Introduction
Discourse markers (DMs) are generally regarded as “linguistic, paralinguistic, or nonverbal elements that signal relations between units of talk” (Schiffrin, 1987, p. 40). Fraser (1999) defined them as lexical expressions that relate discourse segments together and categorized these linguistic features in syntactic classes of conjunctions, adverbs, and prepositional phrases. DMs can have various functions (Grosz & Sidner, 1986; Hummel, 2012; Wang, 2011; Yang, 2011), which vary according to the contexts (Grosz & Sidner, 1986). Different scholars have categorized these lexical phrases based on their various functions. Fung and Carter (2007) developed an inclusive taxonomy in four functional headings: interpersonal, referential, structural, and cognitive DMs. In the interpersonal category, the markers are used to interpret shared knowledge, indicate attitudes, and show responses. In the second category, the markers appear on textual levels marking relationships between the verbal activities which are preceding and following them. The third category contains DMs, which signal links between topics, and markers of the fourth group provide information about the cognitive state of the speaker to help the hearer build a mental representation of the discourse.
Although the use of DMs is often optional (Carter & McCarthy 2015; Fraser 1999; Fung & Carter 2007; Lin 2016; Schiffrin 1987), as linguistic devices, they play an important role in conversational coherence (Christodoulidou, 2011; Das & Taboada, 2018), second language fluency (Neary-Sundquist, 2014), spoken interactions (Carter and McCarthy, 2006), and indicators of the speakers’ attitude (Lin, 2016). Also, their misuse in conversation is reported to lead to semantic or pragmatic misunderstandings (Khandaghi Khameneh & Fakhraee Faruji, 2020). Since forming a coherent discourse is among the crucial factors in English as a Foreign Language (EFL) learners’ communicative competence development (McCarthy & Carter, 1995), there is a pedagogical need for focusing on DM’s acquisition in language teaching programs. Nevertheless, despite their important roles, textbooks still have limited information on them (Namaziandost et al., 2019), and EFL learners are reported to underuse or use a limited number of them in less varied contexts in their speaking (Asic & Cephe, 2013; Götz, 2013).
In recent decades, a large number of studies investigated the use of DMs in various genres. A study by Irfan et al. (2020) investigated them across gender in two novels written by a male and a female writer and concluded that females use DMs more frequently as meaningless fillers. The markers have also been searched in academic writings by Al-khazraji (2019) to examine their use by the learners, and it was concluded that DMs are needed to link sentences and construct knowledge from the known to the unknown. Furthermore, the markers related to the turn transition strategies and pragmatic functions are investigated in chats and instant messaging by Degand and van Bergen (2016), and König (2021), respectively. Regarding the turn transition strategies, Degand and van Bergen (2016) concluded that DMs in utterances' final position can function as turn-transition signaling that the turn is over. König (2021), who investigated two of the markers’ pragmatic functions (HM and EHM), asserted that texters use these markers to state different discourse relations and stances.
This growing body of knowledge investigated multiple factors, such as gender and proficiency, that may affect the use of these linguistic features. However, while many studies are carried out on native speakers, studies investigating L2 learners’ use of DMs remained limited (Tsai & Chu, 2017). Moreover, some existing studies reported conflicting results concerning the above-mentioned factors, which warrant further empirical attention to this important aspect of language use for EFL learners. As an example, while Vanda and Péter (2011) and Fraser (personal communication, 2018) believe that there are no gender differences in DMs’ use, Irfan et al (2020) and Rezaee et al. (2015) found gender differences in using the markers.
The current corpus-based study aimed to contribute to this line of research and examined the use of DMs in second language teachers’ and learners’ speech in the classroom.
Review of the literature
Functions of DMs
According to Tannen et al. (2015), DMs are linguistic elements that function in social, expressive, textual, and cognitive domains. They have an important role in communicative competence, producing meaningful sentences (Rahimi, 2011), and communication (Al Kohlani, 2010). These linguistic features are expressions that connect two parts of discourse, but do not relate to their meaning (Richards & Schmidt, 2011) and are studied under different names such as ‘discourse connectives’ (Blakemore, 1987), ‘pragmatic expressions’ (Erman, 1987), ‘discourse operators’ (Redeker, 1990), ‘continuatives’ (Halliday, 1994), ‘pragmatic markers’ (Fraser, 1996), ‘discourse fillers’ (Navas Brenes, 2005), and ‘discourse markers’ (Tree & Schrock, 1999). Using different terms by different scholars for DMs, and using the same markers differently by different people, Holmes (1990) show their various functions with which they get through discourse (Schiffrin, 2001). As the markers are multifunctional, their proper use could be challenging for the learners (Crible & Cuenca, 2017), which raises a need to make learners aware of their functions (Kaveifard & Allami, 2011).
