Discourse Markers in Academic and Non-Academic Writings of Thai EFL Learners

The ability to use discourse markers (DMs) to create cohesion and coherence of a text is essential for EFL learners at the university level to express ideas and thoughts in various types of writing assignments, such as academic papers and reflections. Hence, this study attempted to shed more light on the use of DMs in academic and non-academic writings of Thai EFL learners. The main objective was to investigate the types, overall frequency, and differences, and similarities of discourse markers in both styles of writing. Sixty essays, consisting of 20 academic essays and 40 non-academic ones, were selected as the primary data. Academic essays were selected from the Critical Reading and Writing course of Xavier Learning Community (XLC), Thailand, while the non-academic ones were selected from the XLC English Newsletter. The data were analyzed based on Fraser’s taxonomy (2009). The results showed that 2.521 DMs distributed in five types, namely contrastive discourse, elaborative discourse, inferential discourse, temporal discourse, and spoken discourse markers, were identified in the 20 academic and 40 non-academic essays. The most frequently used DM was elaborative discourse markers (EDM), F=1,703. This study concluded that raising awareness of DMs would assist Thai EFL learners in producing an effective and coherent piece of writing.


INTRODUCTION
Writing, as a cognitive process that requires a writer to generate ideas and thoughts and position them together into a text (Nunan, 2003), has been commonly recognized by numerous scholars as a demanding task for ESL and EFL learners (Ab Manan & Raslee, 2017;Aidinlou & Mehr, 2012;Arindra & Ardi, 2020;Asprillia & Hardjanto, 2020;Dülger, 2007;Modhish, 2012;Norrish, 1983;Prasetyawati & Ardi, 2020;Sharndama & Yakubu, 2013;Tadayyon & Farahani, 2017) for the reason that they need to "draw upon different grammatical, cognitive, and communicative knowledge" (Tadayyon & Farahani, 2017, p. 133). In line with Prommas and Sinwongsuwat (2011), even English native speakers judge writing as the most problematical language skill among the four because of various essential components, such as word choices, grammatical errors, and overall organization patterns (Dumlao & Wilang, 2019), need to be considered by writers. In writing, most learners face challenges of what and how to write and English grammar (Budjalemba, & Listyani, 2020). Furthermore, what makes writing more perplexing is, as compared to speaking, the lack of "additional means" of nonverbal expressions, such as gesture and facial expression, applied to ensure that the message is accurately understood (Prommas & Sinwongsuwat, 2011, p. 2). Accordingly, teaching EFL learners how to express their ideas in writing is more crucial than only emphasizing the importance of grammar.
It is worth noting that the academic success of EFL learners at the university level is mostly assessed by testing how well they can express themselves in various types of writing assignments, such as academic papers, reports, reflections, and essays (Ab Manan & Raslee, 2017). Therefore, a significant amount of effort has been invested in developing the writing competency of EFL learners since writing is seen as "an essential skill required in almost every activity that EFL learners engage in during their study periods" (Yunus & Haris, 2014, p. 55). Academic and non-academic writings have different styles, and both are fundamental for all EFL learners. Despite the differences, both writing types are equally essential because they serve different and specific purposes (Al-Khazraji, 2019). In academic writing, writers tend to be objective and avoid personal opinions when expressing their thoughts (Hyland & Tse, 2004). In addition, its ultimate goal is to be persuasive, and convincing readers to accept the knowledge revealed. Meanwhile, non-academic writing is more personal, emotional, and subjective. In other words, it is "a kind of individual text in which writers think, feel or believe in something" (Ghanbari et al., 2016(Ghanbari et al., , p. 1451. Often, it is written in an informal tone and not intended for an academic audience. Mature writing, particularly the academic essays, "requires more than just the ability of writers to create sentences appropriately in the standard language, yet they also need an ability to use cohesive devices to create cohesion and coherence of a text" (Prommas & Sinwongsuwat, 2011, p. 2; see also Das & Taboada, 2018;Patriana et al., 2016). Coherence refers to the unity of the ideas and cohesion the unity of structural elements (Chalmers Writing Centre, n.d.). Aidinlou and Mehr (2012) emphasized the significance of cohesive devices by illustrating the relationship between sentences to ensure a text flow. Likewise, Tangkiengsirisin (2010) stated that sentences without coherence are difficult to understand because of the unclear organization of texts. Thus, cohesion and coherence have to be taken into consideration. Conforming to Halliday and Hassan (1976), cohesive devices have been viewed in the literature using different terms, such as linking words, discourse connectors, linking adverbials, cohesive elements, and discourse markers. In this present study, the term discourse markers (DMs henceforth) are used.
