«Item type text; Electronic Dissertation Authors Goertler, Senta Publisher The University of Arizona. Rights Copyright © is held by the author. ...»
While these are all findings of behaviors and events in the physical space, some words of caution need to be expressed. First of all, having an observer present may have changed some of the interactions, and from my classroom observation notes, I can only comment on the events occurring during the times I observed the class. Second, due to the different labeling and formatting of data sources, in almost all cases it was impossible to match subjects from one data set to the next. However, since this was a discussion of events in general, such matching was not necessary. Furthermore, as has been discussed already with the surveys, students did not comment on many occurrences in the physical space. In fact only one student out of 46 mentioned the teacher’s actions in the physical environment on the survey. It may be that students also reported fewer occurrences in the self-report forms than actually were noteworthy. Therefore, it has to be pointed out that the students may not be the most reliable data source. However, in the absence of multiple video cameras, student report forms and observation notes are the best source of data regarding the events in the physical space.
4.6.2 Research Question 5b The question is: What modifications would have to be made when moving SCMC activities to a remote location? As mentioned in response to question 4a, understanding chat implementation in distance learning contexts is of interests to administrators. The answer to question 5a provides the basis for the recommendations for modifications if chatting were to be moved from an in-class activity to a distance learning context.
The areas of modification fall into the following categories: logistical issues, technological aspects, resources, bonding, and commonality. While with no certainty can these factors be claimed as essential for chatting, they did occur in the physical environment during the chat activities of these three classes in this study. I can only provide suggestions for modification, which should then be tested in a later study using actual distance learning environments.
Since procedurals and task explanations were mostly given in the physical environment, it may be necessary to provide a glossary for the task descriptions.
Furthermore, the chat server may need to have a whisper function, which students can use to contact the teacher with any issue directly without having other students see it. To my knowledge, none of the students in this study used the whisper function, so it may be necessary to draw students’ attention to this function. In response to item 38 on the survey, one student also stated that he wished that computer instructions were done in English. While this was only one student, because so many computer problems occurred due to errors in login and logout procedures according to the reports from the Program Developers who receives the error reports, it may be necessary to have one training session for students and teachers, and make procedural guidelines and troubleshooting guidelines easily available.
On a related note, since teacher and Lab Assistant were utilized to fix technological problems, again a training session may be necessary. Furthermore, it may be necessary to find a way in which students can communicate technological problems immediately to their chat partners, so that the chat partners do not feel ignored when a partner is not responding due to technical difficulties. Furthermore, minimum and ideal computer requirements need to be stated explicitly for students, and made available in some form to those students who do not have their own equipment that meets the requirements.
Since students cannot see which resources other students are using, providing a list of resources and making them easily accessible might be helpful for the students.
EveningTeacher also used course management software, where students could share useful links.
Two of the major functions of the physical space seemed to be the commonality of the experience and the bonding. While it is not clear from the data collected for this study whether this is an important function, it was a predominant function. Hence, I consider finding ways to make this possible in a distance learning context important. To allow students to still look over each other’s shoulders, the chat could be set up in a way that students can easily move from channel to channel. Furthermore, an added feature in which students can have a private one-on-one conversation with one of their friends may help with the sharing. In order to establish some commonalities or at least acknowledgement of differences, video software could be added to the chatting. Another less technologically demanding solution, could be to have students listen to the same radio station online while chatting or to ask students to describe their environment at the beginning of each chat session so that they have some common experiences or can visualize each other’s environments.
While these are all recommended modifications based on the findings of the events in the physical environment in this study, there are several limitations to these recommendations. First of all, no level of importance of the events in the physical environment could be established. Second, the effectiveness of the recommendations has to be established first. Third, only two teachers, and three classes were investigated.
Fourth, since the design of the computer lab in which the chatting occurred is so unique, some of the bonding elements may be a result of the collaborative set-up of the lab, rather than a necessity of the chatting.
4.7 Additional Findings In the previous sections, the research questions were addressed. However, the data also revealed some additional findings worth mentioning. In addition, some of the results posed further questions, which were then investigated. The following additional findings will be discussed: students’ attitudes towards technology and chat, students’ recommendations for chat improvement; potential mismatches between students’ desires and actual experiences; the relationship between teacher feedback and error uptake; the relationship between teacher feedback and student self-correction; and the issues of comprehensibility, fluency, accuracy, and target language use.
Looking at the results from the survey, students overwhelmingly liked technology in the classroom, and thought that chat was both beneficial for language learning (NSC 83.33% approval, SSC: 87.5% approval, and ESC: 77.78% approval) and fun (NSC:
83.33%; SSC: 87.5%;ESC: 66.67%). Students who explained why they liked technology gave the following reasons: (1) “visual” aspects, (2) change of pace, (3) “resource,” (4) “the future,” (5) “broadcast information,” and (6) available when the teacher cannot be.
However, two students cautioned that they did not find it “necessary.” Furthermore, one student specified that technology is only helpful when it is working and another student mentioned that the computer can be a hassle. Another student thought language learning is more effective when speaking it. Yet another student cautioned that effective use of technology requires computer literacy. NSCJennifer also mentioned that the computer is her only real way of communicating with people in Germany.
