«Item type text; Electronic Dissertation Authors Goertler, Senta Publisher The University of Arizona. Rights Copyright © is held by the author. ...»
It has also been a concern that perhaps error uptake would be more dramatic in the CMC environment that in F2F interactions. However, Ene et al. found little evidence of error uptake in comparison to the number of errors produced, and determined that a percentage of the error uptake by the interlocutor may have been a reflection of a similar interlanguage stage.
Kelm (1992) voiced caution in regard to the transferability of skills from a written CMC communication to oral communication. However, Kost’s (2005) quasiexperimental dissertation research on overall language development in CMC versus face to face (F2F) role-plays found no significant difference in oral and written proficiencies.
Encouragingly, Payne and Whitney (2002) found CMC beneficial for oral proficiency.
Although more research is needed, these results suggest that CMC is not harmful for language development, even oral language development, and may prove beneficial in language learning overall.
2.4.3 CMC vs. Classroom As mentioned, early studies of CMC claimed to compare the benefits of CMC in contrast to classroom settings, even though such comparisons were not always justified.
Yet, many of the results correlated with each other, and in this sense they validate the research results of the individual studies.
The following benefits of CMC over classroom discussion were suggested by early CMC research: CMC (1) provides a voice to those who do not have one (Batson, 1988); (2) increases language productivity (Beauvois, 1992, 1998; Kern 1995); (3) decreases teacher dominance (Beauvois, 1998; Kern, 1995 though he views that negatively); (4) equalizes participation especially for shyer and minority students (Beauvois, 1992, 1998; Bump, 1990; Kelm, 1992; Kern, 1995; Warschauer, 1996); (5) increases willingness to discuss topics openly and honestly (Beauvois, 1992, 1998); (6) increases student motivation (Batson, 1988; Beauvois, 1992; Bump, 1990; Donaldson & Kötter, 1999; Kelm, 1992); (7) decreases anxiety (Beauvois, 1992, 1998; Donaldson & Kötter, 1999); (8) encourages target language use (Beauvois, 1992, 1997, 1998; Chun, 1994; Kelm, 1992); (9) improves reading comprehension (Kelm, 1992); (10) improves writing ability (Chun, 1994); (11) increases language output (Darhower, 2000; Sullivan & Pratt (1996); (12) increases interactive discourse (Abrams, 2001; Chun, 1994); (13) improves attitude towards language learning in general (Blake, 2000; Kern, 1995;
Warschauer, 1996); (14) improves attitude towards cultural studies (Fraser, 1999;
Schneider & von der Emde, 2000); (15) provides a great range of language functions through discourse (Abrams, 2001; Chun, 1994); and in general (16) can benefit language development (Beauvois, 1992; Chun, 1994; Kelm, 1992; Kern, 1995; Warchauer, 1996).
The features of CMC described earlier (reduced speed, editing capabilities, reading options, etc.) are possible explanations for such benefits.
Beauvois (1997) found that a group of learners utilizing CMC outperformed a group utilizing F2F in an oral exam on pronunciation, grammatical accuracy, lexical choice, accuracy, and content. In the same study, Beauvois also argued that the syntactic complexity, the lexical sophistication, and the amount of speech were good during CMC for the language level of the students. Warschauer (1996) made a similar argument.
The challenge of comparing F2F classroom to CMC interactions, is that usually only small group or pair work is utilized in the CMC portion of the class, while these as well as other interaction forms are used in the classroom. Hence, in order to generate reliable findings, one would have to either compare pair and small group work in the F2F portion of the class with that in CMC, or find ways to also include teacher-fronted, whole group, and individual work in CMC. Moreover, computers and the internet offer learners the opportunity to utilize various kinds of resources, which again changes the way the learners are able to and will interact. It should be noted that computer and internet access for all students may be of concern, as well as that the teacher’s or institutional knowledge of technology and its options influences this availability.
2.4.4 CMC as Part of Class vs. Independent of a Language Class Most research studies to date investigated the use of CMC in an intact language classroom, even though, in many cases, the CMC component of the class was an add-on for research purposes and not an integral, normal portion of the course. Many research studies do not mention if the CMC component of the class was (a) graded and part of the course grade and/or (b) seen by the course instructor. One can assume that students participated in varying degrees of effort depending on both the significance of the task for their course grade and their relationship with the course instructor.
The approach of CMC to language learning, whether presented as an independent component or as part of a course curriculum, naturally changes the requirements of the task design and the nature of the interaction. To this researcher’s knowledge, the only study investigating CMC that was not part of a curriculum was a study by Klocke and Görtler (presented at CALICO, 2003). In this study, the researchers investigated the use of a commercial ESL chat server for further independent study, and found that successful use of such opportunities requires persistent, almost aggressive behavior by the ESL learner, as well as the willingness to take risks. Understandably then, one of the major indicators whether or not the independent engagement with CMC was successful was the personality of the learner.
2.4.5 CMC Asynchronous vs. Synchronous Like the manner in which CMC is presented in the language learning environment, the context of synchronous versus asynchronous CMC is a factor that changes the activity. While some instructors use asynchronous CMC software, such as message boards, for synchronous communication, it remains an asynchronous tool.
Although there might be simultaneous postings, they are not seen as the message is written. Often these programs require the user to hit the “refresh” button or tab in order to update the screen communication rather than automatically updating. Also, the physical presentation of the program usually encourages longer responses than the layout of a small chat window which often only has space for one line. Abrams (2003) summarizes the similarities and differences between SCMC and ACMC, all of which could arguably
influence the amount and the complexity of the language produced, as follows:
Similarities: Extensive learner-to-learner (or learner-learner-teacher) negotiation of meaning; more ‘talk’ time per learner than oral classroom communication;
increased amount of output results in richer and more diverse lexicon; written code; register between those of written and oral styles of communication.