DMs are, as reported in Tadayyon and Farahani (2017, p. 133), "linguistics devices" that are applied to connect clauses and sentences together to create an extended discourse. To put simply, DMs are discourse markers, such as 'though', 'since', 'therefore', 'but', and so on that cohere two different segments of discourse together. Also, DMs, seen as functioning words, cannot deliver meaning on their own or change the meaning of a sentence. Furthermore, in agreement with Sharndama and Yakubu (2013), DMs are comparable to the 'glue' that sticks different parts of texts into one well-organized piece of writing carrying pragmatic meaning and cohesiveness. Therefore, it is worth noting that the proper understanding of DMs is considered necessary for good writing production. In other words, an awareness of discourse markers will assist EFL learners in producing an effective and coherent piece of writing (Povolná, 2012).
The study of DMs has been widely conducted over the past 40 years or so as it is believed to have a prominent contribution to the quality of written texts (Feng, 2010;Martínez, 2004). Even though it has been explored for several decades, this area still draws a good amount of attention from numerous scholars in the field of linguistics until now. Recently, Kusumayati (2020) applied Fraser's taxonomy (2009 to examine the use of DMs in expository essays produced by 10 Indonesian EFL learners. She compared the results between two groups (high and low competency of writing skills) and found out that a different level of writing skill influences the use of DMs. Furthermore, another qualitative study by Al-Khazraji (2019) has been conducted to investigate the use of DMs with coherence and cohesion in ESL learners' essay writing in Dubai. The findings indicated three paramount aspects, namely misused DMs, overused DMs, and advanced used DMs. In addition, Al-Owayid (2018) conducted pre and post-test to collect data and particularly analyzed the use of contrastive markers (CDMs) among 100 Saudi females. Through the writing skill instruction, this study concluded that the knowledge of the different meanings of CDMs prominently contributes to the quality of learners' writing.
In the Thai EFL learners' context, Tangkiengsirisin (2010, p. 2) argued that Thai learners, even at the advanced level, have difficulty in "academic writing at the level of text organization and cohesion" with appropriate DMs. Furthermore, the use of DMs is often ignored by both Thai teachers and learners. That is, Thai teachers rarely teach their learners the ways of using DMs correctly in academic and non-academic writings. By the same token, learners also overlook the significance of using DMs while writing. As such, their writing lacks coherence and unity of content. Accordingly, it is worth conducting more studies regarding DMs since there is, to the best knowledge of the researchers, no previous study that has investigated and compared the use of DMs in academic and non-academic writings. However, some studies examined DMs in Thai learners' writing and compared Thai learners' use of DMs to the use of DMs by learners in other countries. Moreover, some studies have investigated the use of DMs in various contexts. For example, Andayani (2014) studied DMs in the argumentative writing of Thai and Indonesian university learners, Prommas and Sinwongsuwat (2011) compared the DMs used by Thai EFL and English native speakers, and Arya (2020) explored the DMs in Thai college learners' conversation. It is urgent to compare DMs in academic and non-academic writings because "learners need to be familiar with discourse markers and how to use them in different writing styles, such as academic and non-academic writings" (Ghanbari et al., 2016(Ghanbari et al., , p. 1452. In sum, DMs in Thai EFL learners' academic and non-academic writings have been scarcely explored.
This study, therefore, attempted to shed more light on the use of DMs in the academic and non-academic writings of Thai EFL learners. This comparison aimed to identify the types, overall frequency, and similarities, and differences of discourse markers in both writing styles. Thus, the following research question was formulated: • What types of discourse markers, including their frequencies, differences, and similarities, were discovered in Thai EFL learners' academic and non-academic writings? It is expected that the results of this study would be a beneficial contribution to English language teaching and learning, particularly related to writing courses, allowing Thai learners to know various DMs and can use them in their writing. Also, teachers would pay closer attention to teaching DMs and raise Thai learners' awareness of discourse markers when writing academic and non-academic essays.

LITERATURE REVIEW
In this section, related terms used in this paper and previous DM studies, such as academic writing, non-academic writing, and discourse markers, together with their subtypes (e.g., CDMs and EDMs), are defined and reviewed.

