Enhancing English Language Learners ’ Conversation Abilities via CA-informed Sitcom Lessons

The study investigated the effectiveness of Conversation Analysis (CA)-informed sitcom lessons in enhancing conversation abilities of Thai learners of English. The participants included 42 high school students enrolled in an English for Communication course at a public high school in Southern Thailand. Through 15-week sitcom lessons, they were taught how to construct conversation sequences to accomplish such sequential actions as greeting and leave-taking, dis/agreement, new announcement, compliment, invitation, and request, as well as to collaboratively analyze conversations from the sitcoms and role-play them at the end of each lesson. Before and after the series of lessons, the participants were engaged in role-play conversations that were videotaped for subsequent assessment of their conversation abilities. The findings from both comparative statistical and close single-case analyses revealed significant improvements in all the aspects assessed especially regarding grammar and appropriacy. Therefore, it is recommended that EFL teachers should apply CA principles to teaching English conversation, integrating conversations from authentic materials such as sitcoms to strengthen English language learners’ conversation abilities.


Introduction
This study sheds light on the application of conversation analysis (CA), a sociological approach to the study of talk-in-interaction which emerged in the 1960s through collaborative works of sociologists Harvey Sacks, Emanuel Schegloff and Gail Jefferson.CA has been claimed to provide a comprehensive and systematic way of describing the structure of natural conversation and explicating the sequential organization of such common language actions as greetings vs. leave-taking, agreement vs. disagreement, news announcement, compliment vs. compliment response, invitation vs. decline, and apology, which are an essential part of interactional abilities (Barraja-Rohan, 1997, 2011;Wong & Waring, 2010).With regard to integrating CA insights into a language classroom, several studies have suggested that CA can serve as a tool for second language pedagogy especially in teaching interactional competence (Barraja-Rohan, 2011;Fujii, 2012;Markee, 2005;Seedhouse, 2005;Wu, 2013).Markee (2005), in particular, maintained that CA provides a clear and comprehensive understanding of the structure of conversation as a means of producing more effective, comprehensible output.In addition, conducting CA-based classroom research, Barraja-Rohan (2011) contended that the learners became successful in acquiring various aspects of language actions.Further, from the analysis of the development of learners' conversation abilities, it was suggested that while CA served as a powerful analytic tool to identify causes of problematic turns found in conversation, CA-informed instruction be introduced in an EFL classroom in order to enhance learners' conversation abilities.Likewise, Wu (2013)'s investigation of CA principles in teaching conversation via transcribing naturally-occurring data of native speakers' (NS) or non-native speakers' (NNS) conversations also revealed that conversation abilities of their learners increased via explicit CA-informed teaching.Similarly, Fujii (2012) conducted an experimental study applying insights from CA to raise Japanese ESL students' awareness of language-specific aspects.The study also suggested that CA-based instruction increased learners' interest in the mechanism of English conversational interaction while developing their interactional competence.
Despite becoming more widely accepted as a research methodology into L2 pedagogy, the integration of CA insights is still rare with lessons using authentic materials approximating naturally occurring conversation.Among authentic materials often used for teaching speaking in L2 classrooms, sitcoms, situation comedies aired on TV, are found to provide a useful source of input given its abundance of natural spoken language features (Washburn, 2001).As suggested by Martinez and Fernandez (2008), sitcoms can be used as a tool to develop L2 learners' conversation abilities since they allow the learners to be exposed to natural English through the verbal and nonverbal interaction of native speakers.Saito (2013) additionally highlighted that sitcoms such as Friends provide better samples of authentic English conversation than scripted conversation in general ELT textbooks.Investigating the use of Friends sitcoms to teach sequence organization such as refusals and invitations, Bacelar Da Silvia (2003), in particular, asserted that explicit conversation instruction via such sitcoms could facilitate L2 learners' development of conversation abilities.
