國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
57
instructor, not a participant observer; therefore, her observation mostly relied on the video data. In other words, in Cycle I, the video recordings were the main source of data used to document the implementation results of Prototype I. Each lesson to implement Prototype I was recorded on video. To meet this data-collection need, an e-classroom was chosen, for it was equipped with two ceiling-mounted cameras, one at the front and one at the back of the room (See Picture 3.1), which would not be distracting to the students. This helped to eliminate the effects of the video cameras on the students’ learning behaviors.
A ceiling-mounted video camera
Picture 3.1 E-classroom in Cycle I
However, the class could not be recorded without participants’ permission.
According to Bogdan and Biklen’s (2007) ethical principle, “[a] researcher should neither lie to subjects nor record conversations on hidden mechanical devices” (p. 54).
Following this principle, the student participants in the study were informed in the first lesson that classes would be recorded on video for the entire semester, though only the recordings of the implementation of Prototype I entered into analysis.
Prototype I was conducted around the middle of the term, by what time the students
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
58
were used to being recorded and acted naturally in the classroom. To meet the research purposes, the researcher instructed the e-classroom assistants on when and how to take videos from different angles during each class session.
In-depth Interviews
In Cycle I, semi-structured interviews with students and teachers were conducted.
Table 3.3 summarizes all the interviews conducted in Cycle I. First, after the implementation of Prototype I, all groups were invited to participate in group interviews. The purpose of interviewing all the groups was to develop a broader understanding of students’ perspectives on the task. The interview schedules were arranged with the student participants’ agreement and willingness. In total, 34 students from the ten groups participated in the interviews. As there were many student interviewees in Cycle I, assignment of pseudonyms was deemed impractical.
Thus, a numeric label was given to each interview participant, such as “I-34.” The Roman numeral I represented Cycle I, and the following Arabic number 34 referred to the order in which the interviewee spoke. In other words, I-34 was the last student speaking in Cycle I; I-01, the first. Each interview for each group lasted around 30 minutes and was recorded with an audio recording device.
The second interview was conducted with the guest teacher to acquire an understanding of her perspectives on the task, class interactions, and students’
improvement in translation and review competence. The interview was scheduled immediately after the implementation of Prototype I and was conducted one week after the implementation. The interview lasted for 30 minutes and was also recorded.
Table 3.3 summarizes the interview participants and interview guides for all the interviewees. The ten groups of students were encouraged to talk about their feelings and attitudes toward Prototype I of the CoTT and their preferences for the five sessions. For example, the researcher asked the ten groups if they liked Session 4, a
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
59
session at which the two teachers gave suggestions to students. Finally, students were asked to reflect on their own learning of translation or other aspects in the interview.
The interview guide for interviewing the guest teacher aimed to develop an understanding of the user’s perspective. The guest teacher was encouraged to talk about her feelings about Prototype I of the CoTT and give her observations on the learners’ behaviors in class. Based on her responses, the researcher invited the guest teacher to suggest modifications of Prototype I.
Table 3.3 Interview Guides in Cycle I
Interview Participant Interview Guide Ten groups of students 1. Perspectives on Prototype I
2. Preference of the sessions of Prototype I 3. Learning in translation
The guest teacher 1. Perspective on Prototype I
2. Observation on students’ learning behaviors 3. Suggestions on the modification of
Prototype I
The language chose by all the interviewees was Chinese. In this study, the transcriptions of the interviews were translated into English for data analysis in Chapter 4 and 5.
Cycle II Participants
The participants in Cycle II were 25 technological university students who majored in Applied English and two translation teachers. There were 3 males and 22 females taking this course, named English Translation. They were divided into five groups. They were all seniors, but they did not have any translation learning
experience before. The class met two hours each week, and the Prototype II took three weeks, five hours in total. The genre for the translation assignment was the story,
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
60
which was decided by the instructor. However, each group could select a story they liked and translate it from English to Chinese. The average translation words were around 450-500 (see a sample of student-chosen story in Appendix C). A consent form (see Appendix B) written in both Chinese and English was given to the student
participants to ensure that all the student participants knew that they were taking part in a research study. The teacher participants, two female teachers who played the roles of the instructor and the guest teacher, respectively, took part in Cycle II.