Academic and Non-Academic Writings
Academic writing is defined as a mode of expression used by researchers or scientists to define the intellectual boundaries of their disciplines and specific areas of expertise (USC Libraries, 2021). This writing style contains a formal tone, the use of the third person rather than the first person (usually), a clear focus on the research problem under investigation, and precise word choice (USC Libraries, 2021). Arnaudet & Barrett (1984, p. 73) state that "in general, academic writing aims at being 'objective' in its expression of ideas, and thus tries to avoid specific reference to personal opinions" (as cited in Hyland, 2002, p. 352). Examples of academic writing are 20 argumentative essays written by Xavier Learning Community (XCL) students participating in this study. Next, non-academic writing is defined as a piece of writing that focuses on a general topic rather than an academic topic and it is intended for a lay audience or the general public, rather than an academic audience (Davis, 2021). Some examples of non-academic writing are Xavier Learning Community (XLC) newsletter which is published periodically carrying general news on XLC events, and students' essays about their memorable personal experiences.

Discourse Markers
The term discourse markers (DMs henceforth) refer to linguistic expressions which carry pragmatic meanings and create coherence and cohesion between sentences and paragraphs (Ismail, 2012). As a matter of fact, DMs have been under investigation with different terms by numerous researchers (Degand & Evers-Vermeul, 2015). Some alternative terms used for DMs are sentence connectives, discourse connectives, discourse word, discourse operators, filler, discourse particles, discourse signaling devices, phatic connectives, pragmatic connectives, pragmatic expressions, pragmatic formatives, pragmatic markers, pragmatic operators, pragmatic particles, and semantic conjuncts (Kusumayati, 2020;Semahat, 2017). Fraser (2009) proposed a taxonomy, which is classified into three functional classes of DMs (see the first three DMs in Table 2). The first class is called contrastive discourse markers (CDMs), such as 'but', 'although', 'however', and 'on the other hand'. In this case, CDMs signal that the "information carried by the discourse segment might establish direct or indirect contrast" (Dumlao & Wilang, 2019, p. 203). For example, 'I want to go shopping in the mall, but it is raining heavily'. The second class is termed as elaborative discourse marker (EDMs), such as 'and', 'besides', 'in addition', 'furthermore', and 'such as'. Basically, this type of DMs indicates that the information in the discourse segments that host them is an elaboration on the information represented by prior segments (e.g., 'My younger brother loves swimming. Besides, he always goes jogging every morning'.). The third class is inferential discourse markers (IDMs), i.e., 'thus', 'therefore', 'because of', and 'so'. IDMs can function as a basis to make inferences based on the preceding sentence or utterance (Fraser, 2009), as in the following: 'She did not study well this semester. Thus, she failed two subjects'.