Differences: Relatively immediate responses in SCMC vs. extended planning, encoding, decoding time in ACMC; Use of outside resources cumbersome in SCMC versus use of outside resources not limited in ACMC; Social immediacy of interlocutors in SCMC versus interactants not ‘immediately’ present (p. 159).
Some of the differences in findings about the effects of using CMC can be explained by the use of different research tools. For example, Chun (1994), when looking at chat, noted a lack of language complexity, while Gonzales-Bueno (1998), on the other hand, found an increase in complexity when using email for journaling rather than paper and pencil. This difference can be explained by the use of synchronous CMC by Chun versus asynchronous CMC by Gonzales-Bueno. Based on the summary of similarities and differences by Abrams, this discrepancy can be expected.
Abrams (2003) contrasted different media of CMC in respect to some of the benefits identified by numerous CMC researchers as previously discussed. She found that the asynchronous CMC treatment group produced less output than the synchronous group or the F2F control group in regards to lexical richness and density and syntactic complexity. This finding is surprising, considering that a message board, standard to asynchronous CMC, is visually more inviting for larger postings. However, the tasks observed were designed as discussions, and a synchronous form of communication is more natural for discussions than an asynchronous form, which may explain why Abrams (2003) came up with seemingly counterintuitive results.
Perez (2003) compared asynchronous and synchronous CMC for their effect on language productivity. In her within-subject design study, she measured language productivity through new words used and found no significant difference between the two media. The preference of one medium over the other was also similar between the two groups. However, it should be noted that students produced more words in chat than in email messages. And though the activities are poorly explained, it can be assumed that this difference can be attributed to the instructions given to the students. Specifically, email messages were to be at least 90 words long, whereas chatting took place over the course of an hour. It is easily arguable that even a beginning student can write more than 90 words in an hour.
2.4.6 CMC from the Same Location/Time vs. CMC from Different Locations/Times Some research studies have investigated CMC activities occurring in the same physical space, whereas others have reported on CMC in which the participants were not in the same physical space. When using CMC in different locations, students may be required to participate at the same time, or may be allowed to participate at their convenience with time differences. If the chat occurs outside of class time, students may be less likely to participate because it requires extra effort. In addition, if chat occurs at different times, the teacher’s role changes due to the diminished likelihood that any or all groups will be observed, especially when using synchronous CMC. It must also be taken into consideration that computer quality may affect the interaction, as some students have slower, dial-up connections while others are using high-speed internet connections. The sense of community will change depending on where the CMC happens; and, if the chatting occurs outside of the class, the students are no longer able to ask each other and the teacher questions during the chatting in the physical space, affecting the CMC dynamics.
Donaldson and Kötter (1999) reported that technical difficulties that some students experienced, due to the CMC component occurring at different times and locations, minimized the benefit for some students. They also reported that the teacher was viewed as an intruder in a synchronous CMC activity that was to occur outside of class.
2.4.7. CMC Non-Native Speaker & Native Speaker Dyads vs. Non-Native Speaker Dyads Also challenging in comparing CMC studies is the attempt to compare results from different dyads and small groups. Some CMC projects investigated the interaction between NNS and NS, while others attended to the interaction between NSSs. As mentioned earlier in the discussion of the negotiation of meaning, a dyad consisting of NNSs is likely to negotiate more than a NNS and NS pair. On the other hand, a NNS interacting with a NS has more opportunity to receive correct and premodified input.
Additionally, many of the NS/NNS pairs are set-up as a tandem learning project which also changes the dynamic. More code-switching appears to occur in such contexts, if, for example, one compares the results of code-switching in Donaldson and Kötter (1999) with Ene et al. (2005).
2.4.8 CMC Group-Size It has also been found that group-size has an effect on CMC in the classroom.
Böhlke (2003) revisited the issues of participation and language quality in chat versus F2F discussions. He focused on differences according to group-size and language stage.
His cross-over design study measured participation in c-units (a communication unit: “a word, phrase, or sentence that in some way contributed pragmatic or semantic meaning to
a conversation” (Crookes, 1990, p.184), and adopted the five language stages (1:
Canonical word order: subject-verb-complement; 2: Adverb preposing: Adverbial phrasesubject-verb-complement; 3: Verb separation: subject-finite verb-complement-nonfinite verb; 4: Inversion: adverbial phrase-verb-subject-complement; 5: verb end (final verb) in subordinated clause) developed by Tschirner (1996). The five-member groups showed no equalizing effect, while the four-member groups did, suggesting that smaller group sizes are necessary to benefit from CMC. However, these findings could also suggest that an even number of students per group works better than an odd number of students. Based on anecdotal evidence, it appears that small group sizes such as two or three are effective, since students are forced to talk to each other; no student can stay silent without the other students noticing and most likely commenting on the fact.
2.4.9 CMC and Task Type Smith (2001) in his dissertation explored task-based CMC in an ESL context. He concluded that task type and characteristics are important for interactive discourse, negotiation of meaning, and consequential language acquisition. On one hand, he found that more closed tasks produced more incidental negotiation, which is similar to findings from Pica et al. (1993) and also a study by McBride (conversation at CALICO, 2005). On the other hand, Smith found that guided tasks that were more open allowed for more negotiation in general, especially in the way of new lexical items, the direction of the task, and language structures. This difference in research findings may be a result of the open tasks being guided and not open discussions. It appears that students will negotiate in any case, but the more guided they are, the more they have the ability to negotiate as they all know the direction of the task. However, too many restrictions may inhibit the variety of items that can be negotiated. The problem with this research is that the definitions of task types are still problematic, and that outcomes likely depend on the implementation and the introductions provided by the teacher.