Exploring the use of turn-taking and overlap resolution strategies among Vietnamese non-English major students

Faculty of Foreign Languages, Ho Chi Minh City University of Food Industry, Vietnam

Truong, T.N.N. (2023). Exploring the use of turn-taking and overlap resolution strategies among Vietnamese non-English major students.Language Value, 16(1), 145-187. Universitat Jaume I ePress: Castelló, Spain. http://www.languagevalue.uji.es.

June 2023

DOI: https://www.doi.org/10.6035/languagev.7237

ISSN 1989-7103

ABSTRACT

Based on Sacks et al.'s (1978) turn-taking model, the research aims to determine how Vietnamese non-English major students used turn-taking and overlap resolution strategies to manage their discourse during English tutoring sessions. Two Vietnamese students were conveniently selected for the study, and their conversations were recorded, transcribed, and coded in the style of conversation analysis and deductive content analysis. The results show that although the male student used more devices and strategies than the female in taking turns in one-at-a-time talks, both employed latching to reduce transition space most of the time. Also, the male student tended to use overlap resolution strategies (e.g., cutting off his talk and persevering in completing his turns) in simultaneous talks more frequently than the female. Interestingly, the female student used more turn-taking strategies in overlapping speech than in one-at-a-time talks. The findings shed light on Vietnamese non-English major students' strategy use during interactions and are a great boon to English educators at tertiary institutions who should consider training student tutors, providing English language learners with necessary interactional resources, and rethinking speaking assessments.

Keywords:overlap resolution, turn-taking, strategies, Vietnamese students.

I. INTRODUCTION

Language is obviously and essentially a powerful tool that we use to serve our demands and activities in every minute and second of our life, such as communicating, studying, and working. Language is also one of the master keys to helping us improve our life. However, it is not with this assertion that we can allow our language to flow out as much as we want when we interact with others. It is because talks feature a face-to-face conversation in a synchronized manner in which one person stops, and the other starts talking (Wiemann & Knapp, 2008). We sometimes delay or stop our turn to let others speak while conversing with them in various social and cultural contexts. In this way, we have engaged in managing our discourse.

In language education, conversation analysis (CA) within discourse analysis has enabled researchers to assess the relationship between the nature of pair interactions and the success of language learning (Storch & Aldosari, 2013). Several elements determine if this correlation is positive. Firstly, if language learning is to be successful, pair interactions need to be collaborative. The beneficial effects of the collaborative nature of pair interactions for language pairwork or tutoring activities have been recorded in language classrooms. Specifically, cooperative work promotes social and cognitive development (Storch, 2001) and allows second-language learners to use the target language (Long & Porter, 1985; Washington-Nortey et al., 2022).

Pairing students effectively also depends on participants' proficiency levels. Studies in pairwork for L2 learning research indicate that proficiency levels significantly influence achievement and engagement in collaborative pairwork (e.g., Kim & McDonough, 2008; Leeser, 2004). Low-level learners made significant gains in the discussion of L2 when paired with a higher proficiency interlocutor (Storch & Aldosari, 2013) and produced more language-related episodes (Kim & McDonough, 2008; Leeser, 2004). Similarly, research in cross-age tutoring in which an older student with a higher proficiency level acting as a tutor for a younger student suggested that cross-age tutoring benefited both the tutor and the tutees (e.g., Davenport et al., 2004; Hattie, 2006).

Since most studies on pair interactions have investigated the effects of pairing different proficiency levels, very few studies in this line of research have touched on gender. The majority of psychological, sociological, and linguistic literature represents men and women as possessing different characteristics in their approaches and use of language (Yates, 2001). Hence, both sexes can demonstrate salient features in controlling their discourse, affecting the effectiveness of pair interactions. Understanding how each sex manages their discourse, i.e., organizing their talk via turn-taking, will help language researchers clear myths about genderlect and language educators better assign students for pairwork or tutoring activities.

To explain how ordinary conversations and other talk-in-interaction are organized, Sacks et al. (1978) introduced the notion of turn-constructional units (TCUs), the basic unit of talk for CA. TCUs help CA researchers understand turn-taking organization, which is crucial in deciphering human behavior because the organization of talk shapes most actions carried out through talking into speaking turns (Lerner, 2004). Although CA has been applied to linguistics and education in recent decades (Mori & Zuengler, 2008) and many studies concerning this approach have been published for second language teaching research (e.g., Bowles, 2006; Fujii, 2012; Liddicoat, 2004; Saadi Ali, 2021; Wong, 2002), empirical studies on turn-taking organization in South East Asian (SEA) English teaching contexts, where English is mainly taught as a foreign language (EFL), are rare.

Thus, the current study was the first in Vietnam to address the organization of turn-taking during paired cross-age tutoring interactions using conversation analysis. In particular, the study explores how Vietnamese male and female students manage their discourse, especially in turn-taking and repairing overlapping speech during cross-age English tutoring sessions. The results of this study would contribute to the CA and teaching English to speakers of other languages (TESOL) literature. The findings may also offer language educators ideas about the reconceptualization of students' speaking competence and assessment. In particular, the study seeks to answer the following research questions:

What devices and strategies do participants use to take turns in one-at-a-time talks during English tutoring sessions?
What devices and strategies do participants use to repair overlapping talks during English tutoring sessions?

II. LITERATURE REVIEW

II.1. Turn-taking in talk-in-interaction

II.1.1. Deontic modality

This section first elaborates on the organization of turn-taking in talk-in-interaction, including the definitions of talk and turn-taking, essential features of turn-taking organization, followed by devices and strategies for turn-taking in one-at-a-time talks, and the repair of overlapping speech. The literature ends with recent works on genderlect, an essential foundation for understanding how male and female students take turns in their exchanges.

II.1.1.a. Talk

When people converse, they engage in an interactive, meaningful activity called talk, where they can strategically achieve their communicative goals (Sacks, 1992). However, this activity is contextually dependent in that the context shapes it, so speakers can understand what follows their produced speech. In turn, talk can shape the context by restricting and affecting the next bit of talk and determining how they are comprehended. When closely observing what naturally happens in talks, we see that people do not usually speak simultaneously at all times in ordinary conversations. They take turns talking. Thus, a turn can be interpreted as "one party speaking at a time" (Sacks, 2004a, p.37) or actions in sequences (Ford et al., 2002), or an on-record speaking behind which lies an intention to convey a referential and functional message (Edelsky, 1993). However, not all talk can be counted as a turn. A talk, said off-record, usually in a low voice, is just a side comment (Edelsky, 1993). Likewise, talks intended to give feedback, not referential messages (e.g., uh-huh, and uhm), are considered encouragers or back-channel responses (Sacks et al., 1978).

II.1.1.b. Turn-taking

Turn-taking, a type of sequential organization, focuses on the logical and mechanical nature of the conversation. It concerns the "relative ordering of speakers, of turn-constructional units, and different types of utterance" (Schegloff, 2007, p.2), or in other words, the "allocation of opportunities to participate in the conversation and the turn-constructional forms such participation takes" (Lerner, 2004, p.4). Since speakers manipulate their chances to participate in the conversation, which follows specific patterns or rules, turn-taking is a "closely monitored and coordinated joint activity" (Ford et al., 2002, p.15). Although many turn transitions are achieved without overlaps or silence, they can be patterned and explicable even when overlaps or gaps emerge. A gap-free turn transition and changes involving overlaps or gaps are all "interactionally exploited alternatives" (Ford et al., 2002, p.15). Thus, turn-taking considers two cases: pure turns, i.e., one-at-a-time turns, and diffused turns, i.e., overlaps and gaps.