Schiffrin (1987) defines DMs as ‘sequentially dependent elements which bracket units of talk’ (p. 31). She asserts that DMs appear as contextual coordinates by locating utterances on one or more planes of talk containing; action structure, ideational structure, information state, exchange structure, and participation framework. Further categorization of DMs is presented by Fraser (1990, 1996, 1999). Unlike Schiffrin who considers vocalizations, such as ‘oh’ as DM, Fraser limits them to linguistic expressions and defines them as devices that signal the relationship between the utterances proceeding and foregoing the markers.
Fung and Carter (2007) proposed some criteria for a linguistic item to be considered as a DM. These included flexibility in position, prosodic independence of the utterances, multigrammaticality (e.g., coordinate conjunctions, subordinate conjunctions, prepositional phrases, adverbs), optionality, and indexicality. Indexicality of a DM refers to its functioning as indexical expressions categorized as conceptually rich (I guess), conceptually empty (oh), and partly conceptual (so) meanings. Furthermore, Fung and Carter classified the markers in four functional headings, which is called a multi-category framework in the words of Lin (2016) and is developed based on the data from pedagogic discourse, in which DMs are produced by L2 intermediate and advanced NNSs of English, as well as NSs in natural speaking production in Hong Kong. They examined all the markers produced by the participants.
Table 1: A core functional paradigm of discourse markers in pedagogic discourse (Fung and Carter (2007)
The Role of proficiency in the Use of DMs
Proficiency is a factor that can affect DMs use, which is investigated in some research. Some have compared L2 learners' and NSs' use of the markers and concluded that L2 learners do not use DMs as much as native speakers (NSs) (Fung and Carter, 2007; Hasselgreen, 2004; Muller, 2005; Neary-Sundquist, 2014). In Fung and Carter’s (2007) study, the production of DMs by NSs and L2 learners were investigated by gathering the data in group discussions of 49 participants while doing a task comparing to the NSs’ data, which was gathered from a pedagogic sub-corpus in CANCODE (Cambridge and Nottingham Corpus of Discourse in English). It was concluded that NNSs use a limited range of DMs’ linguistic input.
A study by Neary-Sundquist (2014) investigated the use of DMs by L2 learners at different proficiency levels,comparing the data with native speakers. The results revealed that DMs’ use increases across proficiency levels, and the highest proficiency L2 learners used the markers at the same rate as native speakers, but the functions of DMs used in each proficiency level were not examined.
Another study on the role of proficiency in DMs’ use was done by Huang (2019) who examined the use of DM ‘well’ in Chinese learners, Swedish learners, and NSs. He observed that Swedish learners overuse ‘well’ while Chinese learners underuse it; consequently, and, in line with Gilquin and Granger (2015), he concluded that the learners’ mother tongue may influence DMs’ use. Moreover, in his study, the Chinese learners were investigated across three proficiency levels, revealing that the markers’ use does not vary across proficiency. Therefore, he asserted that more investigations have to be done on the effect of proficiency levels on DMs’ use. Also, he stated that there is a need to examine the factor of gender in this regard as well. It is worth noting that not all of the markers were investigated in his study. Despite much research investigating the role of proficiency in DMs’ use, there is a need for more research to understand its various effects.
The Role of gender in the use of discourse markers
The study of gender differences in DMs has attracted considerable attention. Nevertheless, the findings from different studies result in contradictory findings. Some scholars believe in gender differences in using all or some of the markers. Women use the markers to connect consecutive arguments, while men use them as attention-drawing and signaling repair work (Erman, 1992). Kim and Kang (2011) investigated gender differences in DMs use among 65 people, gathered from Sejong Spoken Corpus. They revealed that women use the markers to gain time or hesitation while men prefer to participate actively in communications. Irfan et al. (2020) examined DMs used in two novels written by a male and a female novelist, asserting that males use fewer markers than females and that females use more DMs as meaningless fillers than men.
Liao (2008) examined the use of nine DMs (yeah, oh, you know, like, well, I mean, ok, right, and actually) in six male and female Chinese graduate students in a study-abroad context in 1422 tokens through discussions and sociolinguistic interviews and stated that gender differences exist in the use of DMs by NNSs. Also, quantitative and qualitative differences across gender in the use of the markers among lecturers were reported by Rezaee et al. (2015) while investigating four DMs (well, OK, you know, and I mean) produced by ten male and ten female Iranian lecturers, considering the functions of the markers. Rezaee et al. (2015) then compared the results with a corpus of English native male and female lecturers. Some quantitative and qualitative differences were found between Iranian male and female lecturers, as well as the English native ones. The females in both groups used the four investigated DMs much more frequently and with wider functions.