Videotaping
In Cycle II, the classroom was not equipped with video cameras, so a video camera was mounted in the back of the classroom by the researcher for the
implementation of Prototype II. However, when group work began, the camera could not capture group interactions. Hence, the researcher took the camera in her hands and moved to each seminar table to film the within-group and teacher-student interaction.
Each session in Prototype II was videotaped. Following the same ethical principles as Cycle I, the permission to record the class on video had been obtained from the students before the camera was set up.
In-depth Interviews
In Cycle II, two interviews were conducted. First, weekly interviews were conducted with a group of students in order to monitor the students’ changes in perspectives on Prototype II. Prototype II consumed three weeks of class time, so three interviews with the same group were conducted. The weekly interviews were conducted immediately after class. Since each member of a group went to the five different student seminars, interviewing a single group could provide the researcher sufficient data to understand what happened in each student seminar. Therefore, the present study employed Patton and Patton (2001)’s purposeful random sampling to select one group from the five groups by drawing lots, and this group showed their
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
61
willingness to participate in the weekly interviews.
Following the naming system in Cycle I, these five students were given numeric labels from II-1 to II-5, with Roman numeral II representing Cycle II and the Arabic number referring to the order in which the interviewee spoke. In other words, II-5 was the last student who spoke in the interview. Each interview lasted for around 30 minutes and was recorded.
In addition to the interviews with the students, after the implementation of Prototype II, an interview with the two teachers, the abovementioned translation instructor and guest teacher, was conducted immediately in a faculty lounge in order to give the two teachers a chance to share their perspectives on the feasibility of Prototype II and observations on students’ reactions to the activities. They suggested some modifications on Prototype II. This interview lasted around 30 minutes and was recorded.
Table 3.4 summarizes the interview participants and the interview guides. The group of student participants was encouraged to talk about their feelings and attitudes toward Prototype II and their preferences for the five sessions. It was expected that changes in students’ perspectives on Prototype II could be explored in weekly interviews. Finally, the students were asked to reflect on their own learning of translation and other aspects of the class during the interview.
The interview guide helped the researcher to understand the users’ perspectives (see Table 3.4). The users were encouraged to talk about their feelings during the implementation of Prototype II and provide observations on the learners’ behaviors in class. Based on their responses, the researcher invited the two teachers to suggest modifications to Prototype II. As the guest teacher had participated in both cycles, she was especially encouraged to point out the strengths and the weaknesses of both prototypes. Moreover, the differences in the students’ learning behaviors in both
‧
cycles could be described especially clearly by the guest teacher.
Table 3.4 Interviews Conducted in Cycle II
A group of students 1. Perspectives on Prototype II 2. Preferences for the five sessions 3. Learning in translation
Translation instructor The guest teacher
1. Perspectives on Prototype II
2. Observations on students’ learning behaviors 3. Suggestions on the modification of
Prototype II
4. Comparison of Prototype I and II (guest teacher only)
5. Differences in students’ learning behaviors in Cycle I and II (guest teacher only)
The language chosen by all the interviewees was Chinese. In this study, the transcriptions of the interviews were translated into English for data analysis in Chapter 4 and 5.
Differences in the Design between Cycle I and II Participants
The student participants in the two cycles were from two different technological universities. Student participants in Cycle I had taken translation courses for at least one semester, while the student participants in Cycle II had never taken any
translation courses before. The biggest difference between the student participants in the two cycles was the class size. The class size in Cycle I was larger, 56 students, while the class in Cycle II was smaller, only 25 students.
The two female teachers were from different universities. The instructor had two semesters of teaching experience, and the guest teacher five semesters of experience, in teaching translation. In the study, they played different roles in class. The instructor played the roles of class manager and translation reviewer, while the guest teacher only took responsibility for reviewing translations. The instructor participated only in
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
63
Cycle II, while the guest teacher took part in both Cycle I and Cycle II.