II.1.2. Features of the turn-taking organization

In preserving one party talking at a time, techniques to allocate and construct turns are necessary for just one next speaker and minimize gaps and overlaps between turns (Sacks, 2004a). Therefore, explaining how turn-taking works as a set of rules requires understanding its two components: turn-constructional units (TCUs) and turn-allocation (Sacks et al., 1978).

II.1.2.a. Turn Constructional Units (TCUs)

TCUs are building blocks of turns to allow the projection of a possible completion point, called transition relevance place (TRP), which enables speaker change (Sacks et al.1978). TCUs have two main criteria: syntactic structures and projectability (Selting, 2000). Although syntactic structures include grammatical elements such as words, phrases, clauses, and sentences, they are not structurally defined units because they are context-sensitive. Also, a decision about what constitutes a TCU can only be made in the context because people do not always utter in sentences but tend to deploy a variety of structures (Schegloff, 2007). However, three ways can determine if an utterance is possibly complete: firstly, in terms of grammar; secondly, intonation; and finally, as an action (such as asking a question or offering to help) (Sacks et al., 1978).

Naturally, on engaging themselves in the conversation, participants do not always know when a turn ends; they continue the conversation intuitively. However, this can be accountable. Turn-taking occurs in the transition space, the space between speakers' turns, considered part of a stretch of talk in which transition may occur. Transition space commences just before a TRP and finishes just after the end of a TRP (Sacks et al., 1978). A TCU can be applied at the possible and projectable completion points, i.e., TRP (Schegloff, 2007) or, in other words, places where speaker change could occur. In another sense, the interlocutor knows what it will take to complete the ongoing unit of talk, and they can project where an ongoing TCU will possibly be completed. This projection is essential for the organization of turn-taking because the addressee will not have to wait until the addresser completes their turn to become the next speaker (Liddicoat, 2021). Hence, the projectability of TCUs at TRP is a catalyst for a smooth transition with no pause between turns because a pause may be interpreted as a delay or absent response (Liddicoat, 2021).

Sometimes, the same piece of talk is not a new TCU when not recognized as possibly complete at a particular point in the ongoing talk (Liddicoat, 2021). In other words, they continue the speech through many turns, called a multi-turn TCU, a unit that can spread through many turns at talk. Besides, a multi-turn TCU can contain only one stretched TCU (Schegloff, 2007). Notably, a turn can consist of many TCUs, and thus it is called a multi-TCU turn. For example, if the speaker tells a story, they may produce many TCUs. It means that they continue to produce the next TCU to accomplish their communicative goals after completing a TCU. However, the current speakers may have equal opportunities to take turns because the ability to produce more than one TCU in a turn is "the result of interactional work, not the result of a right to produce more than one TCU" (Schegloff, 2007, p.4).

II.1.2.b. Turn-allocation techniques

In addition to turn-constructional techniques, turn allocation can minimize gaps and overlaps between turns (Sacks, 2004a). If TCU explains where speaker change can happen, turn allocation will account for how it occurs. Sacks (2004a) explained four basic rules of turn allocation to ensure one party talking at a time. The first is that the current speaker selects the next speaker using linguistic and syntactic forms (e.g., you, your, and questions) or gestures (e.g., a gaze). In the second rule, the current speaker stops at the next possible completion point of their sentence construction to allow the next speaker to start. Third, self-selection occurs when the current speaker does not select the next speaker. This process can happen where the previous talk is planned to require someone to speak next but does not restrict who will. The first starter has the right to talk. The fourth rule allows the continuation beyond any TRP if the current speaker does not select the next speaker and self-selection does not occur. The current speaker may stop at any next possible completion point.

II.1.2.c. Overlapping talk

In contrast to one-party-at-a-time talk, overlapping talk is an interactional phenomenon produced by speakers in unison. Unfortunately, sometimes, overlapping can be confused with interruption, which has a negative connotation. Interruption is undesirable behavior violating standard conversational rules (James & Clark, 1993). If two speakers begin a TCU simultaneously, this overlapping talk is deemed problematic and thus is seen as a case of interruption (Sacks, 2004b). In this sense, interruptions occur when entry into the talk is not related to a possible completion (Liddicoat, 2021). Deciding if an overlap is an interruption is just a matter of degree (James & Clark, 1993), and thus the context must be considered (Tannen, 1994). Therefore, as interruption is a part of the overlapping talk, overlaps can be problematic and unproblematic. If the overlap is short, i.e., occurring just before possible completion, it is unproblematic and seen as a collaborative interaction (Liddicoat, 2007).

II.1.3. Turn-taking devices and strategies

Because turn-taking does not work at the level of the whole conversation but at the level of each next bit of talk that unrolls as the conversation continues, it is "locally organized and interactionally managed" during the interaction process (Liddicoat, 2007, p. 54). Strategies for turn-taking organizations examined locally will effectively and concretely explain how speakers manage their discourse. The primary strategies and devices of turn-taking organization for the one-at-a-time talk are expounded as follows.

II.1.3.a. Constructing a Multi-Turn TCU

Firstly, structural patterns can be used to construct a multi-turn TCU. For example, a multi-turn TCU can be realized using complex sentences such as conditional sentences (e.g., if and unless), time clauses (e.g., when and while) (Lerner, 2004), and adverbial clauses (e.g., because, since, before, although, and so that) (Ford, 1993). Let us examine the following example:

Mai: If we win the lottery ticket=

Dane: =we will buy a big house first.

This first component of the construction can imply a possible completion and makes it possible for the current speaker to predict the next. As a result, two participants cooperatively completed a single TCU ('If we win the lottery ticket, we will buy a big house first') over two turns at talk. Another way to achieve a multi-turn TCU is by adding an increment to the talk of the prior speaker so that the produced one will become an integral part of a grammatical unit (Liddicoat, 2007). An increment is any nonmain-clause continuation of a speaker's turn after that speaker has come to what could have been a completion point, or a TRP, using lexical devices (e.g., a noun phrase, a prepositional phrase, and subordinate clauses (Ford et al., 2002). The following example will illustrate how this device is used.

Anna: We can go to Mui Ne resort this weekend to relax

Mary: and to get a suntan

Daisy: and to buy some seafood

We see that the speech produced by Anna can be complete without further talk. However, what Mary and Daisy utter in their turn is just a continuum to fit the grammatical construction of Anna's utterance. Therefore, the utterances "We can go to Mui Ne resort this weekend to relax," "and to get a suntan," and "and to buy some seafood" are attributed to only one TCU, which is stretched through three turns.

II.1.3.b. Constructing a Multi-TCU Turn

For a current speaker to produce more than one TCU in a turn, specific extending TCUs techniques have to be employed (Schegloff, 2007). There are three places to realize a multi-TCU turn.

At the beginning of the turn, transitions and signal words such as 'first of all,' 'in the first place,' 'next,' and 'however' will be helpful to indicate that the speaker may produce a longer than usual piece of talk (Liddicoat, 2007). Likewise, the current speaker can preface their talk with such rhetorical questions as 'Can I ask you a question?' and 'Can I ask you a favor?' to ensure an extended turn in the next turn (Liddicoat, 2007).

In the middle of the turn, a non-linguistic and less overt device such as large audible breathing can also be deployed, implying that there will be a longer than usual bit of talk and more breath is thus necessary to fulfill it (Liddicoat, 2007). Also, deictic devices such as personal pronouns, demonstratives, adverbs, and tenses can refer to some previously produced speech, which needs clarifying, and thus, further talk is necessary (Goodwin & Goodwin, 1992).

At the end of the turn, the current speaker can employ a 'rush through' technique that reduces the transition space between two TCUs and restrains falling intonation (e.g., speeding up speech delivery) (Liddicoat, 2007).