On the other hand, some studies revealed that there is no relationship between gender and DMs’ use (Schleef, 2004; Vanda and Péter, 2011) or believed that the gender difference is only seen in teens and has disappeared by age 23 (Jarret, 2014). Some other studies reported no significant gender differences when DMs’ use is examined within a given activity context (Escalera, 2009) or among attitude markers (Nasri et al., 2018). As can be seen, more research has to be done, regarding the role of gender in DMs’ use in various contexts, considering all the markers produced by the learners, while examining the markers’ functions.
The Present Study
Despite much research on discourse markers, there are still some gaps in this area especially across gender and proficiency (Huang, 2019). According to the crucial role of DMs, which was previously mentioned in the literature, a broader corpus-driven cross-linguistic context, more naturalistic data, and different discourse types is needed to have a more complete view on the markers, as Haselow (2019) believes that research on functional categories of the markers should be expanded. Furthermore, there are still some contradictory results regarding the role of gender, proficiency, and learners’ L1 in this regard.
Since Lin (2016) asserted that the existing gap in the literature regarding DMs can be filled by using a multi-categorical model in a pedagogical context this study attempts to fill this gap by investigating all DMs produced by Iranian teachers and learners across their gender and proficiency in a corpus of 22,603 words within the classroom context through a model proposed by Fung and Carter (2007), including multi categories of the markers, and aims to answer the following questions;
- Is there any significant difference between male and female learners in terms of the frequency of using DMs with different functions?
- Is there any significant difference between high and low proficiency learners in terms of the frequency of using DMs with different functions?
- Is there any interaction between learners’ gender and their level of language proficiency in terms of frequency of using DMs with different functions?
- Is there any significant difference between male and female teachers in terms of the frequency of using DMs with different functions?
Method
Participants and setting
There were two groups of participants in this study: learners and teachers. A corpus with a total amount of 22,603 words was constructed from 20 teachers and 40 learners. The participants’ level of proficiency was determined by the Oxford Placement Test (OPT), and all were aged between 20 and 40 years old. Due to the difficulty of motivating subjects attending the study, a non-probability sampling method was used. Participants volunteered to take part in this study, and then, among the participants, stratified sampling was adopted to have equal numbers of males and females.
The group of teachers consisted of ten males and ten females who were compared across gender in using DMs, in which all had high proficiency levels in English (C2 and C1 levels, according to CEFR). They were audio-recorded while teaching. The learners’ group consisted of 20 males and 20 females. In each gender group, there were ten high proficient (C2 and C1 levels, according to CEFR) and ten intermediate level participants (B2 and B1 according to CEFR). The learners were compared across gender, as well as proficiency in using the markers.
Instruments
The book Functions (Matreyek, 1987) was used in the study. As mentioned in the book, it is suitable for advanced and intermediate levels. One of its chapters named Asking about Expressing was used. As Kim & Kang (2011) stated, women respond more emotionally than men. Since this chapter is about expressing feelings, it was chosen to show the gender differences better. The teachers were asked to present this chapter to students and their voices were recorded. Three questions at the end of the chapter were given to learners to be answered while recording their voices. It is worth noting that the learners who participated in this study were not necessarily in the participating teachers’ classes.
A multi-category framework for DMs’ functions presented by Fung and Carter (2007) was used in this study. It was used since it is drawn according to a corpus of nonnative female and male intermediate-advanced English learners with actual language use that is similar to the corpus of this study. They categorized DMs into four main categories: interpersonal, referential, structural, and cognitive.
Procedure
The study was carried out in two stages for each participant. At first, they were asked to take the OPT test. Ten intermediate males and ten intermediate females (CEFR levels B2 and B1), and ten advanced males and ten females (CEFR levels C2 and C1) were then chosen. The learners whose levels of proficiency were lower than intermediate and the teachers who were lower than advanced level were removed. After categorizing the participants, the teachers were asked to teach the chapter Asking about/Expressing from the book Functions to their students. The copies of the chapter were already given to them to be prepared. The researcher got their approval to record their voices during teaching. The recorded teachers’ voices contained monologues and sometimes dialogue among teachers and learners, but the researcher just considered monologues.
To collect samples of learners' speech, they were asked to complete three related speaking tasks presented at the end of that chapter of the book the teachers had taught. The questions were previously given to the learners to be prepared. Then, the learners were asked to answer the three questions. The researcher also got their approval to record their voices.