Students thought that chatting was fun because it was a form of communication that was different. However, many students who commented expressed caution, such as “only if the other person responds quickly and accurately;” “would rather be getting from the teacher;” or “as long as chat partner has a clue.” Several students mentioned issues with their partners in the survey comments and also in the self-report form which influenced their experience.
Students who explained their opinion thought that chatting was beneficial because it was different, “visual,” and “practice.” However, NSCDanielle felt that only her written German improved. Furthermore, SSCOphelia added “when I had a good partner.” In addition ESCLarissa felt that she did “not learn anything.” As has already become apparent from the previous section, several students had issues with their partners. Naturally several of the recommendations for improvement were regarding partner assignment. In the NSC and SSC students recommended that activities and procedures should be explained more clearly. In ESC students had varying recommendations regarding the timing and duration of the chat sessions. One student recommended that the transcripts could be used for corrective feedback.
Besides the students’ attitudes towards technology, attitudes towards corrective feedback were also gathered and already discussed previously. In addition, actual practices regarding feedback were also discussed. However, it needs to be stated here again, that they did not only perceive more corrective feedback than actual feedback given, but also their preference in receiving a lot of feedback was not matched. As a general trend students would have liked to have every one of their mistakes corrected, however less than 12% of the errors of any case study subject were corrected by others.
Therefore there is a mismatch between students’ desired teacher behavior and actual practices.
Another question that arose out of the observation of low teacher feedback was the question of whether lower teacher feedback then also means a higher error uptake rate. To investigate this further, the data from the six case study subjects were analyzed in regard to correlations in a two-tailed test of significance with the Pearson Correlation coefficient. Correlational procedures are used to establish relationships between two factors; however, they do not suggest a cause and effect relationship. Furthermore, a correlation with only six subjects may receive chance results and has to be viewed with caution. Therefore, this is an exploratory analysis that will need further investigation with more subjects in the future.
Table 4.30 Correlation between Error Uptake Rate and Teacher Feedback Rate
With only six subjects, the correlation between teacher feedback and error uptake was not significant (r = -.086, p.05) (see also table 4.30). This may mean that the absence of teacher feedback does not lead to an increase in error uptake as one might fear. However, as mentioned above the analysis cannot establish a cause and effect relationship and the number of subjects analyzed makes this an exploratory finding which needs further proof to be sustainable.
Another question that arose out of the observation that teacher feedback and selfcorrection were higher in SSCEmily’s case was the question of whether observing more feedback from the teacher has a relationship with self-correction. The data from the six case study subjects were analyzed with regard to correlations in a two-tailed test of significance with the Pearson Correlation coefficient. For the same reasons as discussed above this procedure was chosen and faces the same limitations. However, the correlation between teacher feedback rate and self-correction rate (r =.970, p.01) was significant (see table 4.31). This may mean that when students receive more feedback from the teacher, they may be more encouraged to self-correct or they may gradually develop more awareness of their errors, i.e. maybe teacher correction leads to ‘noticing’.
However, due to the low number of subjects, this is a prelminary finding that needs further exploration with other data or subjects.
Table 4.31 Correlation between Teacher Feedback Rate and Self-Correction Rate
The last question that arose from the data was the question of comprehensibility of chat transcripts and the relationship between fluency and accuracy. For the most part, the transcripts analyzed were comprehensible to the researcher. For the six case study subjects, 1617 errors were found. Of those only 68 (4.21%) were errors that lead to incomprehensibility as determined by my error coding. 31 of those errors were from ESCAmanda’s transcript. Therefore, it can be concluded that comprehensibility was high.
The data of all subjects were then analyzed for any relationship between fluency and accuracy with regard to correlations in a two-tailed test of significance with the Pearson Correlation coefficient. With 44 subjects, this correlation can be assumed to be more accurate in describing a relationship between fluency and accuracy. The correlation between words per minute (as a measure of fluency) and the post-test score (as a measure of accuracy) from all subjects was significant (r =.970, p.01) (see also table 4.32).
While some fear that students who are more fluent pay less attention to accuracy, this analysis suggest that students’ with higher fluency also have higher accuracy.
Table 4.32 Correlation between Fluency and Accuracy
As mentioned earlier, EveningTeacher used significantly more English words in the case study subjects’ transcripts than MorningTeacher. An interesting additional finding is that in comparing the target language use by the six case study subjects, there was also a significant difference found between the four case study subjects taught by MorningTeacher and the two case study subjects taught by EveningTeacher using a chi squared analysis (see table 4.32). MorningTeacher’s case study subjects used significantly more target language than the EveningTeacher’s case study subjects. This matches the difference between the two teachers and may suggest that the teacher’s frequency of the use of the target language may be seen as a model by the students for their own language choices.
Table 4.33 Case Study Subjects’ Target Language Use
Chi-square = 236.26 -- p 0.001.
4.8 Summary In this chapter, the research results in response to the five research questions were presented. In summary, there was a difference in participation patterns between the two case study teachers. However, these differences between the two case study teachers resulted in only a few differences in attitudes and language of the students investigated.