Although the aforementioned studies seemed to confirm the advantages of both CA-informed instruction and the use of sitcoms in developing L2 interactional competence, there is a dearth of empirical studies investigating the effectiveness of the integration of the two (Markee, 2005).Especially, in Thailand, not only have there been a handful of studies investigating the contributions of CA insights to pedagogy in enhancing learners' L2 interactional competence (Abhakorn, 2017;Sitthikoson & Sinwongsuwat 2017;Teng & Sinwongsuwat, 2015), there apparently were also no studies exploring the use of sitcoms for teaching conversation and enhancing Thai learners' interactional competence.Thus, the present study aims at investigating the effectiveness of using CA-informed sitcom lessons in enhancing Thai language learners' conversation abilities.It particularly attempts to address the following questions: Can CA-informed sitcom lessons help to improve Thai students' conversation ability?And if so, in what aspects?

Background
Teaching conversation in Thailand is no doubt a daunting task for second language teachers since most students have little opportunity to interact with English speakers outside the classroom.Similar to other so-called EFL countries, one of the major obstacles is that many English teachers themselves have only limited exposure to the language of natural English conversation (Yunibandhu, 2004).Therefore, such an implementation of the teaching approach to building learners' communication skills as Communicative Language Teaching (CLT) has never been easy.As reported in Saengboon (2004), most teachers still felt more comfortable engaging students with pattern drills and rote memorization of separated sentences, focusing mainly on grammatical competence.Additionally, the washback effect has sadly made grades and examinations the main focus among teachers and students rather than the vital development of students' communicative competence (Islam & Bari, 2012).There is often a wide gap between the aims and methodologies prescribed by official curriculum statements and what actually happens in school classrooms.Thai teachers, in fact, had little understanding of CLT or any curriculum reforms (Tayjasanant & Roger, 2010).Therefore, as argued in Burns (1998) even though CLT has been promoted for many years in EFL contexts, there is still a need for improvement both in pedagogical practice and materials.Especially, when it comes to developing fundamental skills such as conversation, both teachers and learners, in fact, all stakeholders, still need to reexamine CLT-oriented pedagogical practices actually adopted in the classroom and to understand the genuine nature of the communicative skills being taught to reduce the gap and make CLT more successful in developing students with better conversation abilities.
To effectively undertake such CLT activities as role-play, simulation, skits, and games, some understanding of the mechanism of natural conversation is required by all parties involved (Hall, Hellermann, & Doehler, 2011).In order to successfully develop L2 learners' communicative competence, Barraja-Rohan (1997) and Wong and Waring (2010) even argued that features of such basic sequences must be taught explicitly to them.CLT activities can be carried out more efficiently in light of the complexities of naturally-occurring conversation uncovered via the lens of CA.In Thailand, while the Basic Education Core Curriculum B.E. 2551 emphasized Thai learners' communicative competence, the in-depth understanding of genuine nature of spoken interaction such as that revealed via CA has often been ignored.Therefore, Thai teachers have even more limited capacity to provide more comprehensible language input for learners to interact in the classroom so that they can unlock their L2 conversation capacity.
The implementation of CLT in the Thai context is consequently inadequate for learners to master conducting conversation in an effective way.As reported by Kwon (2017), while struggling with how to conduct CLT classrooms, Thai teachers are also confronted with government-prescribed textbooks that apparently do not serve the purpose of CLT.The textbooks used often lack genuine spoken language features and fail to provide real-world language contexts.They are therefore facing the dilemma of whether to follow the textbooks prescribed or to use home-made materials that would better meet the objectives of developing conversation abilities (Kanoksilapatham, 2007).Consequentially, in the Thai context, not only has the implementation of CLT widely been of a major concern, but also the teachers' dilemma also needs to be addressed.As Thai learners are still in dire need for exposure to communicative English through materials and teaching techniques which are more conducive to CLT, Wong and Waring (2010) suggested teachers acquire some background knowledge of naturally occurring conversation to develop teaching techniques and enhance the quality of the materials currently used, making them more appropriate for their CLT classrooms.
