«Item type text; Electronic Dissertation Authors Goertler, Senta Publisher The University of Arizona. Rights Copyright © is held by the author. ...»
2.4.10 CMC and Measurements of Language Quality As mentioned previously, CMC has been found both beneficial and detrimental for language development. The pessimistic view is that CMC leads to lower levels of accuracy (Beauvois, 1992; Kern, 1995), more error uptake, lesser language complexity (Böhlke, 2003; Chun, 1994), and the use of medium-specific non-standard language, such as the use of the medium-specific acronyms like “LOL” (laugh out loud) and “BRB” (be right back). I have observed German native-speakers using the same acronyms based on the English language when chatting with each other. The optimistic view is that CMC is beneficial for language development (Beauvois, 1992; Chun, 1994; Kelm, 1992; Kern, 1995; Warschauer, 1996), and that the negotiation of meaning promotes language learning. Several studies have shown that it improves reading comprehension (Kelm, 1992), improves writing ability (Chun, 1994); increases language output (Darhower, 2000; Sullivan & Pratt, 1996); increases interactive discourse (Abrams, 2001; Chun, 1994); exhibits a great range of language functions (Abrams, 2001; Chun, 1994; Smith, 2001), improves or maintains oral proficiency (Kost, 2005; Payne & Whitney, 2002);
provides lexical choices (Beauvois, 1997); allows for syntactic complexity (Beauvois, 1997); provides lexical sophistication (Beauvois, 1997); increases language quantity (Abrams, 2003; Beauvois, 1997); and improves accuracy (Böhlke, 2003; GonzalesBueno, 1998).
The research findings about the effect of CMC on language learning have been mixed across measurements, studies, and sometimes even within a study. The underlying questions are what language improvement means and how it is measured. This challenge has been explored previously in regard to effectiveness of corrective feedback. The measurements that have been applied include improvement on tests, discourse analysis, word counts, number of errors, and analysis of lexical density and syntactic complexity.
While all of these measures are important and can provide insight into the effectiveness of teaching tools, they only paint a partial picture, especially in the case of CMC, where the medium itself may encourage or discourage certain behaviors. For example, in a fast moving chat conversation, it will most likely seem unnatural to use complex sentence structures that demand lengthier completion times. Also, due to problems with typing and the lack of chat spell check, there might be more errors in a CMC delivery, which may or may not indicate actual knowledge of the language.
2.4.11 CMC and the Role of the Teacher The role of the teacher has rarely been considered in a CMC context. In fact, many studies often do not mention what the teacher was doing during the chat sessions.
In a rare mention, Donaldson and Kötter (1999) discussed the role and influence of the teacher in their analysis of a tandem MOO project, in which they found that the teacher was seen as an intruder (Donaldson & Kötter, 1999 & Kötter, 2003). In this case, the students were engaged in MOO exchanges with tandem partners from a partner class (German – English) outside of class time, which appears to have been a required component of the class. Since this was a tandem CMC project occurring outside of class, it seems logical that the teacher was seen as an intruder, as there is no role for him or her in such an activity design. Furthermore, since CMC provides a learner-centered learning environment, the role of the student becomes more active and potentially decreases the need for a teacher. In addition, Beauvois (1998) noted that students as well as teachers introduced new topics in the chat environment, which was confirmed by Gonzalez-Bueno (1998) in an email context. This can be considered as further evidence that the teacher is not as necessary in CMC as in a F2F classroom context. In general, participation is claimed to be more equal among students in the CMC environment, but also between teachers and students. However, as stated previously, poor research descriptions have presented unclear teacher roles or tasks during the CMC activities. In some studies, it seems that the teachers did not participate at all (e.g., Abrams, 2003).
Ene et al. initially intended to explore that the effect the teacher’s absence versus presence has on focus on task, target language use, corrective feedback, and error uptake.
What they found was that two teachers who, in the regular classroom, have similar teaching styles engaged with their students in different ways during CMC. One teacher provided mostly explicit feedback, while the other provided mostly implicit feedback.
Whether the teacher was present or not made almost no difference in general. However, the teacher who used mostly explicit feedback appeared to have a silencing effect on the students compared to a teacher using implicit feedback. On-task behavior and target language use were high with both teachers in both conditions. Hence, even though students produced less words in the presence of the teacher using explicit feedback, they still were on task and using the target language most of the time.
2.4.12 CMC and Error Correction Research on corrective feedback during CMC has been limited. Sotillo (2005) reported that corrective feedback was available to learners participating in both NNSNNS and NS-NNS dyads. Interestingly, she found that NNSs provided more explicit feedback to their interlocutors than the NSs. The errors responded to were grammatical and lexical. She also found instances of successful uptake in both dyad forms.