II.1.3.c. Increasing Transition Space

During a conversation, when no one may speak at all, there can be short pauses (i.e., gaps). Some silence can happen, lasting for a few seconds or a few minutes. In this case, the long pause, called lapses, is not ascribed to any party involved in the conversation. The reasons for lapses can be attributed to either the form of the prior speaker's turn or the recipient's hearing problems. In this case, the current speaker can tackle silence after TRPs by continuing with further talk or repeating their utterance, thus increasing the transition space (Liddicoat, 2007). The following examples illustrate this consideration.

Conversation 1:

Steve: so are yih gonna be free on the weekend, (0.4)

Steve: say on Saturday evening

Mary: yeah

Conversation 2:

Lan: Have you ever gone to America? (0.3)

Lan: Have you gone to America?

Mai: oh, no. I haven't.

In conversation 1, there is a 0.4-second pause after Steve's suggestion. This silence is an indication of some problem in the talk. In this case, it is a problem with the turn construction unit's form. Therefore, Steve continues his original turn at talk in an attempt to fix the problem in his second turn by clarifying his speech. In this way, Steve has added an increment to his talk to transfer from silence between his and Mary's turn into silence within his turn. Similarly, in the second conversation, there is a 0.3-second silence after Lan's query, attributed to the hearing problem. As a result, Lan has to repeat her question in her second turn to elicit the answer from Mai.

II.1.3.d. Reducing Transition Space

Absolute adjacency pairs can help reduce the transition space because the next speaker can latch their talk to the prior speaker's, so there will be no silence between their turns and no overlap (Jefferson, 1986). An adjacency pair comprises two relatively ordered turns delivered by two different speakers, one after the other. The speaking of the initial utterance (the first-pair part or the first turn) provokes a responding-related utterance (the second-pair part or the second turn) (Schegloff, 2007). Common adjacency pairs include offers-acceptance/rejection, question-answer, greeting-greeting, and invitation-acceptance/refusal (Archer et al., 2013). The moment the first speaker ends their turn (e.g., offering help), the next speaker starts their turn immediately (e.g., accepting an offer). The following example will illustrate this point:

Elle: Would you like a coffee?=

Dave: =Yes, please.

An additional way to reduce transition space involves creating a bit of overlapping talk between the current speaker and the next speaker (Liddicoat, 2007). The speaker starting first is more likely to get their turn if multiple speakers start simultaneously (James & Clark, 1993). The onset of the talk may be put forward to an earlier start than a usual transition space to speed up the chance to become the first speaker. This technique can create interactional effects, e.g., showing understanding, disagreement, or refusal of the prior talk (Liddicoat, 2007).

II.1.4. Overlap resolution

Regarding turn-taking organization, it is not always the case that one speaker talks at a time. Overlaps can occur when interlocutors take turns simultaneously, i.e., simultaneously sharing and contributing their understanding or thoughts to the discussed topic. This engagement of the next speaker during another's talk is considered a "miscue in the turn-taking system" or an interactional problem (Liddicoat, 2007, p. 87). It is clear that sometimes overlap can occur before the start of the transition space in a way that diverges from the ongoing completion of the talk. The extended overlap is inevitable when speakers are engrossed in the ideas they are pursuing and persist in talking in a way that goes beyond the turn-taking system. Hence, overlapping talks need to be resolved.

II.1.4.a. Overlap resolution devices

Schegloff (2000) elaborated that an overlap resolution device comprises three elements. The first element is broken down into resources interrupting the ongoing talk, including hitches and perturbations. Hitches include cutting off the talk by an oral, glottal, or velar stop, prolonging a segment of talk, and repeating just a prior element. Perturbations or prosody of the turn consist of increased volume, faster or slower pace of talk, and higher pitch. Volumes relate to the loudness or softness of the sound, while pitch relates to the high or low notes of the sound (e.g., stressed words having high pitch and unstressed sounds denoting low pitch). The second element deals with places where these resources can be employed. The current speaker can use overlap devices in either two positions: at the onset or the end of an overlap. At the onset phase, the interlocuter can increase the speed of the talk to prevent another person's starting. The beginning of the resolution can also be prefaced with hitches and perturbations to hold the talk before possible completion. They can then delay finishing towards the end of the overlapping speech by decreasing the pace of the talk using sound stretches and repetitions until the other overlapping speaker reaches a completion point. The overlapping resolving devices are occasionally deployed after the speaker's talk has emerged into the clear. After winning the turn, i.e., post-overlap, the speaker can adjust their voice and pitch and speak normally. It should be noted that there are no hard-and-fast rules concerning which position in an overlap that resources can be used because participants can resort to kinesthetic devices (e.g., a gaze and body gestures).

The third element is related to the interactional logic of overlap resolution. The unit concerning the overlap resolution is the beat, a relative equivalence to a syllable. An overlap can comprise a series of emerging beats providing locations for organizing and sequencing the bit of talk. There are three possibilities at the point of overlap. First, either of the two speakers can stop, thus solving the overlap, i.e., "a return to one person speaking at a time" (Liddicoat, 2007, p. 91). When many speakers take turns simultaneously, it is difficult to allocate the turn because there is no turn-taking sign after possible completion. Several explicit devices can also help to tackle the problem. The speaker can say 'Who me?' to decide the next speaker, or any interlocutor can say 'Pardon me' to change the speaker (Liddicoat, 2007, p.73). However, the interlocutor, chosen as the next speaker, is not determined in advance; it happens naturally at the moment of overlap.

The second possibility is that both of them can discontinue speaking, and silence will ensue; as a result, the turn-taking has problems and needs fixing. In this case, the current speaker can continue to repair the silence after a TRP (Schegloff, 2007). In the last possibility, both speakers continue, and the overlap proceeds into a second beat. In this case, either of them is aware that the other is speaking and thus either stops or continues. If they stop, they will return to the first possibility, resolving the overlap in the third beat. Conversely, if they continue, they will compete seriously for the next turn, which can persist into the third and the fourth beat. At this point, both speakers will resort to overlap resolution devices (i.e., hitches and perturbations) to upgrade their talk, and as a result, severe competition for the floor is unavoidable. However, at this point, either of the two interlocutors will usually retreat from the exchange, thus resolving the overlap.

II.1.4.b. Management of overlapping speech

Schegloff (2000) explained three criteria to be successful in overlap management. The first criterion includes persevering in completing by producing a talk relaxingly as if no one is speaking simultaneously as a sole speaker to bring their talk to a projected completion without using hitches or perturbations. The second criterion is to project the thrust of the turn. The final criterion is to achieve sequential implicativeness, which means language is linked to a linear sequence, and conversational turns are logical because they are clarified in sequence.

Ex: (1) Mary: it is hot, today, isn't it? =

(2) Huong: = Yeah. I do not feel like going out tonight

In the above conversation, Mary comments on the weather in her turn to receive agreement from Huong (turn 2), which is successful sequential implicativeness. In some cases, during the overlap, the speaker can delay their talk, i.e., preempt their completion to respond to the other speaker's comment and then turn back to their talk, which can be called a "collaborative turn sequence" (Lerner, 2002).

II.2. Genderlect

The term genderlect is coined to define the language of sexes. Unlike dialect, which refers to the unique language of people in a specific geographical area, genderlect is a variety of languages tied to the speakers' sexual gender (Orasanu et al., 1979). Most literature on genderlect focused on the association between gender roles and gender-associated speech, displaying women's inferior social status (Bilous & Krauss, 1988) with attributes such as timidity, dependency, and incompetence (Lakoff, 1975). Tannen (1990) highlighted that women and men had different conversational styles in that women speak the language of intimacy and connection, whereas men speak the language of independence and status. Besides, Tannen (1990) claimed that women talk more than men in private conversations, while in public, men speak the most and attempt to gain status. For men, talk is for information, while for women, telling things is a way to show involvement, and listening is a way to show interest and caring.