To incorporate CA insights into L2 pedagogy, Huth and Taleghani-Nikazm (2006) suggested that common actions done through turns at talk-in-interaction be systematically introduced via CA-informed instruction.Wong (2010) further postulated that the understanding of these social actions and their sequential organization is required for L2 speakers and should be explicitly taught in language classrooms.Seedhouse (2011) particularly added that concepts such as adjacency pairs, which illustrate the conditional relevance of turns at talk and deal with the basic type of sequence into which turns are usually organized, provides a comprehensive input for learners, allowing them to produce a comprehensive output.
Being introduced to how people systematically initiate and respond to each other in different types of sequential practices, learners are more likely to notice them when they come across these interactions in real life and stand a better chance for successful interaction in the target language.Illustrated in Wong & Waring (2010), one could hardly imagine for instance how learners with limited exposure to L2 interaction outside the classroom would be able to know how a complicated sequence such as a news announcement is typically organized into the following structural sequence: Therefore, CA-informed explicit teaching of conversation is essential (Abhakorn, 2017;Barraja-Rohan, 2011;Fujii 2012;Markee, 2005;Richards & Seedhouse, 2005;Sitthikoson & Sinwongsuwat 2017;Teng & Sinwongsuwat, 2015;Wu, 2013) and should be even more efficient with the use of authentic materials approximating real-life conversations such as sit-coms.It should be noted that since language is learned through cultural practices, sitcoms are full of cultural references and present a daily situation in which spoken language is uttered naturally (Ulusoy & Demirbilek, 2013).Further, as highlighted by Washburn (2001), sitcoms provide more abundant sources of appropriate conversational models compared to other genres available on television such dramas and soap operas since sitcoms mainly portray daily speech routines in which language is used by various users and settings such as at work, at supermarket, at home, and in public places.Morreale (2003), additionally, maintained that sitcoms basically focus on family, a workplace, and a community; consequently, the basic interactions uttered by differing characters can also represent the regular social interaction in everyday cultural context.Thus, the incorporation of sitcom materials into CA-informed conversation lessons should promise to help learners to learn the target language in a more meaningful way and to meet the demand of real-life conversation more easily.

Participants
The participants in this study were 42 students in Mattayomsuksa 6 (Grade 12) who studied English as a foreign language for 12 years following the Basic Education Core Curriculum B.E.2551 (A.D.2008) released by the Ministry of Education of Thailand.All of the students took English for Communication 2 (E30206) as an optional subject in the second semester of the academic year 2017 at Saiburi Cheangprachakarn School, Pattani province, located in the Deep South of Thailand.Their age average was 17 to 18.These students had little opportunity to use English outside the classroom.Most of them were used to being taught English conversation through scripted textbook dialogues, drill-based activities and rote memorization of separated sentences in the dialogues.The students were selected by a purposive sampling method for a quasi-experimental treatment with a one-group pretest-posttest design.

Instruments
The instruments used in this study included a pre-and post-test, 15-week and CA-informed sitcom lesson plans containing CA-based materials, conversation from sitcoms, and roleplay cards.All of the participants were asked to carry out 1-2 minute target language action sequences in role-play conversation with their peers before and after the instruction as pre-and post-tests.The sequences included greeting and leave-taking, dis/agreement, news announcement, compliment, invitation, and request.The students' role-play conversations were assessed in real time using a rubric of discrete speaking-related items with descriptors adapted form Barraja-Rohan (2011), Luoma (2004), andO' Loughlin (2001), previously used in Saniboo & Sinwongsuwat (2016), Sitthikoson & Sinwongsuwat (2017), Teng & Sinwongsuwat (2015), and Ussama & Sinwongsuwat (2014).The scoring criteria cover five domains: fluency, vocabulary, appropriacy, comprehensibility, and grammar (see Appendix A).The students' conversations were also recorded for subsequent close analysis assessment of their conversation abilities to perform the target sequences.