Video Recording
The video recordings differed in two aspects. First, the roles of the videos in the two cycles were different. In Cycle I, the videos were the primary data because the researcher could not act as an observer. The researcher’s analysis of classroom interaction mainly relied on the videos. In Cycle II, the researcher herself was a research instrument for taking field notes. The researchers’ field notes were the primary data.
The second difference was the effects of the video cameras on the students. In Cycle I, the cameras were part of the classroom, so the students hardly noticed the cameras, and the impact could be reduced to a minimum. On the other hand, in Cycle II, the researcher had to carry the camera around the classroom, the lens sometimes only a meter away from the students. This proved to be a major distraction to some students. Student Participant II-3 complained in the first weekly interview that she felt really uncomfortable when the camera was focused on her. In one video segment, this participant can be seen to display anger; when the camera nears her table, and she glares at it. It seems that the camera had some negative effects on the students’
emotions. Although there were some complaints, most of the students attempted to ignore the existence of the camera when completing their tasks.
In-depth interviews
The interview questions for the guest teacher, who participated in both Cycle I and Cycle II, were similar, but the questions in Cycle II encouraged her to address the differences in the implementation results of Prototypes I and II. Her responses were especially important, as she had observed both implementations and could thus draw such comparisons.
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
64
Another difference was the number of interviews with students. In Cycle I, to deeply explore the students’ perspectives on Prototype I and gain input for making major modifications, ten groups of students were invited, and 34 students eventually participated. All together, there were ten interviews in Cycle I. On the other hand, to investigate gradual changes in students’ perspectives on Prototype II, weekly
interviews with the same group were conducted, three interviews in total.
Analysis of Data
The present study aimed to design a translation task through iterative implementations and modifications. To find problems in implementation, it is
necessary to have a framework to systematically analyze the results in Cycles I and II.
As Collins, Josephy and Bielaczyc (2004) suggested, a researcher needs to identify the goals and elements of the design, settings to implement, each phase and change, the outcomes found, and lessons learned when reporting on design research.
Engeström’s Activity theory, which includes five components, objects, rules, community, division of labor, and mediating artifacts, is well matched with the intended goals of design research. Moreover, it is suggested by Thorne (2004) that researchers use Activity theory “to define and analyze a given activity system, to diagnose possible problems, and to provide a framework for implementing
innovations.” (p. 65). Therefore, in order to diagnose implementation problems, and to provide guidelines for successful implementation of the CoTT, the current study utilized Activity theory as a tool to analyze the implementation results of the two prototypes. Figure 3.1 illustrates the six main components in an activity system and their connections.
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
65
Figure 3.2 Components of the Activity System
Regarding the issues of validity and reliability of a qualitative study, there are no standardized instruments to rely on, as in quantitative studies. Instead, validity is placed on a researcher’s analysis and data as accurately representing the social world (Neuman, 2000). Woods (2006) summarized three main features which qualitative research commonly rest upon for trustworthiness: unobtrusive and sustained methods, respondent validation, and triangulation. Unobtrusive and sustained methods refer to
“the methods that leave the situation undisturbed as far as possible” (p. 4). The researcher can focus on long-term participant or non-participant observation,
unstructured interviews or conversations, the use of key informants, and the study of documents. Respondent validation occurs when a researcher takes field notes back to the respondents for them to judge the adequacy (Neuman, 2000). If the respondent can recognize the researcher’s description as reflecting their intimate social world, the study is respondent or member valid. (Neuman, 2000). Finally, triangulation refers to
“the use of different researchers or methods, at different moments of time, in different places, among different people and so on” (Woods, 2006, p. 4), which strengthens the researcher’s account.
Subjects
Mediating Artifacts
Object
Rules Community Division of Labor (horizontal & vertical)
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
66
In the present study, the researcher adopted triangulation to meet the requirement of trustworthiness. Multiple data were collected in these two cycles to support or contradict the interpretation, including video recordings, audio recordings, interview data, and the student documents such as their translation works. It was expected that these triangulation methods would help the researcher to eliminate biases that might result from using only one data-collection method, source, analyst, or theory (Borg &
Gall, 2003), and in turn gives qualitative research validity (Woods, 2006).