The way men and women use language to communicate verbally is very dissimilar. Women are more active than men in supportive roles in conversation by using a lot of back-channel support such as the use of hedges (sort of, kind of, I think), fillers (e.g., you know, sort of, well, you see) and epistemic modal forms (e.g., should, would, might, could, may) to indicate indirectness, uncertainty, hesitation and reluctance (Thomas & Wareing, 2004). Furthermore, women use more empty adjectives (e.g., divine, charming, cute), intensifiers (e.g., so, just, quite), and rising intonation in declarations to attract attention, seek agreement, and show emotion (Holmes & Wilson, 2017). Regarding verbal communication when listening, women tend to give more listening responses (e.g., mhm, uh-uh, yeah) than men, and the signals they provide also have different meanings (Tannen, 1990). While women use 'yeah' to mean 'I'm with you and I follow', men tend to say 'yeah' only when they agree (Tannen, 1990). Hence, when a man confronts a woman who says 'yeah', he interprets it as showing agreement, but if the woman turns out not to agree, he may conclude that she is insincere or agrees without listening. Vice versa, when a woman confronts a man who does not say 'yeah', she might suppose he had not been listening.

In a review of studies conducted between 1965 and 1991 on gender differences, James and Clark (1993) concluded that women tend to produce more cooperative overlapping talk than men and that women have a higher tendency than men to use simultaneous talk to show rapport and involvement. Also, the authors found mixed results, which seems inconsistent with Tannen's (1990) report regarding the claim that men tended to interrupt more than women. While in some studies (e.g., Bilous & Krausse, 1988; Dindia 1987, as cited in James & Clark, 1993), no significant differences between genders in the number of interruptions were documented, some studies revealed that men interrupted females more significantly (e.g., Bohn & Stutman, 1983; Esposito, 1979, as cited in James & Clark, 1993) and other studies reported that interruptions initiated by women were more significant than those by men (e.g., Sayers, 1987; Murray & Covelli, 1988, as cited in James & Clark, 1993).

Nevertheless, whether gender differences exist is a very controversial issue. While some researchers posit that gender differences are compatible with the frequent use of certain speech forms, others argue that the preconceptions and prescriptions about gender differences do not reflect the genuine picture of gender differences in speech. Moshman (2013) reasoned that males and females are psychologically rather than categorically different, and only subtle mean differences warrant claims about the differences between the two sexes. Therefore, qualitative conclusions about gender differences may be relative, and genderlect or turn-taking style results should be interpreted cautiously because contextual, situational, and cultural factors could be accountable. In light of the literature review on turn-taking and genderlect, men and women seem pretty different in their speech and how they carry out their speech. However, results about genderlect are inconsistent, and it does not seem easy to generalize speaking styles inherent in men and women. However, several of these findings provide insights into the speaking patterns that are widely found for each sex and thus would be helpful for educational researchers to explore further. Understanding how English as a foreign language (EFL) learners of different sexes take-turn in conversation to manage discourse would help English instructors to decide which interactional resources need to be reinforced for their students and rethink speaking competence assessments.

III. METHODOLOGY

III.1. Participants

The sample is a convenient one. The two Vietnamese university students who studied at the same public university and registered for the peer tutoring program for the English subject were invited to participate in the research. The peer tutoring program, a cooperative learning method, created chances for students with a higher level of English proficiency to help students with a lower level of English abilities to enhance their English skills irrespective of their ages and disciplines. The 19-year-old male was a freshman and needed a tutor; the 22-year-old female was in her fourth year and volunteered to be a tutor. The tutoring sessions were conducted thrice weekly, each lasting two hours. Participants signed the consent form, and pseudonyms were used to protect their identities. The researcher ensured that their personal information and recorded conversation were kept confidential. Table 1 provides the demographic data of the two participants.

Table 1. Participants' Demographic Information.

No	Name	Gender	Age	Native language	Years at university	English proficiency level	Majors	Years of formal English learning
1	Henry	Male	19	Vietnamese	first	IELTS 5.0	Business Adminisration	8
2	Rose	Female	22	Vietnamese	fourth	IELTS 7.0	Business Adminisration	12

Some gender-based literature posited that "gender is more likely to be salient in initial encounters between strangers when people notice gender and have little information to draw on to form expectations about each other" (Aries, 1996, p.182). For those reasons, the participants selected in this research come from the same culture but are strangers to each other. Their conversation was recorded in the first three tutoring sessions. They interacted in both informal and formal dyads, i.e., the interaction between two unknown people under the collaborative floor.

III.2. Data collection method

The two students completed the background questionnaire intended to elicit demographic background information such as sex, age, majors, years of learning English, and self-rated English proficiency. All the exchanges during the first three days of the meeting were recorded with the participants' permission. After that, the recording was transcribed into computer files with their names coded R, representing the female, and H, the male. Speakers participated in three recording sessions, each lasting two hours. Out of those, eight pieces of dialogue between the two students were chosen randomly for detailed analysis. The data were chosen on the grounds of good sound quality. The participants were asked to hold conversations naturally at the time of the recording. They were told that their conversations would be recorded but were assured that details of their conversations would not be disclosed to outsiders. The researcher confirmed to the students that the tapes from their discussions would be destroyed once the final research report was written.

III.3. Data analysis method

The study employed conversation and deductive content analyses. It is essential that the study did not distinguish between formal and informal settings in which the talks are underway. It is because conversation analysts see the talk in interaction as a social process. Also, the turn-taking model can be applied to all conversations regardless of factors such as age, the topic of discussion, type of setting, number, and speaker identity (Liddicoat, 2007). The conversations were transcribed verbatim. The researcher read the transcripts several times to get familiar with the data before coding it in conversation analysis following Gail Jefferson's transcription system (Appendix B) and Liddicoat's (2021) guidelines. The strategy is to make a detailed inspection of tape recordings and transcriptions of participants' conversations. The next step is to use deductive content analysis to identify patterns to see how participants manage their conversations locally, turn by turn. From this, inductive comments about social organization can be made. The analytical procedure was elaborated as follows.

In the first step, transcribing, the audio-taped data were transcribed in as much detail as possible, including the points where interruptions and overlaps began and finished, laughter, and some non-verbal behaviors such as breathing and external noise (e.g., typing something). Conversational gaps were transcribed within and between turns and were timed. No attempt was made to temper the transcripts, for instance, by excluding incomplete utterances or restoring what was said into grammatical forms. However, the transcripts do not include detailed descriptions of body movements (e.g., gaze and gesture) and supra-segmental features (except audible breathing, increased volume, and higher pitch) because their inclusion in the transcripts and their analysis were beyond the scope of this study.

Besides, not any talk can be counted as a turn. For example, uh huh, uhm, and yeah are considered side comments (Edelsky, 1993) or encouragers (Sacks et al., 1978). However, there is some evidence in the transcripts that the current speaker stalls for some time after completing one TCU. Only after receiving some signal from the listener does the interlocutor continue their bits of talk (e.g., turns 85 and 107 in Appendix A). Therefore, in this research, such side comments are seen as a turn to indicate agreement as long as they happen after or just before the current speaker finishes their utterance. It should be noted that deciding whether the speaker uses a particular device or strategy for what purpose is often a matter of interpretation, even with straightforwardly descriptive categories as discussed in the literature. The database consists of 215 turns, selected continuously from eight randomly chosen dialogues extracted from three sets of two-hour recordings for analysis.