The lesson plans were executed throughout 15 weeks, with two hours of instruction per week.The main objectives were: (1) to enhance students' awareness of sequential organization of the focused language actions, (2) to use CA insights to explicitly teach the students how to construct the target sequence practices; (3) to enable the students to become both conversationalists and conversational analysts via observing conversations in American sitcoms, Friends, the video clips of which were prepared by the teacher to introduce focused language actions via CA lens; and finally, to enable the students to transfer skills in sequence practices into real-world situations via roleplay activities.
CA-based handouts given to the students were designed based on the concept of sequential organization discussed in Wong and Waring (2010), which targeted language actions such as greeting and leave-taking, dis/agreement, news announcement, compliment, invitation, and request (see Appendix B).
Excerpts of sitcoms conversations were taken from a famous American sitcom, Friends.The justification of incorporating this sitcom for pedagogical purposes lies in the fact that, as Quaglio (2007) argued, conversation in Friends is very close to naturally occurring conversation.Saito (2013) added that Friends presents natural conversation closings compared with conversation closings in typical ELT textbooks.Thus, Friends was chosen to give learners opportunities to observe, analyze, and discuss conversation through the CA lens.In the study, students were required to access the sitcom both individually and in group using their electronic devices such as smartphones and laptops.

Procedures
Before and after the instruction, participants' conversation abilities were assessed via a pre-and post-test in which they were asked to converse with their peers to carry out common language actions such as greetings vs. leave-taking, agreement vs. disagreement, news announcement, compliment vs. compliment response, invitation vs. decline, and apology.Videotaped for subsequent scoring by the researcher, learners' interactions were assessed by two Thai English teachers with respect to the following features: fluency, vocabulary, appropriacy, comprehensibility and grammar.The scores obtained from the two raters in the pre-test and post-test were averaged and statistically calculated for mean and standard deviation, and an independent t-test was used to determine significant differences in students' conversation abilities before and after the pedagogical intervention.In order to see the students' progress, close analysis of students' video recorded conversations in pre-and post-tests was carried out following CA principles.The conversations were also transcribed according to the transcription convention adopted by CA analysts Seedhouse (2005) and Schegloff (2007).
In each lesson, the teacher first tried to elicit the students' prior basic knowledge and the ability to perform focused language actions.Some sitcom excerpts were presented in order to engage the students in analyzing and discussing how turn-taking occurs.In the presentation stage, the learners received lectures explicitly introducing them to language actions based on CA perspectives, while CA-based materials were also provided.Then, the participants were engaged in observing, discussing, and analyzing a sitcom episode subtitled in English.At this stage, when speech acts were identified, the excerpts were repeatedly replayed until the learners could relate the conversation to CA concepts.In the production stage, the learners were assigned to perform role-plays in pairs.

Results and Discussion
After the 15-week instruction aiming at investigating the effectiveness of using CA-informed sitcom lessons, the post-test scores were processed in the same way as the pre-test scores.An independent t-test was used to compare the pre-and post-test scores.As shown in Table 1, the statistical results indicate that the instruction helped to enhance learners' conversation abilities in all the aspects assessed: fluency, vocabulary, appropriacy, comprehensibility, and grammar (see Appendix B for descriptors) with a significant degree of difference at a level of 0.01.The difference in the overall mean score and Cohen's (1988) effect size of 1.25, in particular, show that incorporating CA-informed sitcoms lessons into second language pedagogy could obviously enhance the English language learner's conversation abilities.After such an instruction, not only the students' overall conversation performance but the performance related to each item were significantly improved; therefore, the first research question was confirmed.** Significant at level 0.01.