Studies in Discourse Markers
DMs in academic and non-academic writings have been profoundly explored by Ghanbari et al. (2016). Following Fraser's taxonomy (2009), this study examined the types of DMs found in 60 academic and non-academic papers produced by Iranian EFL learners. Primarily, it aimed to find out how DMs were employed in both types of writing to determine similarities and differences. Similar to the previous study, the findings showed that elaborative and inferential markers were dominantly used in academic writing, whereas only elaborative markers were highly used in non-academic ones. The results were affected by "the nature and style of writing" (Ghanbari et al., 2016(Ghanbari et al., , p. 1456. In academic writing, learners "knew the style of academic writing and the importance of coherence and cohesion in their text, while the non-academic text is a kind of personal text without any special style" (Ghanbari et al., 2016(Ghanbari et al., , p. 1457. This study concluded that academic papers required more DMs. Likewise, Tadayyon and Farahani (2017) studied the use of DMs in the academic writing of Iranians and English native writers. Although the participants were Iranians and native writers of English, Tadayyon and Farahani's (2017) research is relevant to review because it also focused on academic writing. The study selected 30 articles from each group and employed Fraser's taxonomy (2009) to categorize and analyze the data. The results showed that DMs used in Iranian's articles were higher than native learners' articles, and Tadayyon and Farahani (2017, p. 144) concluded that since Iranian learners "are non-native speakers of English, they lack the native-like competency of DMs, which might have resulted in their overuse of DMs in general".
Furthermore, Rahimi (2011) and Ghanbari et al. (2016) argued that studies on DMs can be identified into three main purposes. First, various studies examined the relationship between DMs used and the quality of learners' writing (Fox Tree, 2015). Second, some scholars attempted to find out similarities and differences of DMs in the writing of native and non-native speakers. Third, many researchers studied the use of DMs in different types of writing, such as argumentative, expository, reports, and essays. As this present study mainly focused on written discourse, several empirical studies, which have been conducted to explore learners' writings in the past ten years, will be reviewed. Previously, Dumlao and Wilang (2019) employed Fraser's taxonomy (2009) to investigate the use of DMs in 24 academic essays written by L1 and L2 English learners. Interestingly, this study included the other two markers, namely temporal discourse, and spoken discourse markers, because they display discourse relations of the essays. The results of this study revealed that L1 English learners most frequently used elaborative discourse markers. On the other hand, L2 learners' essays overused some markers, such as temporal and inferential discourse markers, which affected the coherence of their essays. Regarding their findings, Dumlao and Wilang (2019, p. 204-205) concluded that L2 English learners "omitted to give examples or support their ideas to some extent. L1 English users, on the other hand, employed all sets of elaborative markers". Dumlao and Wilang (2019, p. 208) also said "it is possible that the students are not familiar with all types of discourse markers to the same degree, so they only utilize those that they are familiar with".
Some studies have also been conducted to investigate DMs used in other types of writing. For example, Al Gurkosh and Badie (2016) analyzed DMs in descriptive compositions of Iraqi learners by mainly focusing on three types of discourse markers, namely contrastive, elaborative, and inferential. This study affirmed that DMs have a significant impact on the quality of writing since they develop the cohesion and coherence of the writing. Lastly, Andayani (2014) investigated the use of DMs in argumentative essays composed by EFL Indonesian and Thai learners in higher education. This study analyzed 46 argumentative essays by following the framework of Halliday and Hasan (1976) and Halliday and Matthiessen (2004). This study indicated some important problems that influence the use of DMs, such as run-on sentences, missing verbs, grammatical errors, and overuse of DMs.

Participants and Data Collection Procedure
The participants in this study were 20 third-year undergraduates taking the Critical Reading and Writing course and majoring in English at Xavier Learning Community, Chiang Rai, Thailand. The reason for the purposeful selection of the participants was that they had been exposed to several courses involving writing skills such as Basic Structure, Grammar, Writing 1 and 2, and Critical Reading and Writing. The primary obtained data were 60 writing essays, which also functioned as an instrument. As shown in Table 1, 20 academic writings were taken from the learners' argumentative essays, which were the focus of the Critical Reading and Writing 1 course. They were asked to choose one of the following topics, such as (1) Thai government should allow collegians to wear normal clothes in the university, (2) Abolish the death penalty because everyone has the right to live, (3) Stopping burning garbage will lead people to have a healthy life, and (4) Agricultural chemical should be banned in Thailand. On average, the participants wrote around 1,000 words.
Meanwhile, 40 non-academic writings were taken from the XLC Newsletter, issues 12-18. Non-academic writing, including the XLC newsletter, focuses on a general topic rather than any academic topic, and thus, it can be said the newsletter is intended for a lay audience or the mass public (Davis, 2021). The reason why the quantity of non-academic essays was larger than the academic ones was to provide a balance of the total number of words for both writing types for the purpose of comparison (see Table 1).

Data Analysis Technique
This study primarily focused on DMs indicating the relationships among clauses in the learners' essays. Furthermore, to effectively analyze the similarities and differences in EFL learners' academic and non-academic essays, the researchers utilized Fraser's taxonomy (2009), which was explained in Dumlao and Wilang's (2019) study as it was believed to be the most reliable framework (Kusumayati, 2020). Referring to Table 2, Fraser (2009) classified the use of DMs into three categories, namely contrastive discourse markers, elaborative discourse markers, and inferential discourse markers. Nonetheless, the researchers included temporal discourse markers and spoken discourse markers in this study since they also present discourse relations and help create cohesion in the text. Furthermore, the researchers took several steps to analyze the obtained data. First, the researchers thoroughly read the learners' essays to inspect discourses containing DMs. Second, all DMs found in the essays were highlighted, and their frequency was calculated manually. Third, all DMs found were classified in tables based on the five categories. Finally, the findings were carefully described and interpreted based on their categories.