The second step relates to labeling and coding. After completing the transcript, the researcher wrote notes, identified strategies that belong to one-at-a-time and overlapping talks, and labeled and coded them, referring to the turn-taking system from the literature review. As discussed in the literature review, a model of turn-taking consists of two main components: turn-allocation (the current speaker selects the next speaker, and the next speaker self-selects) and turn-construction units. Since the current research focuses on dyads, turn-allocation will not be observed because it tends to be applied to multi-speaker exchanges. Therefore, in this study, only turn-construction units were examined.

In the third step, categorizing data, the researcher listed the devices and strategies that participants employed. The researcher then grouped the codes under two broad categories, pure turns (i.e., one-at-a-time talks) and diffused turns (i.e., overlapping speech), to achieve better and closer examination. Next, sub-categories of strategies were accordingly generated and put under suitable headings. The sub-heading numbers to code the devices and specific strategies are employed to mark their occurrences (e.g., 1.1 for Multi-turn TCUs, 1.1.1 for structures, and 1.1.2. for increment). Under these labels, all of the data were accounted for (see Appendix C). The final steps involve recording the numbered turns for each category for later retrieval. Appendix C delineates each participant's main categories, sub-categories, and numbered turns. After obtaining the total number of occurrences of strategy use for each sub- and main category, the researcher compared the results between the two sexes.

IV. FINDINGS AND DISCUSSION

IV.1. Devices and strategies participants use to take turns in one-at-a-time talks during english tutoring sessions

Turn-taking strategies and devices in pure turns include constructing multi-turn TCUs and a multi-TCU turn and reducing transition space (Table 2 in Appendix C). Tables 3 and 4 in Appendix D provide more information regarding the frequencies of devices and strategies used to take turns for both participants. Interestingly, the male student employed more turn-taking devices and strategies in pure turns than the female student. Specifically, to make multi-turn TCUs, he used increments more frequently (e.g., turns 87 and 89 below).

Extract (5)

87 H: =Especially, especially they're from Italy as well.

(.)

88 R: and Yeah=

89 H: =The way they speak English is so hard to hear=

Besides, the male student also used structures to take turns. However, structures were less frequently used than increments and recorded in only one turn (e.g., turn 211 below).

Extract (8)

210 R: =Yeah::, But it has two ways of (.)

211 H: WRIT[ING]

Meanwhile, the female student used only increments to make multi-turn TCUs. However, she used this strategy in only one reported turn (turn 70).

Extract (4)

68 R: =Because some of the topics seem too personal to you,=

69 H:= Yeah.=

70 R:= ( ) a taboo:: Okay?=

Regarding constructing a multi-TCU turn, the male student used audible breathing and deictic devices in the middle of the turn and sped up speech delivery at the end. Among these three choices, speeding up speech delivery (e.g., turns 109, 111, and 115) was the most frequently used, followed by deictic devices (e.g., the use of like as discourse particles or fillers in turns 111 and 113 and anaphoric references such as it and she in turn 111). Interestingly, both participants did not use list beginners and rhetorical questions at the beginning of the turn to ensure an extended turn in the next turn.

Extract (6)

109 H: = hh you know hh, the way I study English like I ( ) that (. ) they think they think the same way same as like ( ) into the mind like the one I read manga they think like they think me I am just ignorant because I am hearing the music but the music I am hearing it's like the (. ) English music,=

110 R: =o uhm o =

111 H: = I hear and I hear the conversation as well I hear twenty-four and twenty-four (.) and the teacher >you know< the guy even the cousin think I am ignorant I think like I am a smart thing I am a smart guy and doesn't need to study. ( ) just- one thing you don't understand English (. ) how can you study things ( ) Even she says ( ) in English I can't understand it I need to ↑study,=

112 R: =ouhmo=

113 H: = and I study in my way like I study (0.25) I'm read the like- I read the manga I read- >because< I can't study things as you guys like read (.) like took the book, like took [the::=

114 R: [hehh[hhh-

115 H:= [> dictionary < and read every word and write down, I don't have patience.=

Surprisingly, the female student employed no strategies or devices to construct a multi-TCU turn and instead tended to control the conversation by only reducing transition space, also used by the male student. This strategy includes repeat, absolute adjacency, creating overlapping talk, and latching. It should be noted that among all strategies to reduce transition space, the male participant used a bit of overlapping talk (e.g., turn 162) and latching (e.g., turns 164 and 166) more often.

Extract (7)

162 R: eh:: that’s what he is , [right?]

162 H: [y e a h] I don't really sure .hh why doing but he must involve into biodroid (.) I think ( ) bi biochemist biochemist support ( ) he do >something about [bio-<

163 R: [Sssso he mentions that he will come back to Vietnam?=

164 H: =Yeah, s (.)the thing [maybe-]

165 R: [hehhhh]hh=

166 H: = his ( ) his uncle don't want to

Meanwhile, the female student used mainly latching strategies to win the turn (e.g., turns 45 and 47 below). Although she used the repeating strategy, this strategy was recorded in one turn (i.e., turn 43). It should be noted that latching was the most frequently used turn-taking strategy for both participants.

Extract (3)

43 R: You put the stress, right? (.) odoo you [put the STRESS?]

44 H: [o( ] )o=Ah, yeah.

45 R: =Ooh=

46 H: =We just write a little bit like write fast a paragraph.=

47 R: = Uh[: : : ]

In short, in one-at-a-time talks, the male student employed various strategies to take turns compared to the female student, who mainly controlled her turn by reducing the transition space. Although she used increments and repeated her talk to win the turn to talk, these strategies were recorded in only one turn. As latching was the most frequently used by both participants, it can be inferred that reducing transition space as a turn-taking strategy was more popular and easier than constructing multi-turn TCUs and multi-TCU turns for both participants.

IV.2. Devices and strategies participants use to repair overlapping talks during english tutoring sessions

In diffused turns, overlap resolution devices and strategies, i.e., hitches, perturbations, and overlap management were found for both participants. Hitches include cutting off the talk, prolonging a segment, and repeating a prior element. Among these hitches strategies, cutting off the talk was the most frequently used by both participants, exemplified by turn 22 for the female student and turn 25 for the male student and as follows.

Extract 1

21 H: [TODAY T O D AY ] Everyone tired,= I don't know=

22 R: = Oh.Yeah. May be [( )-

23 H: [The weather OI think O the weather changing, [OI thinkingO

24 R: [The weather? huh huh huh. [I'm s]till all ri:gh[t huhh.]

25 H: [ehh-] [b u t ] Everyone the same in my school and (.) today (.) Ohe said just like O let everyone do the uh uh uh uh uh involved to talk,=

Also, both participants repeated a prior element, with the male student employing these strategies more often than the female student (e.g., turns 105 for the boy and turn 104 for the girl).

Extract (5)

104 R: the [politics?]

105 H: [political] yeah political polit[ics y[eah polit]ics.

A similar result was found for the use of increased volume or higher pitch to control the turn (e.g., turn 193 for the male student and turn 196 for the female student).

Extract (8)

193 H: Why .hh the guy the American guy doesn't o understand o I forgot. I [SAY=

194 R: [oprogrameeo

195 H: =PROGRAMEE [you say-

196 R: [AH YOU SAY PROGRAMEE , not programming::, ing=

Meanwhile, the female tended to prolong a segment of talk more frequently than the male student (e.g., turn 189 for the female student and turn 190 for the male student) because the female student showed the male student how to pronounce a word.