However, despite significant performance improvement shown by these statistical results at a level of 0.01 on every aspect assessed, the degree of improvement noticeably varied among the five aspects; appropricacy (t= 14.42, sig = 0.01) and grammar (t= 24.52, sig = 0.01) apparently were most prominently developed via CA-informed sitcom lessons.Regarding appropricacy, since the students were allowed to repeatedly see the clear pictures of sequential organization of different speech acts throughout the course, they were able to initiate the conversation more naturally, respond to each other's turns more relevantly, and expand upon the turns meaningfully, thus conversing more appropriately with their interlocutors.Furthermore, the lessons helped the students not only to improve their ability to respond to the turns appropriately but also to employ a wide range of structures to accomplish each interactional goal.Asked to analyze conversations based on CA principles throughout the course, the students became more aware of a wide range of structures employed to perform different actions and were able to put them into use in their own talk.Therefore, it can be argued that CA-informed sitcom instruction could boost students' abilities to use English more appropriately in conversation, enabling them not only to construct grammatically accurate, conversationally suitable, and sequentially relevant turns but also to organize them into appropriate sequences of actions.These findings were strengthened through the close analysis of the student's conversations.As will be shown in the analysis below, after receiving explicit CA-informed explanations on the sequential organization of speech actions, students became more aware of the overall structure of conversation and were able to organize it with an appropriate opening, followed by centering and a closing sequence.Such structuring also allowed them to perform the target actions more appropriately and effectively.
While appropriacy and grammar apparently were the two areas in which the learners could benefit the most from the CA-informed lessons, similar to Ussama's (2013) and Teng and Sinwongsuwat's (2015), comprehensibility of turns remained the area to be worked on as the students had difficulties in understanding their peers' utterances.With regular exposure to natural English as appeared in sitcoms, the students could however more easily master the pronunciation of words and sentences and especially the intonation of speech in the target language and make their utterances more intelligible.Despite more challenges to be overcome in this aspect, it can still be maintained that CA-informed sitcom lessons are beneficial as CA pedagogical concepts allow L2 learners to understand various conversation practices, while sitcoms provide them with models of natural near-natural conversation to be observed and analyzed regarding appropriate language use.
Based on the statistical results, CA-informed lessons can satisfactorily improve students' conversation performance overall and the improvement can be further attested especially since their videotaped conversations were closely scrutinized.All of the 21 pairs of students showed noticeable improvement particularly in their turn construction and conversation expansion.As illustrated in talk excerpts 1 and 2 below, where two students, Nasran and Fitriya, were asked to deliver good news through role-play, their conversation was appropriately organized and expanded.Compared to the pre-test, Nasran was able to effectively announce his good news about getting an AFS scholarship to his friend Fitriya, while the latter also seemed to be able to respond to her interlocutor's turns more appropriately, not only keeping the conversation going smoothly but also rendering a more expanded, well-organized sequence than in the post-test.
01 Nasran: Hey Friend↓ I got an AFS scholarship.
04 Fitriya: ((Laughing)) As seen in their role-play conversation from the pre-test, Nasran jumped into the news right after getting Fitriya's attention with the turn preface, 'Hey friend,' without even preparing the latter for his delivery.Forced to respond immediately to the news, Fitriya could only provide a brief formulaic uptake with a repetition of the news.The sequence is then abruptly ended with a minimal third-turn expansion expressing gratitude and laughter, suggesting a failure to show interest in the partner's good news and to keep the conversation going.However, in the post-test, the two students were able to carry out a more extended conversation via a pre-and a post-announcement sequence.Via the pre-announcement sequence in lines 1-3, Nasran knew how to observe his partner's interest and to get her to pay close attention to the news he was going to deliver.The latter also knew how to signal her interest in the good news via the token of news receipt in line 5 before appropriately responding to it in line 7. Via high pitch and intonation, both parties were able to convey their excitement about the good news being delivered.Fitriya, in particular, was not only able to spontaneously respond to the news but to find out more details about the news from her partner with the question 'When will you go abroad?' making it possible for both parties to continue the conversation via the post-announcement sequence, lines 9-12.Compared to the excerpt taken from their pre-test, the conversation in the post-test was obviously more meaningful and elaborate.