RESULTS AND DISCUSSION
From 20 academic essays and 40 non-academic essays written by Thai EFL learners in this study, five discourse markers (DMs) were identified, namely contrastive discourse, elaborative discourse, inferential discourse, temporal discourse, and spoken discourse markers. The findings can be seen in Figure 1 below.

Figure 1. Types of DMs.
As shown in Figure 1, the findings showed that Thai EFL learners employed all types of DMs in their academic and non-academic essays. Further, there was 2.521 total DMs found in their essays. However, the learners used more DMs in academic essays than non-academic ones. To be more specific, the total frequency of DMs in academic essays was 1.524. On the other hand, only 997 occurrences were found in non-academic essays. Moreover, the first dominant markers in both styles of writing were elaborative discourse markers (F=1,703, 67.55%). For the second and third frequencies, inferential discourse markers and contrastive discourse markers were respectively used with 361 (14.31%) and 324 (12.85%) frequencies. Last but not least, the temporal discourse markers and spoken discourse markers were scarcely used in this study, with a total of 94 (3.72%) and 39 (1.54%), respectively. In addition, it is worth noting that the findings above showed reconfirmation of some other studies (Kusumayati, 2020;Modhish, 2012;Rahimi, 2011;Semahat, 2017;Tadayyon & Farahani, 2017) which discovered that elaborative discourse markers were dominantly employed by their participants. Further discussion and details of each type of DMs employed by Thai EFL learners in their essays will be elaborated more profoundly in the following section.

Contrastive Discourse Markers (CDMs)
The total frequency of CDMs used by Thai EFL learners was 324. However, Table 3 indicated that the learners prominently employed more CDMs in academic essays with 212 occurrences than they did in non-academic essays. Al-Khazraji (2019, p. 561) stated that CDMs are "acknowledged through conjunctions and interpret a clear explanation of the following sentence differences from the preceding one". Hence,

Academic Essays Non-academic Essays
CDMs indicate contrasts between discourse segments in the texts. In this study, the researchers found eight CDMs in academic essays and six CDMs in non-academic essays, namely 'although', 'but', 'even though', 'however', 'instead of', 'on the other hand', 'still', and 'while'. As seen in Table 3, the marker 'but' was highly used in both genres with a slightly different number of frequencies, namely 112 and 95. Moreover, the researchers found out that Thai EFL learners extensively relied on the use of 'but' instead of utilizing a wide range of other CDMs. Mostly, 'but' was used to contrast ideas and sentences and could be positioned either in the beginning and the middle of the sentences, as shown in examples [1] and [2]. More interestingly, the findings of this study also indicated some grammatical errors of Thai EFL learners in using the DMs, punctuations, and tenses. In example [3], instead of writing 'on the other hand', the learner wrote 'in the other hand'. Also, as shown in example [4], the learner incorrectly used a full stop in the dependent clause and forgot to convert the verb 'reduce' to 'reduces' to conform to the subject of the sentence, 'it'. These findings were supported by other studies, such as Syahabuddin and Zikra (2018, p. 72), who discovered that the misuses of CDMs included "incorrect position of certain markers, inappropriate punctuation, and the misunderstanding of the use of markers". In addition, the words 'on the other hand' and 'while' were not used in non-academic essays which can be implied that the learners attempted to keep the essays concise and casual.
[1] They believe that they can save a lot of money from their works in other countries, but it isn't easy for them to get work permits.
[2] But, not all participants at the demonstration support the movement.
[3] In the other hand, wearing a uniform makes learners feel pressured like somebody always controls them.
[4] In the other hands, if we use public transportation. It also reduces air pollution.
In addition to the use of CDMs in academic and non-academic discussed above, it is necessary to point out that the learners avoided using some important CDMs while writing their essays, such as 'despite', 'nonetheless', 'yet', and 'though'. This situation occurred possibly because the learners still lack familiarity with various types of CDMs and need more allocated time outside the class to practice how to use them correctly and properly. This finding is in line with Sitthirak (2013, p. 875), who asserted that "Thai learners found a variety of DMs difficult to distinguish and the apparent interchangeability of the CDMs can also be the cause of confusion". Additionally, the finding that 'but' was the most frequent CDMs used in this study also resonated and supported other studies, which revealed the extreme reliance on the use of 'but' among EFL learners, such as the studies of Povolná (2012) and Dumlao and Wilang (2019).