Extract (8)

188 H: =the last the last word ((he means the last syllable))?( ) go up, right?, [programmee]

189 R: [programming]= NO, progra::mming, [it< i:s second sy]llable,>=

190 H: [< p r o g r a:m >]

Like hitches, the male student also resorted to perturbations more than the female student. Perturbations include increased volume or higher pitch and faster or slower pace of talk. While the first sub-category of perturbation was found for both participants, with the male student (e.g., turn 18) tending to use this strategy more than the female student (e.g., turn 20), holding a faster or slower pace of talk was used by only the female student (i.e., turn 20).

Extract (2)

18 H: [no one do it.] (.) [(eh I) Even I try to, when I going home I try to remember what he give me like >WHICH essay he give me.< Oh whaat=

19 R:= ( ) huh huh huh

(0.5)

20 R: Just someTIMES th[ey’re out of m < i n ]D>.

21 H: [TODAY T O D AY ] Everyone tired,= I don't know=

To solve overlaps, participants also persevered in completing the talk. However, the male student seemed more determined to complete the turns than the female student. Nine instances of using these strategies were found for the male student (e.g., turn 141) compared to six cases for female students (e.g., turn 142) (Table 5 in Appendix D).

Extract (6)

139 H: =She just ask me::: talk more:: and try to:: like like write your own words.=WHEN I write my own words, she thinks I am trying to copy someone.=That's really much [( ), really hard for me] (.) to improving [because-

140 R: [Eh::::::::::::::::::::::::::::::] [justsssss it lacks of (.) understanding and[:: communi ]

141 H: [(like understand) yeah] >communication <. (.) WHEN I THINK like some hard words, I try to: oyao know: I try to: like I >try to< put into my conversation >or may be I put into my writing<, some sometimes of course li:ke she will think like from different way, (. ) because I am just first like first learner like I am a second language, second English (.) eh[::-

142 R:[a second language learner=

143 H: = second language learner,Yeah and this really hard really ↓

To sum up, in one-at-a-time talk, although the male tended to use more devices and strategies than the female in taking turns, both employed latching to reduce transition space most of the time. Moreover, the male tended to push the onset of their utterance to an early start to create a bit of overlapping talk, speed up his delivery, and use increments more frequently than the female. Therefore, this may be interpreted that the male student tended to be more dominant than the female in managing non-overlapping discourse. Similarly, in the overlapping talk, it's interesting to notice that while, in most cases, the male tended to cut off his talk to let the female usurp the turn, he also persevered in completing his talk more often than the female. Furthermore, the male was inclined to increase his pitch and volume or repeat his utterance to win the turn more frequently than the female. These results contrasted with what Holmes and Wilson (2017) reported about women who often used rising intonations in declarations to control their turn-taking. In this study, the male student, instead of the female one, used different paralinguistic features, such as increasing the volume and using a higher pitch more frequently to win the turn in overlapping talks.

Although the literature on genderlect documented that women have a higher frequency than men of using overlapping talk to show rapport (James & Clark, 1993), this study shows that the female participant used varied strategies during simultaneous talks to solve overlaps to revert overlapping talks to one-at-a-time talks. Likewise, in line with Tannen's (1990) report and results from other empirical studies regarding that men interrupted females more significantly (e.g., Bohn & Stutman, 1983; Esposito, 1979, as cited in James & Clark, 1993), the findings in this study show that the male student tended to interrupt the talk by initiating a bit of overlapping speech more often than the female student.

V. IMPLICATIONS

The results from this study indicate that the male participant, the tutee, tended to dominate the conversation in the non-overlapping speech by using various strategies to control his turn. Although the female participant, the tutor, used diverse strategies to solve overlaps, she mainly cut off her talk to let the male student win the turn. These findings have important implications for tutoring and TESOL practitioners.

For the most part, university student tutors and tutees at Vietnamese universities are not required to follow a stipulated syllabus or highly specified program but instead follow the general goal of the tutoring program to help tutees improve their English level, including mastering fundamental knowledge of the English language, passing formative and summative English tests at their faculty, and teaching to tutees' needs. Hence, in most cases, many student tutors play the role of the listener and only answer the tutee's questions if asked and share information if necessary, especially when they are fatigued due to a heavy university workload or that they may think it is not pleasant to interrupt the tutee, which could explain why the tutee from this study tended to dominate the conversation and employed more strategies to control his turns in one-at-a-time talks most of the time. For this reason, before joining the tutoring program, student tutors must be well-trained in various turn-taking and conversational strategies to get the floor, gain and manage time, and keep the floor while thinking. Also, they should be instructed to recognize transition relevance places to intervene in the discussion flow appropriately and to initiate discourse to avoid going off-topic.

Although turn-taking, presented as interactional competence, is one of the aspects of discourse competence in addition to the flexibility to circumstances, thematic development, coherence, and cohesion (Council of Europe Council for Cultural Co-operation Education Committee Modern Languages, 2001), it is rarely included as one of the speaking assessment criteria in formally designed speaking tests at most Vietnamese universities, in some international language tests such as International English Language Testing System (IELTS) and Test of English for International Competence (TOEIC) and localized speaking assessment frameworks (e.g., Vietnamese Standardized Test of English Proficiency) (Truong et al., 2021). Thus, it is time to reconceptualize speaking assessment. Including interactional competence in assessing students’ speaking performance can help speaking assessors ensure the interactive nature and authenticity of paired tasks, identify candidates memorizing prepared notes, and assess their discourse competence.

VI. CONCLUSIONS AND LIMITATIONS

Since conversation is a medium through which people socialize and maintain their relationships, they engage in linguistic interaction, called turn-taking. Turn-taking has been long discussed under the view of sociolinguistics and conversation analysis with mixed results. Also, findings about how Vietnamese students of different sexes managed their discourse are lacking. This research about turn-taking and overlap management strategies among Vietnamese students of different sexes can reveal whether Sacks et al.'s (1978) turn-taking model can be applied to English language teaching. The results show that in one-at-a-time talk, the male student used more devices and strategies than the female in taking turns, and both employed latchings to reduce transition space most of the time. In the overlapping talk, although the female used fewer strategies than the male in most cases, she used more various strategies in simultaneous speech than in one-at-a-time form. These results indicate that male students might dominate the conversation and use more strategies than female students to win the turn to talk. The findings provide useful information for English instructors to be aware of differences in turn-taking and overlap resolution strategies among students of different sexes and provide them with interactional resources to maximize their participation in various assigned groups. Also, the findings can motivate speaking assessors to consider interactional competence as one of the speaking competencies for pairwork assessment.

Because of the nature of the sample, it is impossible to investigate all social variations in the use of discourse features (e.g., dialect, culture, social class, and situation) that may be related to the shape of discourse. Also, due to the limited number of participants, findings may not be applied to all Vietnamese non-English major students. Therefore, it is necessary to have extended samples of speech recorded under similar educational circumstances from individuals belonging to different social categories, such as age, gender, and social class. Future studies can also explore if a relevant correlation exists between each gender's turn-taking strategies with extra-linguistic variants such as age, social status, personality, and the number of pure and diffused turns. Admittedly, interpreting these discrepancies and estimating their significance remains a considerable challenge and requires a more comprehensive investigation. The evidence presented here may provide a suitable foundation to explore these inquiries further.

VII. DECLARATION OF INTERST

The author confirmed that there was no conflict of interest involved.

VIII. FUNDING

This research is funded by Ho Chi Minh University of Food Industry, Ho Chi Minh City, Vietnam

IX. REFERENCES

Archer, D., Aijmer, K., & Wichmann, A. (2013). Pragmatics: An Advanced Resource Book for Students. Routledge.

Bilous, F. R., & Krauss, R. M. (1988). Dominance and accommodation in the conversational behaviours of same- and mixed-gender dyads. Language & Communication, 8(3), 183-194. https://doi.org/10.1016/0271-5309(88)90016-X https://doi.org/10.1016/0271-5309(88)90016-X.