Excerpt 12 Nasran: Thanks:: Similar improvements were also evidenced when the students were engaged in an invitation sequence.Shown in excerpt 3 taken from the pre-test, Asmah delivered an invite turn right from the start of the conversation.However, not immediately responded to, she had to repair her invite by providing a reason for it, prompting her co-participant Rusna to simply accept it.Instead of accepting the invite in the first pair-part, the latter chose to abruptly end the conversation with a leave-taking turn in line 4.No expansion was done with questions, such as "when" and "where", in order to inquire more information about the event; thus, the post-invitation sequence was missing.The two parties instead burst into laughter due to their wrong word choice of time in lines 5 and 6. (Sure) I will Apparently, with their awareness of pre-and post-expansion sequences, students could structure a more appropriate and meaningful conversation.These close analyses also affirmed the findings of the previous studies by Barraja-Rohan (2011), Fujii (2012), Sitthikoson and Sinwongsuwat (2017), and Teng & Sinwongsuwat (2015).

Conclusion
This study investigated the effectiveness of CA-informed sitcom lessons in enhancing English language learners' conversation abilities of Thai high school students.Based on both statistical results and close analyses, performance improvements were noticeable overall and especially in grammar and appropriacy, which is in line with previous studies by Sitthikoson and Sinwongsuwat (2017) and Teng and Sinwongsuwat (2015).It was thus confirmed that CA-based instruction indeed helps L2 students enhance their conversation abilities.
CA-informed instruction together with an authentic material such as sitcoms, referred to in this study as CA-formed sitcom lessons, could be wisely introduced as a promising approach to boosting L2 learners' conversation abilities in a CLT classroom.While CA could serve as a powerful pedagogical tool to help learners gain more understanding of the complexities of sequential organization of naturally-occurring conversation, natural or near-natural conversations from sitcoms could provide a rich resource for acquiring spoken language features such as vocabulary and expressions particular to talks as well as stress and intonation.It is, therefore, suggested that EFL teachers implement the principles of CA in language teaching particularly to enhance their students' interactional competence and integrating talk-in-interactions from sitcoms into their lessons to boost students' motivation to learn the target language as they can better relate to how the language is actually used in real life.
However, despite the apparent effectiveness of CA-informed sitcom lessons, there were some inevitable constraints in their implementation that need to be considered.First, this research was conducted with high school students and some of them seemed to have problems in understanding the complexities of said CA concepts and terms, such as sequential organization of conversation.It is therefore recommended that explicit CA-informed teaching be implemented to more advanced students.Second, teachers should be aware of cultural specificity of particular sitcoms.Therefore, to appropriately utilize them they might choose some excerpts that illustrate the sequential organization of focused acts of speech for students to analyze and discuss and then point out to them the linguistic and cultural variations that may occur in conversation in other cultural settings.
Despite these constraints, it has been highly recommended that excerpts from sitcoms be integrated into EFL classrooms apart from scripted dialogues in textbooks (Washburn, 2001) to help students notice not only features of natural talk-in-interaction but cultural differences in the target language (Huth & Taleghani-Nikazm, 2006;Wong, 2002).CA can serve as a powerful teaching tool to help students gain more understanding of the complexities of sequential organization of naturally-occurring conversation and become more confident when making English conversation.EFL teachers are recommended to implement the principles of CA in language teaching particularly to enhance their students' interactional competence and integrate talk-in-interactions from sitcoms into their lessons to boost students' motivation to learn the target language as they can better relate to how the language is actually used in real life.

Table 1 .
Comparison between Pre-Test and Post-Test conversation performance Evidently, after receiving the explicit introduction to doing invitation through CA-informed sitcom lessons, the same pair of students could conduct a more appropriate, well-structured conversation compared to their pre-test, showing the same development as in the aforementioned pair.Illustrated by the excerpt taken from the post-test below, Asmah and Rusna began the conversation with an opening constructing 'greeting' and 'how are you' sequences, lines 1-3.With better awareness of sequential structure of an invitation sequence, Asmah, in particular, knew how to ask questions such as 'What are you doing this weekend?' to assess the interlocutor's response to her invite.With Rusna's go-ahead turn in line 4, Asmah then gave information about her upcoming birthday in line 5, to which the partner responded with a positive comment, projecting an imminent acceptance of the invitation.The acceptance was in fact confirmed in line 10 after the invitation.'Will you come to join us?' was launched in line 9.