«Item type text; Electronic Dissertation Authors Goertler, Senta Publisher The University of Arizona. Rights Copyright © is held by the author. ...»
The research results will be summarized in chapter five in relation to the following topics: role of the teacher in synchronous computer-mediated communication (SCMC), corrective feedback in SCMC, language use by teachers and students in SCMC, students’ preferences and actual practices, implications for teaching, implications for program administration, implications for teacher training, connections to SLA research, connections to CMC research, directions for further research, and limitations of the study.
5.1. Introduction After presenting the research results in response to the questions in detail in chapter four, in this chapter the results will be summarized according to the following themes: role of the teacher in SCMC, corrective feedback in SCMC, language use in SCMC, students’ preferences and actual practices, implications for teaching, implications for program administration, implications for teacher training, connections to SLA research, connections to CMC research, limitations and suggestions for future research.
Furthermore, the previous research discussed in chapter two, as well as some additional research, will be taken into consideration in this concluding chapter. However, first a summary of the study will be provided.
As described in chapter three, this study is a multiple case study using a mixed design analyzing a multi-faceted data set. Participants in this study are 46 students and two teachers from three classes. The students and teachers were originally intended to be involved in ten to fifteen 20-minute chat sessions. However, due to curriculum constraints and computer problems that led to transcript loss, students were engaged in 10 or fewer chat sessions, with some of them lasting more and some less than 20 minutes.
Overall students chatted for a total of two to three hours on average during the semester.
The two students who chatted the most in each class were selected as case study subjects for a more in-depth analysis of their chat transcripts. The classes were labeled according to the level of technological support they received during the lab session: No-Support Class (NSC), Some-Support Class (SSC), and Expert-Support Class (ESC). NSC and SSC were held in the morning and were taught by the same teacher, who was assigned the pseudonym MorningTeacher. In similar fashion the teacher teaching ESC in the evening was named EveningTeacher. The students had self-selected screen names for the chatting, which were exchanged with pseudonyms for the purposes of reporting the data.
The chatting occurred in the collaborative computer lab of the College of Humanities at the University of Arizona. The chat activities were designed in adaptation from the textbook and covered all relevant structures and vocabulary. The data sets included the transcripts, self-report forms, pre- and post tests, pre- and post-surveys, classroom observations, and informal conversations with the instructors.
As has been mentioned already in chapter four, this study added new findings to the research, even though it faces several limitations. In the following sections the findings, the connection to the field, the implications, the limitations, and the directions for future research will be discussed.
5.2. Role of the Teacher in SCMC As has been discussed in chapter four, the two different teachers investigated in this study appeared to play some of the same roles and some differentiated roles in the SCMC environment. In this section the research results of the different roles will be reviewed.
The task that I gave the two teachers was to implement the chat activities in their teaching. While the Instructor’s Manual instructed them to introduce the topic and any unknown words, MorningTeacher did not introduce the activities during the classes observed, while EveningTeacher did.
As has been mentioned in chapter three, the teacher’s role was intended to be that of instructor or facilitator, but not that of technological staff. However, in the absence of a Lab Assistant in the NSC, and with a Lab Assistant with limited knowledge of the program in the SSC, MorningTeacher experienced more technological problems than EveningTeacher. While EveningTeacher utilized the Lab Assistant for group assignment, procedural instructions, and computer assistance, the MorningTeacher almost exclusively controlled those aspects in NSC and SSC.
Furthermore, the two teachers utilized the physical space differently. The MorningTeacher walked around and assisted students during the chat, while EveningTeacher remained seated at the teacher station and commented out loud. Both teachers used the projector screen to show the activities. The physical space was used for the following purposes: group assignment, task introduction, assistance in case of questions or computer problems, common experience, to see and hear each other, to comment, to play music, to laugh, to read each other’s conversations, and to utilize resources.
On a related note, it should also be reiterated that students’ descriptions of their teacher’s role and behavior was a good match with that found in the transcripts and in my classroom observations. However, students hardly ever commented on the teacher’s function and role in the physical space during the chat, while the classroom observations revealed some differentiation in the teachers’ approaches to using the physical space.
In regards to the amount of language the students were exposed to from their teacher, there was a significant difference between groups for teacher words per minute according to an ANOVA, with the NSC experiencing the least (0.45 word per minute) and ESC the most (2.12 words per minute) teacher input in words per minute. Hence, it can be argued that the ESC experienced more active participation from their teacher than the NSC and SSC students. If we approach an interpretation of these data from a perspective that emphasizes the importance of input, such as Krashen (1985), then the conditions in the ESC were more conducive to learning. On the other hand other models of input processing acknowledge that the input as such is not the only factor for language acquisition. Gass (1997), as was discussed in chapter two, defines several stages through which the input has to be processed in order to result in language learning. Furthermore, VanPatten (1995) suggested, based on his research, that input without attention to form may not lead to intake, at least not as frequently as input with attention to form.
Considering the low frequency of corrective feedback by students and teachers observed in this study, the effects of the teacher input may be limited. On the other hand, all activities were based on the grammar topics in discussion, which may suggest the presence of a focus on form. Furthermore, during the chatting students not only received the teacher’s input, but were also able to interact with her and negotiate meaning, which is considered beneficial for language learning (see for example Long, 1996). In terms of output opportunities, which Swain (1985) considers essential for language learning, there were no differences between the classes (see also Swain & Lapkin, 1995; Izumi, Bigelow, Fujimara, & Fearnow, 1999). The discussions in the literature concerning output may weaken an argument that the conditions in the ESC where there was indeed more teacher input were more conducive to language learning than the classes where there was less teacher input, namely SSC and NSC.