Bowles, H. (2006). Bridging the gap between conversation analysis and ESP–an applied study of the opening sequences of NS and NNS service telephone calls. English for Specific Purposes, 25(3), 332-357. https://doi.org/10.1016/j.esp.2005.03.003.

Council of Europe Council for Cultural Co-operation Education Committee Modern Languages. (2001). Common European Framework of Reference for Languages: Learning, Teaching, Assessment. Cambridge University Press.

Davenport, S., Arnold, M., & Lassmann, M. (2004). The Impact of Cross-Age Tutoring on Reading Attitudes and Reading Achievement. Reading Improvement, 41(1).

Edelsky, C. (1993). Who's got the floor. In D. Tannen (Ed.), Gender and Conversational Interaction (pp. 189-227). Oxford University Press.

Ford, C. E. (1993). Grammar in Interaction: Adverbial Clauses in American English Conversations. Cambridge University Press. https://doi.org/10.1017/s0272263100013358.

Ford, C. E., Fox, B. A., & Thompson, S. A. (2002). The Language of Turn and Sequence. Oxford University Press.

Fujii, Y. (2012). Raising awareness of interactional practices in L2 conversations: Insights from conversation analysis. International Journal of Language Studies, 6(3), 99-126.

Goodwin, C., & Goodwin, M. H. (1992). Assessments and the construction of context. In A. Duranti & C. Goodwin (Eds.), Rethinking Context: Language as an Interactive Phenomenon (pp.147-189). Cambridge University Press.

Hattie, J. (2006). Cross-age tutoring and the reading together program. Studies in Educational Evaluation, 32(2), 100-124. https://doi.org/10.1016/j.stueduc.2006.04.003.

Holmes, J., & Wilson, N. (2017). An Introduction to Sociolinguistics. Routledge.

James, D., & Clark, S. (1993). Women, men, and interruptions: A critical review. In D. Tannen (Ed.), Gender and Conversational Interaction (pp. 231–280). Oxford University Press.

Jefferson, G. (1986). Notes on'latency'in overlap onset. Human Studies, 9, 153–183. https://doi.org/10.1007/BF00148125 .

Kim, Y., & McDonough, K. (2008). The effect of interlocutor proficiency on the collaborative dialogue between Korean as second language learners. Language Teaching Research, 12(2), 211-234. https://doi.org/10.1177/1362168807086288.

Lakoff, R. (1975). Language and woman's place. Language in Society, 2(1), 45-79. https://doi.org/10.1017/S0047404500000051.

Leeser, M. J. (2004). Learner proficiency and focus on form during collaborative dialogue. Language Teaching Research, 8(1), 55-81. https://doi.org/10.1191/1362168804lr134oa.

Lerner, G. H. (2002). Turn-sharing: The Choral Co-Production of Talk-in-Interaction. In B. F. Cecilia Ford (Ed.), The Language of Turn and Sequence

Lerner, G. H. (2004). Conversation Analysis: Studies from the First Generation (Vol. 125). John Benjamins Publishing.

Liddicoat, A. J. (2004). The projectability of turn constructional units and the role of prediction in listening. Discourse Studies, 6(4), 449-469. https://doi.org/10.1177/1461445604046589.

Liddicoat, A. J. (2021). An Introduction to Conversation Analysis. Bloomsbury Publishing. https://doi.org/10.5040/9781350090675.

Long, M. H., & Porter, P. A. (1985). Group work, interlanguage talk, and second language acquisition. TESOL Quarterly, 19(2), 207-228. https://doi.org/10.2307/3586827.

Mori, J., & Zuengler, J. (2008). Conversation analysis and talk-in-interaction in classrooms. In N. H. Hornberger (Ed.), Encyclopaedia of Language and Education (pp. 773-785). Springer. https://doi.org/10.1007/978-0-387-30424-3_58.

Moshman, D. (2013). Adolescent rationality. In R. M. Lerner & J. B. Benson (Eds.), Embodiment and Epigenesis: Theoretical and Methodological Issues in Understanding the Role of Biology within the Relational Developmental System Part B: Ontogenetic Dimensions (Vol. 45, pp. 155–183). Elsevier.

Orasanu, J., Slater, M. K., & Adler, L. L. (1979). Introduction. Annals of the New York Academy of Sciences, 327(1), vii–x. https://doi.org/10.1111/j.1749-6632.1979.tb17746.x.

Saadi Ali, F. (2021). Overlap and repair of turn-taking system during collaborative oral peer feedback in an EFL writing course. International Journal of Social Sciences & Educational Studies, 8(2), 128-134

Sacks, H. (1992). Lectures on conversation: Volume I & II (G. Jefferson & E. A. Schegloff, Eds.). Blackwell Publishing. https://doi.org/10.1002/9781444328301.

Sacks, H. (2004a). An initial characterization of the organization of speaker turn-taking in conversation. In G. H. Lerner (Ed.), Conversation Analysis: Studies from the First Generation (pp. 35-42). John Benjamin Publishing. https://doi.org/10.1075/pbns.125.04sac.

Sacks, H., Schegloff, E. A., & Jefferson, G. (1978). A simplest systematics for the organization of turn-taking for conversation. Language, 50(4), 696-735.

Schegloff, E. A. (2000). Overlapping talk and the organization of turn-taking for conversation. Language in Society, 29(1), 1-63. https://doi.org/10.1017/s0047404500001019.

Schegloff, E. A. (2007). Sequence organization in interaction: Volume 1: A primer in conversation analysis (Vol. 1). Cambridge University Press. https://doi.org/10.1017/CBO9780511791208.

Selting, M. (2000). The construction of units in conversational talk. Language in Society, 29(4), 477-517. https://doi.org/10.1017/S0047404500004012.

Storch, N. (2001). How collaborative is pair work? ESL tertiary students composing in pairs. Language Teaching Research, 5(1), 29-53. https://doi.org/10.1177/136216880100500103.

Storch, N., & Aldosari, A. (2013). Pairing learners in pair work activity. Language Teaching Research, 17(1), 31-48. https://doi.org/10.1177/1362168812457530.

Tannen, D. (1994). Gender and Discourse. Oxford University Press.

Thomas, L., & Wareing, S. (2004). Language, Society and Power: An Introduction. Routledge. https://doi.org/10.4324/9780203426968.

Truong, T. N. N., Samad, A. A., & Phan, T. T. (2021). Perspectives of Test Examiners of the Localized Speaking Assessment Framework: A Case Study in Vietnam. Pertanika Journal of Social Sciences & Humanities, 29 (S3) , 223-242. https://doi.org/10.47836/pjssh.29.s3.12.

Washington-Nortey, P.-M., Zhang, F., Xu, Y., Ruiz, A. B., Chen, C.-C., & Spence, C. (2022). The Impact of Peer Interactions on Language Development Among Preschool English Language Learners: A Systematic Review. Early Childhood Education Journal, 50(1), 49-59. https://doi.org/10.1007/s10643-020-01126-5.

Wiemann, J. M., & Knapp, M. L. (2008). Turn-taking in conversations. In C. D. Mortensen (Ed.), Communication Theory (2nd ed.). Routledge. https://doi.org/10.1111/j.1460-2466.1975.tb00582.x.

Wong, J. (2002). Applying conversation analysis in applied linguistics: Evaluating dialogue in English as a second language textbooks. International Review of Applied Linguistics in Language Teaching, 40(1), 37-60. https://doi.org/10.1515/iral.2002.003.

Yates, S. J. (2001). Gender, language and CMC for education. Learning and Instruction, 11(1), 21-34. https://doi.org/10.1016/s0959-4752(00)00012-8.