After recording the voices, these were transferred to the computer to be transcribed. There was a total of 22,603 words. Each voice was listened to three times in order not to miss any words. After counting the total amount of each participant’s words using Microsoft Word, the number of words was written at the end of the transcription following by determining the markers that were highlighted in bold to indicate their distribution. Then, the DMs were categorized based on their functions (Fung and Carter, 2007). Apart from those markers, all the other markers that were produced by the participants and that were not in Fung & Carter’s categorization were also taken into account. Those markers were considered based on having the same meanings or functions as the markers of the Fung & Carter study. Since not all occurrences of the considered words were DM, they were investigated according to the context by listening to the voices several times and also according to the characteristics of DMs that are presented in the literature.
Results
All the data in this study were in the form of simple counts or frequencies in corpora produced by different participants. Since these counts were in productions of different sizes by individuals, it was not possible to simply compare the simple frequencies across different individuals and groups. Therefore, all the frequencies were divided by the total size of each production in terms of total word count to come up with relative frequencies. Then, the relative frequencies were multiplied by 100 to change all the frequencies into a percentage. In the end, these percentages were used in all the analyses as follows.
Investigation of Question 1: Is there any significant difference between male and female learners in terms of frequency of using DMs with different functions?
To answer this question, first, the descriptives of different DMs functions’ percentages in males and females were computed (Table 1). Then, males and females were compared in terms of mean DM percentages by either one-way ANOVA (Parametric test) or Man-Whitney (Non-Parametric test). To choose between these two tests, first, the normality of the data was checked by computing skewness and kurtosis ratios from Table 1 by dividing the skewness and kurtosis values by their standard error. As highlighted in Table 1, the ratios for structural and interpersonal DMs were beyond -+1.96, hence, using the non-parametric test for them. For the rest of the data which are normally distributed sufficiently, an ANOVA as a parametric test was run.
Table 2: Descriptive statistics
Table 3 presents Levene’s test results on the homogeneity of variances as an assumption of ANOVA, which are not significant (p > .05), hence, meeting this assumption.
Table 3: Levene's test of equality of error variancesa
Tables 4 and 5 present the ANOVA and Mann-Whitney test results on the comparison of males and females in terms of DM percentages. As highlighted in these tables, none of the results is significant (p > .05), indicating no difference between males and females in terms of DMs’ use. Therefore, the null hypothesis to this research question is supported. That is to say, there is no significant difference between male and female learners in terms of the frequency of using DMs with different functions.
Table 4: Tests of between-subjects effects
Table 5 Test Statisticsa
Investigating of Question 2: Is there any significant difference between high and low proficiency learners in terms of frequency of using DMs with different functions?
To answer this question, an analysis similar to that of Question 1 was done. First, the descriptives of different DMs functions’ percentages in high- and low-proficiency learners were computed (Table 6). Then, high- and low-proficiency learners were compared in terms of mean DM percentages by either one-way ANOVA (Parametric test) or Man-Whitney (Non-Parametric test). To do so, first, the normality of the data was checked by computing skewness and kurtosis ratios from Table 6, showing that the ratios for structural and interpersonal DMs were beyond -+1.96, hence, using the non-parametric test for them. For the rest of the data that are normally distributed sufficiently, ANOVA as a parametric test was run.
Table 6: Descriptive statistics
Table 7 presents Levene’s test results on the homogeneity of variances as an assumption of ANOVA, which are not significant (p > .05), thus, meeting this assumption.
Table 7: Levene's test of equality of error variancesa
Tables 8 and 9 present the ANOVA and Mann-Whitney test results on the comparison of high- and low-proficiency learners in terms of DM percentages. As highlighted in these tables, high- and low-proficiency learners differ significantly only in terms of interpersonal DMs (p < .05). With regard to the descriptives in Table 6, advanced learners make much higher use of interpersonal DMs than intermediate learners. Therefore, the null hypothesis to this research question is rejected only for interpersonal DMs. That is to say, there is a significant difference between high- and low-proficiency learners in terms of the frequency of using interpersonal DMs.
Table 8: Tests of between-subjects effects
Table 9: Test statisticsa
Investigating Question 3: Is there any interaction between learners’ gender and their level of language proficiency in terms of frequency of using DMs with different functions?
Answering this question required running factorial ANOVA with gender and proficiency level as the factors. Therefore, initially, first the descriptives of DMs across all the levels of the two factors were computed (Table 10). Evidently, some differences exist across the levels of the two factors, but to check the significance of these differences factorial ANOVA results were confirmed in Table 11.
Table 10: Descriptive statistics
According to Table 11, there is a significant interaction between the two factors in terms of interpersonal DMs (p < .05). In order to examine the nature of this significant interaction for interpersonal DMs and even other DMs, the interaction plots for all the DMs were drawn.