Additionally, according to a chi-square analysis of the corrective feedback moves in the case study subjects’ transcripts, MorningTeacher used significantly more feedback (3.18% of errors were corrected in contrast to 0.75% by EveningTeacher) than EveningTeacher. Both feedback rates are lower than those reported from face-to-face interactions (for example Lyster & Ranta, 1997). The study by Ene et al. (2005), the only other CMC study on teacher feedback, did not measure errors, so no feedback rate was investigated in that study. Furthermore, MorningTeacher had a clearly defined and high frequency form of feedback (repetitions with correction, which in the research are often referred to as recasts, and models), while EveningTeacher used varied forms of feedback, each less clearly defined and used with much less frequency. MorningTeacher’s preferred feedback style matches that reported in other studies (see for example Lyster & Ranta, 1997).
Both teachers used some of the same and some differing teacher moves. Both teachers used corrective feedback moves, moves that modeled the language and the activity, moves that focused on conversing with the students, and moves that provided TL vocabulary items to the students. In addition, MorningTeacher used teacher turns to control language, on-off-task behavior, topic focus, and to praise students.
EveningTeacher did not use those moves, but instead she used the following additional teacher moves: procedural help, personal sharing, and expanding the topic.
MorningTeacher used her turns to control the task, while EveningTeacher used her turns to widen the task. MorningTeacher did not introduce new topics, while EveningTeacher did. Beauvois (1998) argues that CMC allows more equality in terms of topic introduction between students and teachers, which was only observed in NSC and SSC here.
Based on these observations, MorningTeacher can be described as having the role of teacher and facilitator both in the physical and in the virtual space. EveningTeacher can be described as having the role of conversation partner both in the physical and virtual space, except for her short introductions in the beginning and the occasional corrections. The students described MorningTeacher’s role as that of a source of correction and facilitation in a discrete form. EveningTeacher was described by her students as participating and stimulating the conversation. Students also stated preferring an active teacher, which is in contrast to Donaldson and Kötter’s (1999) findings who found that the teacher was seen as an intruder.
Another interesting issue that arose out of the data analysis was the topic of teacher power. This is especially interesting given the ‘democratization effect’ often attributed to CMC (for example, Beauvois, 1998). In fact, despite some possible democratization with CMC, the teacher is still in an authority role, for he or she decides the grades of the students. By definition, any teaching setting is automatically comprised
of roles for teacher and student (Lantolf & Genung, 2002). Auerbach (2000) states:
All classrooms are “teacher centered” to the extent that it is the teacher’s conception of education that shapes how the learning community develops.
Clearly teachers have their own goals, their own understanding of effective L2 pedagogy, and most importantly, they have power. To deny it is both irresponsible and disingenuous: students know it and teachers act on it whether or not they acknowledge it (pp.144-145).
Indeed, the role of the teacher is inseparable from that of the role of power in a classroom setting, no matter what the specific activity is. Lantolf and Genung (2002) define power as: “the capacity (and privilege) to project and impose one’s perspective on others without taking account of others’ perspectives.” (p. 178). For the purposes of this study, that power is manifested in each teacher’s choices about how and when to interject in the virtual space.
In addition to the fact that the two teachers inherently had a power role in the classroom, there were also specific components of their behavior that exhibited these power relationships. EveningTeacher and MorningTeacher both used features of the chat server that were only available to the instructor: EveningTeacher the “to all” function and MorningTeacher the “invisible” function. This is an obvious manifestation of the teacher power assumed to be inherent to the role, as this instructional software was designed to possess certain capabilities for use only by the teacher. Not only were teachers the only ones allowed to address everyone or be invisible in a room, only the teacher could easily switch from room to room and therefore be omnipresent in all chat rooms. These extra functions for the teacher clearly exhibit an imbalance of power, since the students do not have the capability of using these functions.
Additionally, the teachers also implicitly or explicitly set the rules of the interaction. EveningTeacher modeled acceptable behavior, while MorningTeacher used her teacher turns explicitly to monitor the interactions of the students. For example, and as briefly mentioned above, MorningTeacher used teacher moves to narrow the task and often used the “invisible” function, meaning that she could read what the students were typing without their knowing that she was doing so. These two implantation choices (narrowing task and “invisible” function) can be seen as elements of a teacher role which take a stance of power in interaction with the students. With her teacher moves as evidenced in the chat transcripts, MorningTeacher monitored the conversation of the students, including more feedback for her students than EveningTeacher did for hers.
Furthermore, as just stated, the “invisible” function can only be used by the teacher, and allows the teacher to secretly observe the students’ conversation. EveningTeacher, on the other hand, used the “to all” function, which also can be seen as a move expressing the teacher’s hierarchical power, for it interrupts the flow of the conversation and can only be used by the teacher. Also, EveningTeacher exposed students to more of her own words per minute. Even though they did not have a significant impact on students’ words per minute, EveningTeacher’s frequent contributions to the chat written conversations may still suggest a more involved role than MorningTeacher. Since she expanded topics, she introduced new topics more often than MorningTeacher.
MorningTeacher and EveningTeacher both used power functions but used them differently. It can be argued that either of them or both of them are dominating styles. It may be just as dominating to know that MorningTeacher could be reading your chat and not knowing for sure when she’s doing so as it is to experience more topic introduction and a more active teacher.