Elaborative Discourse Markers (EDMs)
In this present study, EDMs were identified as the most frequent DMs type (67.55% of the total DMs) employed by the Thai EFL learners in their academic and non-academic essays, with 1.005 and 698 counts, respectively. Table 4 presents that the learners used fourteen EDMs in both styles of writing. However, it also illustrated that there were three EDMs not found in academic essays, namely 'above all', 'besides', and 'namely', whereas, only 'likewise' was not used in the non-academic essays. Fraser (2009as cited in Dumlao & Wilang, 2019 enlighten that "EDMs indicate that the information contained in the discourse segments that host them is an elaboration on the information represented by prior segments". Fundamentally, these EDMs were used to add more information and explanations to their writing and clarify the arguments by providing some examples. As shown in Table 4, the three most common EDMs in academic essays were 'and', 'or', and 'also'. Conversely, 'and', 'also', and 'such as' were mostly found in the non-academic essays. Moreover, it is worth noting that the marker 'and' was exceedingly overused in both genres compared to the other EDMs. Modhish (2012) denoted that the amount of the marker 'and' is obviously conspicuous in the data because this certain DM is generally used in speaking. Example [5] shows the overuse of the marker 'and'. It is due to the fact that such a marker was unintentionally uttered when the learners were searching for words to convey their ideas in their everyday conversation (Ismail, 2012). Accordingly, it can be inferred that L1 interference constantly influences the excessive use of 'and' in their written essays as well. Besides,examples [5] and [6] revealed that the marker 'and' was mostly placed in the middle of the sentences even though a few sentences used it in the beginning. In addition to the marker 'and', there were some other EDMs used in this study. Similarly, the marker 'or' (example [7] and [8]), 'also' (example [9]), and 'such as' (example [10]) could be positioned either in the initial or middle parts of the sentences.
[5] Wastes can be easily categorized into three types: solid, liquid, and gas. And each type has different methods of disposal and management.
[6] And the next one is the EV charger or a power charger for the car battery that uses electrical energy.
[7] Or they were once very happy, but the marriage didn't last.
[8] She has been not just a teacher or colleague but a role model, and she loves animals.
[9] Also, in the case of doing farming, I suggest that using fertilizer is better to save ourselves and the environment.
[10] Eating a lot of meat can cause diseases such as hypertension, diabetes, and osteoarthritis.
As discussed above, it is worth noting that EDMs were more extensively used than the other four categories by Thai EFL learners. This extreme use of EDMs in academic essays could be in accordance with Tadayyon and Farahani (2017, p. 15), who revealed in their study that it "may be due to the fact that academic writing requires an explanation of ideas and clarity of reasoning, which largely depends on the use of such type of DMs to establish a sense of organization and coherence between different parts of the written discourse". Therefore, this could imply that the nature of argumentative essays requires elaborations and examples to support the author's arguments. Accordingly, it is understandable that academic essays require more EDMs than non-academic ones.