Table 11: Tests of between-subjects effects
As the interaction plot for interpersonal DMs shows in Figure 1, males and females were different from each other in terms of interpersonal DM’s used in different levels of language proficiency. Specifically, intermediate male learners used fewer interpersonal DMs than females at an intermediate level, but males used more DMs than females at advanced level. It should be noted that plots for other DMs were also drawn despite the insignificant interaction results in Table 11.
This was done since some insight could be gained through the examination of the plots. Specifically, the plot for cognitive DMs also showed that the two lines crossed each other. This is a sign that if a larger sample size had been employed, some significant interaction could have been found in cognitive DMs. Therefore, replicating this study with a larger sample size specifically in terms of cognitive DMs is highly recommended. In sum, the results above indicate that the null hypothesis to the research question is partly rejected. Therefore, there was an interaction between learners’ gender and their level of language proficiency in terms of the frequency of using interpersonal DMs.
Figure 1: Interaction plots
Investigating Question 4: Is there any significant difference between male and female teachers regarding frequency of using DMs with different functions?
To answer this question, male and female teachers were compared with mean DM percentages by Man-Whitney (Non-Parametric test). As highlighted in Table 12, the skewness and kurtosis ratios for females for all DMs were beyond -+1.96, using the non-parametric test for them.
Table 12: Descriptive statistics
Tables 13 presents the Mann-Whitney test results on the comparison of male and female teachers in terms of DM percentages. As highlighted in these tables, none of the results is significant (p > .05), indicating no difference between male and female teachers in terms of DM’s use. Therefore, the null hypothesis to this research question is supported. Therefore, there was no significant difference between male and female teachers in terms of the frequency of using DMs with different functions.
Table 13: Test statisticsa
Discussion
Investigating discourse and DMs has been receiving growing attention in recent years. The crucial role of DMs in discourse (Liao, 2009) has caused many researchers to study this area each viewing it from different aspects in native and non-native contexts, but none of them have enough coverage of this issue at least in Iranian L2 learners’ contexts. This led the researchers to investigate two factors that may have effects on the use of DMs: gender and the level of proficiency. The main purpose of this study was to see if there are any significant gender or proficiency differences between male vs. female L2 teachers and learners in terms of the frequency of using DMs with different functions.
As stated, no significant differences were found among teachers or learners in terms of frequency of using DMs with different functions, in line with the results of previous research (B. Fraser, personal communication, 3 April, 2018, Escalera, 2006; Koczogh and Furko, 2011; Schleef, 2008; Vanda and Péter, 2011). The findings concerning the second question indicate a significant difference between high- and low-proficiency learners in terms of frequency of using interpersonal DMs; advanced learners made much higher use of interpersonal DMs than intermediate ones. This result is in line with Hasselgreen (2004) who conclude that learners’ proficiency level affects the frequency and range of using DMs. Neary-Sundquist (2014) concluded that DM’s use increased with proficiency level. In this study, the proficiency difference was seen in some specific markers (interpersonal markers) that were found to be used more by highly proficient learners. Interpersonal markers, according to Fung and Carter’s (2007) classifications of the markers, consisted of some markers including ‘you see’ and ‘you know,’ which Romero Trillo (2002) stated were used by NSs more than NNSs. Since Mougeon et al. (2004) suggested highly proficient L2 learners had the same patterns as NS, it can be inferred that this result is in line with Romero Trillo (2002) and Muller (2005). Muller stated NSs and NNSs prefer different DMs and differences were found in the usage of individual functions between them.
The third question is concerned with whether learners’ gender has any significant interaction with their level of language proficiency in terms of frequency of using DMs with different functions. The results indicated a significant interaction between the two factors in terms of interpersonal DMs. Males and females were different in terms of interpersonal DM’s use in different levels of language proficiency. Male learners were also found to use fewer interpersonal DMs than females at the intermediate level, but at the advanced level, males use more DMs than females. As no significant gender difference was found between male and female learners in using DMs which may be due to the focus on different levels of proficiency that is neutralized when considering all levels of proficiency together. The result of this research question is in line with Erman (1992), who stated that men use DMs more often than women, and there was a gender difference in the functions of the markers that each of them used. There are other scholars who believe that gender difference in using DMs differs with different functions that each marker has (Chun, 2008; Croucher, 2004; Kim and Kang, 2011; Mei, 2006; Winkler, 2008).