In addition to the silencing effect of an explicit feedback style, Ene et al. also found that explicit feedback led to more learner uptake. However, they suggested that due to the medium, the learners who received implicit feedback may not have felt the need to show evidence of their noticing and understanding of the feedback they received, whereas the explicit feedback was more inviting. In a survey, students receiving implicit feedback reported that they noticed the implicit feedback, but did not feel that they should respond to it. Error uptake was low considering the number of errors made in both classes under both conditions. Furthermore, self-correction and peer-correction were also low, suggesting that students relied on the teacher for corrective feedback. In the survey, the students generally reported appreciation of the corrective feedback received, although some negative comments were received from students taught by the instructor using explicit feedback. The difference in corrective feedback style can be attributed to the teacher’s different interpretation of the medium. The teacher utilizing implicit feedback viewed the chat activities as spontaneous interaction which should not be corrected on the spot in an explicit form, but, if at all, rather with the help of the transcripts. The teacher utilizing explicit feedback viewed the medium as a written form of communication entailing the risk of error uptake, and hence considered it important to provide immediate explicit corrective feedback.
2.5 Gaps in the Literature CMC research has increased over the last two decades and has addressed many diverse issues. However, the field is still working on finding appropriate research methodologies that allow for a fair investigation of the medium as well as recommendations for implementation of the tools in a language classroom. Most studies do not consider the events in the physical space in their investigation of CMC, which leaves a great deal of important information unconsidered. Furthermore, many studies do not provide details about how the CMC activities were implemented, which further challenges the possibility of comparing research results between studies. This lack of information makes it impossible to provide concrete recommendations for the implementation of CMC activities in language programs for teachers, administrators, and students.
To date to this researcher’s knowledge, the only studies that have investigated the role of the teacher and corrective feedback in CMC have been this dissertation and the study discussed earlier by Ene et al. More research on the way in which teachers are participating in CMC activities, both in the virtual and in the physical space, is needed to provide specific recommendations for implementation, and more research on the use and effectiveness of corrective feedback during CMC must be compiled.
2.6 Methodological Issues As has been discussed throughout this chapter, the investigations into CMC and into the effectiveness of corrective feedback faces several methodological challenges relating to measurements, terminology, instruments, and procedures. The field has not yet agreed on effective measures of gains due to either CMC or corrective feedback. In the case of CMC, improvement on language tests, lexical density, syntactical complexity, amount of language produced, instances of negation of meaning, and error counts have been used to measure effectiveness. In the case of corrective feedback, instances of uptake, improvement on language tests, analysis of learner language in regards to specific elements and their stages on the interlanguage, and noticing (i.e., stimulated recall, acknowledgement of receiving feedback in transcripts) have been used as measures of effectiveness. New research needs to use multiple measures of effectiveness to contribute to be able to triangulate the data and contribute to the field.
An additional problem in research on corrective feedback and CMC is the issue of terminology. Researchers have used varying definitions of terminology which lead them to differing research results. For example, whether one defines uptake as an instance of correcting oneself immediately following the correction or within the same conversation will have an effect on the number of instances of uptake a researcher includes in his or her findings. Furthermore, such definitions may require differences depending on the context and the medium. For example, since chatting often has disjointed negotiation routines (Smith, 2001), uptake may be harder to identify.
Instruments and procedures have also been of concern. Since, as discussed above, the context can have an impact on the research results, it is important to describe the context and the procedures in detail. Language learning is a complex phenomenon, and the more we know about the factors involved in subjects’ language learning processes, the better we can assess what triggered an improvement or lack thereof. Additionally, general proficiency tests are often ineffective to measure improvement on language skills in single semester studies, which account for the majority of studies reported. More specific instruments must be designed in order to understand the physical and the virtual interactions and their effects on the analysis of data.
2.7 Summary In conclusion, more research is needed, and more precise and multi-faceted research studies must be designed. Currently, research suggests that input, output, interaction, and the negotiation of meaning are beneficial for language learning.
Corrective feedback, when noticed and engaging for the learner, benefits the language learning process. However, the research community is still conflicted as to the type of feedback that is effective and in which instances, as there may be a relationship between error type and feedback type, or student and feedback type. Research methodologies used make it difficult to evaluate the effectiveness of corrective feedback unless used in combination. In regards to CMC, many benefits have been found, although the issue of language quality is still viewed skeptically. Furthermore, most research has not provided enough information about the implementation of the CMC activities to fully understand the beneficial and the detrimental aspects of using CMC in the classroom, and it is clear that there are medium-specific issues. Thus, it is important to revisit issues that may or may not have been researched or analyzed sufficiently in F2F conversations in the CMC context. More specific recommendations for teaching must be identified and provided in terms of the role of the teacher, corrective feedback, and the use of CMC from a remote location.
As will be described in the next chapter, this study investigates the role of the teacher in both the physical and the virtual environment, the corrective feedback used, and its relations to error type and language improvement. These issues are important to the field, since they have not been sufficiently explored in the CMC context, and are still considered controversial in the F2F context. A multi-faceted research design was chosen to allow for multiple measurements of effectiveness. In addition, since this study aims to allow for comparison of research findings, and since this study strives to provide recommendations for implementations, the procedures and the events in both physical and virtual environment will be described from multiple perspectives using multiple instruments.
3.1 Introduction This dissertation is a descriptive multiple-case study utilizing a mixed design. It investigates the role of the teacher and the use of corrective feedback, and their relationship with these have on classroom chatting and language learning. The classroom chatting considered occurred in three third-semester German classes taught by two different instructors. The data sets stem from a variety of instruments, and are analyzed both quantitatively and qualitatively.