Inferential Discourse Markers (IDMs)
Dumlao and Wilang (2019, p. 204) pointed out that inferential discourse markers (IDMs) were used to "imply significant results in satisfying conversational coherence"; in other words, Thai EFL learners utilized these markers to "establish a causal relationship among clauses and to draw a conclusion". In this study, the total number of IDMs frequency was 361, or 14.31% of the whole DMs. From the data analysis, the researchers found thirteen IDMs used in both academic and non-academic essays, such as 'so', 'of course', 'as a result', 'because of', 'consequently', 'for this (that) reason', 'hence', 'it can be concluded that', 'therefore', 'thus', 'then', 'because', and 'since'. Table 5 depicts all IDMs and their occurrences in detail. Based on the results displayed in Table 5, IDMs were more employed in academic essays by the learners. As IDMs signal that the current utterance conveys a message that is, in a sense, consequential to some aspect of the foregoing, the marker 'because' was dominantly used. In example [11], the learners used the marker 'because' to create a cause-effect relationship in the discourse, and it was positioned in the middle of the sentence. Furthermore, the conspicuous overuse of this marker was found in example [12], where the learner relied heavily on the use of the marker because and did not realize that it could make the sentence wordy. Interestingly, the findings revealed that the marker 'therefore' was barely recognized by the learners in the non-academic essays. Meanwhile, 38 occurrences of this marker were found in the academic essays (see example [15]). However, the researchers also found the overuse of the marker 'so' in the non-academic essays instead of using the other IDMs that serve the same function, such as 'thus', 'hence', 'as a result', and 'consequently', as presented in example [13] and [14]. This situation occurred possibly due to learners' preference to use a more informal IDM and to sound more personal, which is in line with Ghanbari et al.'s (2016) finding. When writing in a non-academic style, learners attempt to be "more personal, emotional, and subjective" (Ghanbari et al., 2016(Ghanbari et al., , p. 1451. Additionally, this marker could be found either in the beginning or the middle of the sentences.
[11] The fact is that there are a large number of crimes because the penalty is not severe enough.
[12] …because undergraduate learners can get jobs because their education is good and complete, and they study because they want to get a good job.
[13] We looked for a boat so we could cross to the Karen State (Myanmar).
[14] So, the key that will lead to the success of this process is to "start" from the beginning.
[15] Therefore, in our system of education, we start learning English from a young age.
Evidently, Table 5 also indicated that several IDMs were underused in the nonacademic essays, such as 'as a result', 'consequently', 'for this (that) reason', 'hence', 'it can be concluded that', and 'therefore'. This finding might be clarified by looking at the characteristics of non-academic writing, as explained by Ghanbari et al. (2016Ghanbari et al. ( , p. 1451. They stated that "non-academic writing is more personal, emotional, and subjective, which shows how writers think, feel, or believe something". Accordingly, the results could imply that Thai EFL learners attempted to keep their writing in an informal tone and avoid applying long markers to keep the stories concise and easily comprehensible for the readers who are not an 'academic or scholarly audience'.

Temporal Discourse Markers (TDMs)
As explained previously, the researchers included temporal discourse markers (TDMs) in this study since the researchers found that these markers also established discourse relations and helped create cohesion in the essays. Dumlao and Wilang (2019) pointed out that TDMs are typically used to indicate the sequences of the text. Based on the data analysis, the researchers found nine TDMs used in the academic and non-academic essays with 94 occurrences (3.72%) in total. Further, Thai EFL learners employed TDMs slightly differently in each type of writing. Table 6 illustrates their occurrences in detail. It is worth mentioning that TDMs was the only category employed more in the non-academic essays than in the academic ones among all the five types of DMs examined in this study. This was possibly due to the genres of texts; in the academic style, the learners were required to argue in their argumentative essays which may need fewer TDMs, whereas, in the non-academic writing, the learners were asked to write personal stories which might involve a chronological order. In Table 6, the marker 'after' was the most frequent TDMs used by the learners. 'After' was used to signal the sequences in the essays. For example, the learner used the marker 'after' at the beginning (example [16]) and in the middle (example [17]) to show the sequences of actions.
[16] After a few weeks working there as a Thai language teacher, I often asked the villagers, "Are you afraid of the Burma Army?" [17] I washed the dishes and cleaned the tables after the customers left.
[18] Before I start to talk about this wonderful opportunity, I would like to send my big thanks to all great people.
[19] Finally, back at XLC, the learners drew up presentations of what they had learned during the seminar.
Likewise, the other two common TDMs found in this study, 'before' and 'finally', were used to indicate the sequences in both academic and non-academic essays. Often, these markers were found at the beginning of the sentences, as shown in examples [18] and [19]. In sum, Ali andMahadin (2016, as cited in Kusumayati, 2020, p. 53) emphasized that "TDMs serve the function of signaling structural relations between the segments that host them and prior discourse segments". Therefore, it could be concluded based on the findings that the nature of the writing of non-academic essays, for instance, writing personal stories involving time sequence, might require more TDMs for the learners to create the flow of events in their stories than that of academic essays, such as argumentative compositions which may require few TDMs.