Conclusion
This project was conducted to develop a description of spoken English. According to the results, the higher use of interpersonal DMs by high proficiency learners may be due to the fact that DMs are not taught explicitly in the classrooms, and the learners have to pick them up themselves in the process of learning their second/foreign language. Due to the important role of the markers, which was presented in the literature, the teachers can teach the markers explicitly even at the starting point of learning English by some awareness-raising tasks on the role of the markers in texts or speech. Hallermann and Vergun (2007) believe that the students who are more acculturated to the US use more DMs. The teachers can help learners be more acculturated by using authentic materials in class, such as using TV shows or radio, for listening or using newspapers or articles as reading materials.
As presented in the literature, DMs are prosodically separate from the utterances they introduce, which meant they have different pauses (Fung & Carter, 2007; Schiffrin, 1987), and this pause sometimes can indicate if a word is DM or not. According to the facilitating role of DMs, it is worth it for learners to consider the markers in the native speech by paying attention to their usage and function in authentic materials.
By comparing DMs, which were produced by high- and low-proficiency learners, this study addresses researchers finding the pattern of learning the markers in the process of L2 learning, regarding the functions of the markers. The other implication of this study addresses researchers who are looking for the reason for the differences which were found in this research. Why do the advanced learners use more interpersonal markers compared to intermediate learners? Why do the highly-proficient students use more referential and cognitive markers rather than the teachers with the same level of proficiency?
This study was limited in terms of sample size. According to the results, replicating this study with a larger sample size, specifically in terms of cognitive DMs among learners considering their gender and the level of proficiency, and also between teachers and highly proficient learners considering the level of proficiency, is highly recommended. Further research can be done on just one group of teachers to appear both as teachers and learners to investigate if the social role can cause any difference in using different DMs with different functions. This study investigated the role of gender and proficiency in using DMs in spoken discourse. The role of DMs in the written form of language can be investigated across gender and proficiency levels for further research. With regards to the obtained results of this research and, according to the increasing number of L2 learners, since discourse and discourse markers play an important role in this area, this study suggests interested teachers, students, and researchers carry out further studies to include more lines of knowledge in this regard.
References
Al-khazraji, A. (2019). Analysis of discourse markers in essays writing in ESL classroom. International Journal of Instruction, 12(2), 559–572. https://doi.org/10.29333/iji.2019.12235a
Al Kohlani, F. A. (2010). The function of discourse markers in Arabic newspaper opinion articles. Georgetown University.
Asik, A., & Cephe, P. T. (2013). Discourse markers and spoken English: Nonnative use in the Turkish EFL setting. English Language Teaching, 6(12), 144–159. https://doi.org/10.5539/elt.v6n12p144
Blakemore, D. (1987). Semantic constraints on relevance. Blackwell.
Carter, R., & McCarthy, M. J. (2006). Cambridge grammar of English: A comprehensive guide to spoken and written grammar and usage. Cambridge University Press.
Christodoulidou, M. (2011). Lexical markers within the university lecture. Novitas-ROYAL, 5(1), 143-160. https://novitasroyal.org/volume-5-issue-1-april-2011-special-issue/#
Chun, C. X. (2008). An investigation of discourse markers across gender in non-English majors’ argumentative writings in English[Unpublished master’s thesis]. Xinjiang Normal University.
Crible, L., & Cuenca, M.-J. (2017). Discourse markers in speech: Distinctive features and corpus annotation. Dialogue & Discourse, 8(2), 149–166. https://doi.org/10.5087/dad.2017.207
Croucher, S. M. (2004). Like, you know, what I’m saying: A study of discourse marker frequency in extemporaneous and impromptu speaking. National Forensic Journal, 22(2), 38–47.
Das, D., & Taboada, M. (2018). RST Signalling corpus: A corpus of signals of coherence relations. Language Resources and Evaluation, 52, 149–184. https://doi.org/10.1007/s10579-017-9383-x
Degand, L., & van Bergen, G. (2016). Discourse markers as turn-transition devices: Evidence from speech and instant messaging.Discourse Processes, 55(1), 47–71. https://doi.org/10.1080/0163853X.2016.1198136
Erman, B. (1987). Pragmatic expressions in English. A study of you know, you see and I mean in face-to-face conversation. Almqvist & Wiksell.
Erman, B. (1992). Female and male usage of pragmatic expressions in same-sex and mixed-sex interaction. Language Variation and Change, 4(2), 217–234. https://doi.org/10.1017/S0954394500000764
Escalera, E. A. (2009). Gender differences in children’s use of discourse markers: Separate worlds or different contexts? Journal of Pragmatics, 41(12), 2479–2495. https://doi.org/10.1016/j.pragma.2006.08.013
Fraser, B. (1990). An approach to discourse markers. Journal of pragmatics, 14(3), 383-398.