Received: 08 March 2023

Accepted: 18 May 2023

APPENDICES

APPENDIX A: TRANSCRIPTS

Link to the transcript:

https://drive.google.com/file/d/1yBY2pPGyZhqBOXfDuGZk8hEBjEumEIAX/view?usp=sharing8.

APPENDIX B: TRANSCRIPTION SYMBOLS

The following transcription symbols, developed by Gail Jefferson and common to conversation analytic research, were used in the data analysis.

→

[	Left square brackets indicate the onset of overlapping or simultaneous, speech by two or more speakers.
]	Right square brackets indicate the point where overlapping speech ends. This may not be marked if it is not analytically important to show where one person's speaking "in the clear" begins or resumes.
(0.4)	Numbers in parentheses indicate a timed pause (within a turn) or gap (between turns) represented in tenths of a second.
:	Colons are used to indicate the prolongation or stretching of the sound just preceding them. The more colons, the longer the stretching.
-	A hyphen after a word or part of a word indicates a cut-off or self-interruption, often done with a glottal or dental stop.
.	A period indicates a falling, or final, intonation contour, not necessarily the end of a sentence.
?	A question mark indicates rising intonation, not necessarily a question.
,	A comma indicates "continuing" intonation, not necessarily a clause boundary.
(())	Matter within double parentheses is a transcriber's comment or description.
=	equal signs within or between turns mark speaking as "latched," with no break or pause, when a speaker makes two grammatical units vocally continuous, or the onset of a next speaker's turn follows the prior speaker's turn immediately without break or pause.
=...=	Two equal signs are used to show the continuation of an utterance from the end of one line to the start of a successive line when overlapping speech comes between the two lines.
word	Underlining is used to indicate some form of contrastive vocal stress or emphasis.
< word	The pre-positioned left carat indicates a hurried start. A common locus of this phenomenon is "self-repair."
WORD	Capital letters are used to indicate markedly higher volume.
ºwordº	The degree sign indicates that the talk following it was markedly quiet or soft. When there are two-degree signs, the talk between them is markedly softer.
↑↓	The up and down arrows occur prior to marked rises or falls in pitch.
><	The stretch of talk between inequality signs in the order "more than" / "less than" indicates that the talk between them is compressed or rushed.
<>	The stretch of talk between inequality signs in the order "less than" / "more than" indicates that the talk between them is markedly slowed or drawn out.
hhh	Hearable aspiration or laugh particles; the more "h"'s, the longer the aspiration. Aspiration or laugh particles within words may appear within parentheses.
hh	Hearable inbreaths are marked with h's prefaced with a dot (or a raised dot).
(word)	Parentheses around all or part of an utterance, or a speaker's identification, indicates transcriber uncertainty, but a likely possibility.

APPENDIX C

Table 2. Turn-taking devices and strategies in pure turns.

Turn-taking devices and strategies in pure terms
Gender's turns	Multi-turn TCUs (1.1.)		Multi-turn TCUs (1.2.)			Reducing transition space (1.3.)
Gender's turns	structures (1.1.1)	Increments (1.1.2)	Audible breathing (1.2.1)	Deictic device (1.2.2)	Speeding up speech delivery (1.2.3)	Repeat (1.3.1)	Absolute adjancency (1.3.2)	A bit of overlapping talk (1.3.3)	Latching (1.3.4)
Male's turns	211	78, 80,87, 89, 127, 181	42	40, 111, 113, 137	18, 111, 115, 125, 128, 133, 141, 162, 169	160, 169	12	7, 34, 50, 85, 150, 162, 167, 177, 186, 212	5, 12, 16, 21, 30, 32, 38, 40, 46, 54, 56, 59, 67, 69, 71, 74, 76, 80, 83, 87, 89, 95, 102, 109, 111, 113, 117, 121, 123, 129, 131, 137, 139, 143, 148, 152, 154, 158, 164, 166, 173, 175, 179, 183, 188, 191, 197, 200, 207, 209, 213
Female's turns		70					43		26, 29, 31, 33, 39, 45, 47, 49, 53, 57, 60, 66, 68, 70, 79, 81, 84, 94, 96, 99, 103, 110, 112, 116, 120, 126, 130, 132, 138, 145, 147, 149, 151, 157, 159, 172, 174, 176, 180, 184,185, 187, 192, 198, 201, 206, 208

Explanations

In pure turn-taking, only one speaker speaks at a time. Two parent categories and four sub-categories were identified, under which specific devices and strategies are listed.

1.1. Multi-turn TCU: structures (if… when) or increment (continuation of a speaker's turn after that speaker has come to what could have been a completion point).

1.2. Multi-TCU turn: at the beginning of the turn (list beginner and "rhetorical question" to ensure an extended turn in the next turn) at the middle (audible breathing and deictic devices), at the end (rush through and speed up speech delivery).

1.3. Reduce transition space: repeat, absolute adjacency pairs, a bit of overlapping talk (put the onset to an early start), and latching.

Table 3. Overlap resolution devices and strategies in diffused turns.

Turn-taking devices and strategies in pure terms
Gender's turns	Hitches (2.1)			Perturbations (2.2)		Managing overlarps(2.3)
Gender's turns	Cutting off the talk (2.1.1)	Prolonging a segment of talk (2.1.2.)	Repeating a prior element (2.1.3)	Intensified volume or higher pitch (2.2.1)	Faster or slower pace of talk (2.2.2.)	Persevering in completing the turn (2.3.1)
Male's turns	2, 25, 35, 82, 117, 139, 141, 146, 150, 162, 164, 195, 202, 203	190	32, 48, 100, 105, 119, 146, 214	18, 20, 25, 100, 119, 133, 193, 203		23, 27, 44, 51, 64, 141, 149, 156, 173
Female's turns	17, 22, 26, 62, 75, 92, 106, 155, 180	36, 140, 189	24, 104, 106, 204, 215	3, 36, 196, 204	20	81, 118, 142, 163, 189, 201

Explanations

In diffused turns, there is more than one speaker at a time, thus unclear talk ensues. Two parent categories and three sub-categories were identified, under which specific devices and strategies are listed.

2.1. Hitches: cutting off the talk by an oral, glottal, or velar stop; prolonging a segment of talk; repeating a just prior element and resources which depart from the prosody of the turn.

2.2. Perturbations: intensified volume, faster or slower pace of talk, and higher pitch.

2.3 Managing overlaps: persevering in completing the turn.

APPENDIX D

Table 4. Number of occurrences of device and strategy use in pure turns.

Turn-taking devices and strategies in pure terms
Gender's turns	Multi-turn TCUs (1.1)		Multi-TCU turn (1.2)			Reducing transition space (1.3)
Gender's turns	Structure (1.1.1)	Increments (1.1.2)	Audible breathing (1.2.1)	Deitic device (1.2.2))	Speeding up speech delivery (1.2.3)	Repeat (1.3.1)	Absolute adjacency (1.3.2)	A bit of overlapping talk (1.3.3)	Latching (1.3.4)
Male	1	6	1	4	9	2	1	10	51
Female	-	1	-	-	-	1	-	-	47

Table 5. Number of occurrences of device and strategy use in diffused turns.

Number of occurrences of device and strategy use in diffused turns
Gender	Hitches (2.1)			Perturbations (2.2)		Managing overlaps (2.3)
Gender	Cutting off the talk (2.1.1)	Prolongng a segment of talk (2.1.2.)	Repeating a prior elelment (2.1.3)	Intensivied volume or higher pitch (2.2.1)	Faster or slower pace of talk (2.2.2.)	Persevering in completing the turn (2.3.1)
Male	14	1	7	8	0	9
Female	9	3	5	4	1	6