Spoken Discourse Markers (SDMs)
Ismail (2012) asserted that SDMs could be applied to both written and spoken language. Therefore, the researchers added SDMs in this study. These markers indicated "the closeness between the topic and learners' ideas" (Dumlao & Wilang, 2019, p. 206) and were used to express their attitude and how Thai EFL learners conveyed their thoughts into their academic and non-academic writing. From the data analysis, the researchers found only five SDMs in this study. Hence, some SDMs were not recognized by the learners, such as 'from my aspects', 'from my point of view', 'it is my view', 'let's start', 'Oh', and 'well'. Furthermore, among the five categories of DMs in this study, SDMs were the least used category with only 39 occurrences in 60 essays. These SDMs are used to show the speaker's attitude or viewpoint, particularly when speaking and writing. Thus, such attitudinal SDMs tend to appear more frequently in informal or casual conversations than in (non-)academic written language. Table 7 revealed the three most commonly used SDMs, namely 'I think', 'actually', and 'indeed'. The following are examples of some common SDMs employed in the essays with explanations.
[20] I think it is time for the farmers to change their method and use the fertilizer to protect their family's health and the consumer.
[21] To solve this problem, I think we should manage our time before leaving home.
[22] Actually, English is more important than just learning a lot from textbooks.
[23] It was indeed a very hot day, thus we were glad that this place proposes an air-conditioned room.
From the examples provided above, SDMs were used for different purposes. The marker 'I think' in example [20] was used by the learner to express his or her own opinions towards the issue being discussed. Likewise, positioned in the middle of the sentence, 'I think' in example [21] was used to give the learner's personal solution for the problem. Additionally, example [22] indicated that the learner employed the marker 'actually' to emphasize his or her opinion. Frequently, this marker is found at the beginning or the end of the sentences. However, the researchers did not find this marker used at the end of the sentences in any essays in this study. Lastly, the marker 'indeed' in example [23] was also used to emphasize the statement that what he or she said was definitely true.

CONCLUSION
The results of this present study indicated that 60 essays produced by the Thai EFL learners at the Xavier Learning Community presented the 'superiority' of four categories of DMs in academic writing (F=1.524) over the non-academic writing (F=997), except for TDMs that were employed more frequently in the non-academic essays. To put it differently, a significant difference in the use of DMs between academic and non-academic essays was discovered. Furthermore, the data analysis revealed that EDMs, with 1.703 occurrences, appeared to be the most dominant category, which was extensively used in both types of writing, followed by IDMs, CDMs, TDMs, and SDMs, respectively. Additionally, the findings indicated five most commonly used markers across the essays, namely 'and', 'or', 'because', 'but', and 'also'. Thai EFL learners exceedingly relied on the use of these particular markers due to the lack of familiarity with other variants. The CMD 'but' appeared the most frequently in the academic and non-academic essays, with a total of 112 and 95 times, respectively. Next, the EDM 'and' was the most frequent, with 661 occurrences in the academic essays and 584 instances in the non-academic ones. The findings further showed that the IDM 'because', as the most frequent DM, was used 123 times in the academic essays and 64 in the non-academic ones. The most frequent TDM 'before' occurred nine times in the academic-style essays and the TDM 'after' 27 times in the non-academic style ones. Note that different TDMs occurred the most frequently in each style because of the nature of the texts, namely argumentative essays. The research participants were asked to choose one of the four topics for their essays: (1) Thai government should allow collegians to wear normal clothes in the university, (2) Abolish the death penalty because everyone has the right to live, (3) Stopping burning garbage will lead people to have a healthy life, and (4) Agricultural chemical should be banned in Thailand. Lastly, 'I think', as the most frequent SDM, occurred 15 times in the academic style and five times in the non-academic style.
To diminish the overuse of certain DMs, alternative teaching methods and strategies are required for English teachers to raise learners' awareness of wideranging types of markers in their writing. Also, the teachers should focus more on teaching the functions and purposes of each discourse marker so that the learners will be able to use them in appropriate contexts. Finally, the researchers expected that this study would provide a better understanding of the use of each type of DMs and would enhance the learners' competency in producing effective and coherent academic and non-academic essays.
The limitations of this study are the causes of the overuses, underuses, and misuses of certain DMs both in academic and non-academic genres were not investigated since the main focus of this study was to explore the types of discourse markers, including their frequencies, differences, and similarities, in the academic and non-academic essays produced by Thai EFL learners at the Xavier Learning Community in Chiang Rai, Thailand. The number of participants of this study was also relatively small (only 20 Thai EFL learners), and as a result, no robust generalization could be drawn concerning the DMs used by the Thai EFL learners. Future DM researchers, accordingly, are strongly encouraged to explore more thoroughly the underlying factors of the overuses, underuses, and misuses of DMs by Thai EFL learners involving more respondents representing different Thai higher education institutions. By doing so, it is expected that the findings would be more robust and generalizable.