Fraser, B. (1996). Pragmatic markers. Pragmatics, 6(2), 167–190. https://doi.org/10.1075/prag.6.2.03fra
Fraser, B. (1999). What are discourse markers? Journal of Pragmatics, 31(7), 931–952. https://doi.org/10.1016/S0378-2166(98)00101-5
Fung, L., & Carter, R. (2007). Discourse markers and spoken English: Native and learner use in pedagogic settings. Applied Linguistics,28(3), 410-439. https://doi.org/10.1093/applin/amm030
Gilquin, G., & Granger, S. (2015). Learner language. In D. Biber & R. Reppen (Eds.), The Cambridge handbook of English corpus linguistics (pp. 418–435). Cambridge University Press.
Götz, S. (2013). Fluency in native and nonnative English speech. John Benjamins.
Grosz, B. J., & Sidner, C. L. (1986). Attention, intentions, and the structure of discourse. Computational Linguistics, 12(3), 175–204. http://nrs.harvard.edu/urn-3:HUL.InstRepos:2579648
Halliday, M. A. K. (1994). An introduction to functional grammar (2nd ed.). Arnold.
Haselow, A. (2019). Discourse marker sequences: Insights into the serial order of communicative tasks in real-time turn production.Journal of Pragmatics, 146, 1-18. https://doi.org/10.1016/j.pragma.2019.04.003
Hasselgreen, A. (2004). Testing the spoken English of young Norwegians: A study of test validity and the role of "smallwords" in contributing to pupil’s fluency. Cambridge University Press.
Holmes, J. (1990). Hedges and boosters in women’s and men’s speech. Language & Communication, 10(3), 185–205. https://doi.org/10.1016/0271-5309(90)90002-S
Huang, L.-F. (2019). A corpus-based exploration of the discourse marker well in spoken interlanguage. Language and Speech, 62(3), 570–593. https://doi.org/10.1177/0023830918798863 PMID: 30232924
Hummel, M. (2012). Polifuncionalidad, polisemia y estrategia retórica: los signos discursivos con base atributiva entre oralidad y escritura; acerca de esp. bueno, claro, total, realmente, etc. [Polyfunctionality, polysemy and rhetorical strategy: discursive signs with an attributive basis between orality and writing; about esp. well, of course, totally, really] De Gruyter.
Irfan, F., Shahzadi, M., Talib, N., & Awan, T. H. (2020). A comparative corpus based analysis of discourse markers for gender description in the alchemist and pride and prejudice. PalArch's Journal of Archaeology of Egypt/Egyptology, 17(11), 358-376.https://archives.palarch.nl/index.php/jae/article/view/5352
Jarret, C. (2014, 2 June). I mean, you know, I’m a conscientious person: links between use of “speech fillers” and personality.: Why should use of phrases such as “like” and “you know” be related to conscientiousness. The Psychologist.https://www.bps.org.uk/research-digest/i-mean-you-know-im-conscientious-person-links-between-use-speech-fillers-and
Kaveifard, E., & Allami, H. (2011). Inferential discourse markers in discussion section of psychology research articles across English and Persian. Theory and Practice in Language Studies, 1(12). https://doi.org/10.4304/tpls.1.12.1786-1791
König, K. (2021). HM and EHM as discourse markers in German WhatsApp chats. Discourse. Context & Media, 39. https://doi.org/10.1016/j.dcm.2020.100457
Khandaghi Khameneh, A., & Fakhraee Faruji, L. (2020). The effect of teaching discourse markers (DMs) on speaking achievement among Iranian intermediate EFL learners. International Journal of Research in English Education, 5(4), 1–13. https://doi.org/10.29252/ijree.5.4.1
Kim, H., & Kang, B. (2011, August). Gender and usage of discourse markers in spoken Korean. In Proceedings of the 16th Conference of Pan-Pacific Association of Applied Linguistics (pp. 30-31). https://paaljapan.org/conference2011/ProcNewest2011/pdf/oral/1B-3.pdf
Liao, S. (2009). Variation in the use of discourse markers by Chinese teaching assistants in the US. Journal of Pragmatics, 41(7), 1313–1328. https://doi.org/10.1016/j.pragma.2008.09.026
Lin, Y.-L. (2016). Discourse marking in spoken intercultural communication between British and Taiwanese adolescent learners. Pragmatics, 26(2), 221–245. https://doi.org/10.1075/prag.26.2.03lin
McCarthy, M., & Carter, R. (1995). Spoken grammar: what is it and how can we teach it?. ELT Journal, 49(3), 207-218. https://doi.org/10.1093/elt/49.3.207
Matreyek, W. A. (1987). Communicating in English: Examples and models (Volume 1: Functions). Prentice-Hall International.
