisionswere shown clearly and separately in the system.

P09, P18, P42, P56, P67, P75, P93

C Through the system, I could know my peers’ comments onmy texts.

P15, P17, P22, P42, P56, P58

D The system not only helped my classmates and me improvewriting skills but also equip us with other skills such associal skills in communications.

P05, P15, P22, P29, P37, P47, P56,P60, P69, P86

E With the Diff engine, I could notice the differences betweentwo texts and saved a lot of time in comparing differentversions of a text.

P09, P24, P42, P58, P73, P93, P95

F With the trace result in the system, I would be able to reflecton my writing process and know how to improve it.

P03, P18, P35, P53, P67, P84

G The system aroused students’ awareness to be responsiblefor writing since they had to double check their revisions.

P02, P17, P33, P51, P63, P83



Braine, G. (1997). Beyond word processing: networked computers in ESL writing classes. Computers andComposition, 14, 1, 45–58.

Chee, Y. S. (1995). Cognitive apprenticeship and its application to the teaching of Smalltalk in a multimediainteractive learning environment. Instructional Science, 23, 133–161.

Cheng, Y. C. & Ku, H. Y. (2009). An investigation of the effects of reciprocal peer tutoring. Computers inHuman Behavior, 25, 1, 40–49.

Cho, K. & Schunn, C. D. (2007). Scaffolded writing and rewriting in the discipline: a web-based reciprocalpeer review system. Computers & Education, 48, 409–426.

Collins, A., Brown, J. S. & Newman, S. E. (1989). Cognitive apprenticeship: teaching the craft of reading,writing, and mathematics. In L. B. Resnick (Ed.), Knowing, learning, and instruction: essays in honor of RobertGlaser (pp. 453–494). Hillsdale, NJ: Lawrence Erlbaum Associates.

Duran, D. & Monereo, C. (2005). Styles and sequences of cooperative interaction in fixed and reciprocal peertutoring. Learning and Instruction, 15, 3, 179–199.

Ferris, D. R. (2003). Response to student writing. Mahwah, NJ: Lawrence Erlbaum Associates.Hansen, J. G. & Liu, J. (2005). Guiding principles for effective peer response. ELT Journal, 59, 1, 31–38.Jarvela, S. (1995). The cognitive apprenticeship model in a technologically rich learning environment:

interpreting the learning interaction. Learning and Instruction, 5, 231–259.Kay, R. H. (2007). Learning performance and computer software: an exploration of knowledge transfer.

Computers in Human Behavior, 23, 1, 333–352.Li, J. (2006). The mediation of technology in ESL writing and its implications for writing assessment.

Assessing Writing, 11, 5–21.Liou, H. C. & Peng, Z. Y. (2009). Training effects on computer-mediated peer review. System, 37, 514–

525.Liu, T. C. (2005). Web-based cognitive apprenticeship model for improving pre-service teachers’ perfor-

mances and attitudes towards instructional planning: design and field experiment. Educational Technology& Society, 8, 2, 136–149.

Lundstrom, K. & Baker, W. (2009). To give is better than to receive: the benefits of peer review to thereviewer’s own writing. Journal of Second Language Writing, 18, 30–43.

Min, H. T. (2006). The effects of trained peer review on EFL students’ revision types and writing quality.Journal of Second Language Writing, 15, 118–141.

Patel, A., Kinshuk, & Russell, D. (2003). Cognitive apprenticeship learning—ensuring far transfer of knowl-edge through computer-based assessment. In S. Naidu (Ed.), Learning & Teaching with Technology—Principles and Practices (pp. 233–246). London: Kogan Page.

Patton, M. Q. (1990). Qualitative Evaluation and Research Methods (2nd Ed.), Newbury Park, CA: Sage.Paulus, T. M. (1999). The effect of peer and teacher feedback on student writing. Journal of Second Language

Writing, 8, 3, 265–289.Storch, N. (2005). Collaborative writing: product, process, and students’ reflections. Journal of Second Lan-

guage Writing, 14, 153–173.van Leeuwen, R., Tiesinga, L. J., Jochemsen, H. & Post, D. (2009). Learning effects of thematic peer-review:

a qualitative analysis of reflective journals on spiritual care. Nurse Education Today, 29, 4, 413–422.Vass, E., Littleton, K., Miell, D. & Jones, A. (2008). The discourse of collaborative creative writing: peer

collaboration as a context for mutual inspiration. Thinking Skills and Creativity, 3, 192–202.Wang, F. K. & Bonk, C. J. (2001). A design framework for electronic cognitive apprenticeship. JALN, 5, 2,

131–151.Weber, R. P. (1990). Basic Content Analysis (2nd Ed.), Newbury Park, CA: Sage.Williams, S. M. (1992). Putting case-based instruction into context: examples from legal and medical

education. The Journal of the Learning Sciences, 2, 4, 367–427.Woolley, N. N. & Jarvis, Y. (2007). Situated cognition and cognitive apprenticeship: a model for teaching and

learning clinical skills in a technologically rich and authentic learning environment. Nurse EducationToday, 27, 73–79.

Xiao, Y. & Lucking, R. (2008). The impact of two types of peer assessment on students’ performance andsatisfaction within a Wiki environment. Internet and Higher Education, 11, 186–193.

Yoshimura, F. (2009). Effects of connecting reading and writing and a checklist to guide the reading processon EFL learners’ learning about English writing. Social and Behavioral Sciences, 1, 1871–1883.



Copyright of British Journal of Educational Technology is the property of Wiley-Blackwell and its content may

not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written

permission. However, users may print, download, or email articles for individual use.

Documents

A reciprocal peer review system to support college ... Student... · A reciprocal peer review system to support college students’ writing ... Lundstrom & Baker ... Three learning