Summary of Chapter 3
Chapter 3 addressed the design of the Cooperative Translation Task (CoTT) and the design of the present study. The student participants in both cycles were
technological university students, including 56 students in Cycle I and 25 in Cycle II.
Two translation teachers participated in the study. For data collection, triangulation data were collected, including videos, interviews, and student documents. The data was put into the framework of Activity theory to diagnose implementation problems in terms of community, mediating artifacts, and division of labor, and innovations with solutions were provided. Finally, the trustworthiness of this study was achieved by triangulation.
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
67
CHAPTER 4
IMPLEMENTATION OF PROTOTYPE I OF THE CoTT
This chapter provides answers to research question 1: What were the results of implementing the first prototype of the Cooperative Translation Task (CoTT)?
Prototype I was conducted in a technological university in central Taiwan. The participants included 56 students who were majoring in applied English and a translator teacher, who was the guest teacher for translation evaluation. There were five sessions in Prototype I. Table 4.1 presents the five sessions in three stages.
Session 1 was implemented before students’ presentations of their translation products.
At the presentation stage, Sessions 2, 3, and 4 were conducted after students’
presentations. Finally, Session 5 was conducted after class.
Table 4.1 Five Sessions of Prototype I of the CoTT in Three Stages
Stage Session
Pre-presentation Session 1: Written Peer Response
Presentation Session 2: Within-group Discussion & Translator Seminar Session 3: Oral Peer Response
Session 4: Oral Teacher Response Post-presentation Session 5: Final Revisions
Each session was analyzed based on Engeström’s (1987, 1993, 1999) Activity theory in (a) Community (b) Mediating Artifacts, and (c) Division of Labor (see Figure 4.1). After the analysis, the modification principles of each session for Cycle II were constructed.
In the end of the chapter, a summary of the findings and modifications to the five sections will be provided.
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
68
Figure 4.1 Activity System
Pre-presentation Stage
The Pre-presentation Stage was a stage to prepare the translation students for participating in the activities in the next stage. The only activity included was Session 1: Written Peer Response. Each session started with an introduction of its
implementation process within the Activity Theory framework, and then the results of the implementation were analyzed according to three components: Community, Mediating Artifacts, and Division of Labor. After the analysis, modification principles were provided for each session.
Session 1: Written Peer Response
Students (the Subjects) learned how to give and use written peer response (Object), two randomly selected groups exchanged translations for peer editing (Rules). Each group gave comments and suggested translations on a peer response sheet (Mediating Artifacts) and sent it back to the translator group by email
(Mediating Artifacts). The translator group made decisions on whether to accept the suggestions and made revisions (Division of Labor). Figure 4.2 illustrates Session 1 within the framework of the Activity system and the interactive relationships among components.
Subjects
(b) Mediating Artifacts
Object
Rules (a) Community (c) Division of Labor (horizontal & vertical)
‧
Figure 4.2 Session 1 in Activity System
Community: Individual Peers, Instructor, and Guest Teacher
In Activity Theory, community refers to the participants who share the same object that shapes and lends direction to individual and shared activity (Engeström, 1987, 1993, 1999). It mediates all elements shown in Figure 4.1. In a community, all the members share labor by utilizing the mediating artifacts to meet the object under the activity rules. It is crucial to consider who should or should not be involved in the community. Therefore, the influences from each member in a community should be analyzed. Here the analysis focused on if the community members helped the subjects give and use written peer response.
The community in Session 1 included the individual members of a peer group and the two teachers. In this community, peer group played the role of translation editor, who helped the translator group to revise a translation. The two teachers here were not actively involved in written peer response activity, but played the role of
The community in Session 1 included the individual members of a peer group and the two teachers. In this community, peer group played the role of translation editor, who helped the translator group to revise a translation. The two teachers here were not actively involved in written peer response activity, but played the role of