Mei, L. Y. (2006). Research on gender differences in the employment of discourse markers [Unpublished masters thesis], Chengdu University of Technology.
Muller, S. (2005). Discourse markers in native and non-native English discourse. John Benjamins.
Mougeon, R., Rehner, K., & Nadasdi, T. (2004). The learning of spoken French variation by immersion students from Toronto, Canada. Journal of Sociolinguistics, 8(3), 408-432. https://doi.org/10.1111/j.1467-9841.2004.00267.x
Namaziandost, E., Shatalebi, V., & Nasri, M. (2019). The impact of cooperative learning on developing speaking ability and motivation toward learning English. Journal of Language and Education, 5(3), 83–101. https://doi.org/10.17323/jle.2019.9809
Nasri, M., Biria, R., & Karimi, M. (2018). Projecting gender identity in argumentative written discourse. International Journal of Applied Linguistics and English Literature, 7(3), 201–205. https://doi.org/10.7575/aiac.ijalel.v.7n.3p.201
Navas Brenes, C. A. (2011). Analyzing an oral narrative using discourse analysis tools: Observing how spoken language works. Actualidades Investigativas en Educación, 5(1). https://doi.org/10.15517/aie.v5i1.9119
Neary-Sundquist, C. (2014). The use of pragmatic markers across proficiency levels in second language speech. Studies in Second Language Learning and Teaching, 4(4), 637– 663. https://doi.org/10.14746/ssllt.2014.4.4.4
Rahimi, M. (2011). Discourse markers in argumentative and expository writing of Iranian EFL learners. World Journal of English Language, 1(2). https://doi.org/10.5430/wjel.v1n2p68
Redeker, G. (1990). Ideational and pragmatic markers of discourse structure. Journal of Pragmatics, 14(3), 367–381. https://doi.org/10.1016/0378-2166(90)90095-U
Rezaee, M., Aghagolzadeh, F., & Birjandi, P. (2015). The effect of lecturers’ gender on the use of discourse markers. International Journal of Research Studies in Language Learning, 4(2), 69–87. https://doi.org/10.5861/ijrsll.2014.900
Richards, J. C., & Schmidt, R. W. (2011). Longman dictionary of language teaching and applied linguistics. Routledge.
Romero Trillo, J. (2002). The pragmatic fossilization of discourse markers in non-native speakers of English. Journal of Pragmatics, 34(6), 769-784. https://doi.org/10.1016/S0378-2166(02)00022-X
Schiffrin, D. (1987). Discourse markers. Cambridge University Press.
Schiffrin, D. (2001). Discourse markers: Language, meaning and context. In D. Schiffrin, D. Tannen, & H. E. Hamilton (Eds.), Handbook of discourse analysis (pp. 54-75). Blackwell.
Schleef, E. (2004). Gender, power, discipline and context: On the sociolinguistic variation of okay, right, like, and you know in English academic discourse. Texas Linguistic Forum, 48, 177-186.
Schleef, E. (2008). Gender and academic discourse: Global restrictions and local possibilities. Language in Society, 37(4), 515-538.
Tannen, D., Hamilton, H. E., & Schiffrin, D. (2015). The handbook of discourse analysis. Wiley.
Tree, J. E. F., & Schrock, J. C. (1999). Discourse markers in spontaneous speech: Oh what a difference an oh makes. Journal of Memory and Language, 40(2), 280–295. https://doi.org/10.1006/jmla.1998.2613
Tsai, P.-S., & Chu, W.-H. (2017). The use of discourse markers among Mandarin Chinese Teachers, and Chinese as a second language and Chinese as a foreign language learners. Applied Linguistics, 38(5), 638–665. https://doi.org/10.1093/applin/amv057
Vanda, K. H., & Péter, F. B. (2011). Gender differences in the use of the discourse markers you know and I mean. Argumentum, 7, 1–18. https://argumentum.unideb.hu/2011-anyagok/FurkoP_KoczoghH.pdf
Wang, Y. (2011). A discourse-pragmatic functional study of the discourse markers Japanese Ano and Chinese Nage. Intercultural Communication Studies, 20(2), 41–61. https://www-s3-live.kent.edu/s3fs-root/s3fs-public/file/05YanWang.pdf
Winkler, E. G. (2008). A gender-based analysis of discourse markers in Limonese creole. Sargasso: A Journal of Caribbean Literature, Language & Culture, 1, 53-72.
Yang, S. (2011). Investigating discourse markers in pedagogical setting: A literature review. Annual Review of Education, Communication & Language Sciences, 8, 95– 108. https://research.ncl.ac.uk/media/sites/researchwebsites/arecls/yang_vol8.pdf
