Şahin Kızıl, A., & Savran, Z. (2018). The integration of corpus into EFL speaking instruction: A study of learner perceptions. International Online Journal of Education and Teaching (IOJET), 5(2), 376-389. http://iojet.org/index.php/IOJET/article/view/368/241 Received: 02.01.2018 Received in revised form: 01.02.2018 Accepted: 19.02.2018 THE INTEGRATION OF CORPUS INTO EFL SPEAKING INSTRUCTION: A STUDY OF LEARNER PERCEPTIONS Aysel Şahin Kızıl Fırat University, Elazığ, Turkey ayselsahin1@gmail.com Zehra Savran Fırat University, Elazığ, Turkey zehrasavran@gmail.com Aysel Şahin Kızıl is currently an Assistant Professor in the Department of English Language Teaching at Fırat University, Turkey. She holds an MA degree on Applied Linguistics and her PhD is on English Language Teaching. Her research interests cover Computer Assisted Language Learning, Second Language Acquisition and Corpus Linguistics in English Language Teaching. She has published and presented scholarly papers dealing with the integration of technology into the language curriculum, corpus applications in language education and student attitudes towards the use of CALL. Zehra Savran is an English language instructor at the School of Foreign Languages, Fırat University, Turkey. She completed her BA in English language teaching and MA in Applied Linguistics. Her research interests include Computer Assisted Language Learning and Corpus Linguistics. Copyright by Informascope. Material published and so copyrighted may not be published elsewhere without the written permission of IOJET. http://iojet.org/index.php/IOJET/article/view/368/241 mailto:ayselsahin1@gmail.com mailto:zehrasavran@gmail.com https://orcid.org/0000-0001-6277-6208 https://orcid.org/0000-0002-3827-7676 Şahin Kızıl & Savran 376 THE INTEGRATION OF CORPUS INTO EFL SPEAKING INSTRUCTION: A STUDY OF LEARNER PERCEPTIONS Aysel Şahin Kızıl ayselsahin1@gmail.com Zehra Savran zehrasavran@gmail.com Abstract Recent years have shown a growing interest in using corpora in language instruction, enhancing data-driven learning (DDL) pedagogy by promoting the use of tools and techniques of corpus linguistics in language classrooms. Many studies have tested the impact of corpus tools in EFL writing or vocabulary instruction; however, little attention has been paid to the integration of corpus tools into EFL speaking instruction. This paper describes a small-scale study of corpus use in enhancing English speaking performance of EFL learners with a focus on their perceptions towards web-based concordancing. Drawn in accordance with convenience sampling procedures, the participants of the study were 31 university level EFL learners who experienced DDL activities in the speaking classroom. Data collected through a post-instruction perception questionnaire were analysed using descriptive statistics. Results indicatethat students benefited from concordance-based learning activities, and also hold positive attitudes towards using it in learning speaking. The findings also point out some challenges to overcome while using web-based concordancing in EFL instruction. Keywords: EFL learners, data-driven learning (DDL), web-based concordancing, learner perceptions, speaking 1. Introduction Since technology has made it possible to compile, store and analyse larger bodies of systematic and computerized collections of written/spoken language data, corpora have been acknowledged to have an impact not only on branches of linguistics but also on the context of foreign/second language teaching. Römer (2011) puts forward that the use of corpora in language teaching contains the use of corpus tools and corpus methods, which leads to a distinction between indirect (e.g. teaching syllabus & teaching materials and direct (teacher- corpus & learner corpus interaction) pedagogical applications. Being a direct application of corpora use, data-driven learning (DDL) approach was developed by Tim Johns in 1991. In this approach, learners are regarded as ‘language detectives’ (Johns, 1997, p. 101) to explore language on their own. Although there has been a considerable progress on integrating corpora into language classroom, Römer (2011) highlights the urge to seek if the learners are willing and able to work with corpus. Therefore, consulting student’s opinions to reveal their attitudes towards corpora could be of vital importance. However, Geluso and Yamaguchi (2014) assert that student beliefs and attitudes towards the use of DDL based methods in language classroom have been mainly researched on written language (Akkoyunlu & Kilimci, 2017; Can, 2009; Kilimci, 2017) but there are relatively fewer of them on spoken language. This study focuses on ‘epistemic stance markers’, a highly common and crucial component of spoken language mailto:ayselsahin1@gmail.com mailto:zehrasavran@gmail.com International Online Journal of Education and Teaching (IOJET) 2018, 5(2), 376-389. 377 which indicates the “degree of commitment to what one is saying” (Kärkkäinen, 2006, p. 705). It is reported in the literature that spoken language is dominated by epistemic devices (Brezina, 2009) which are crucial for learners to achieve a better fluency and foreign- soundingness. Drawing on the relevant literature, the present study aims to introduce the DDL approach into speaking instruction in an EFL context. More specifically, this study aims to find out the perceptions of EFL learners who consulted web-based concordancing through BNCweb (http://corpus.byu.edu/bnc/) in their speaking courses. However, it must be noted that this study is a part of a more comprehensive study and the fact that students show a positive attitude towards use of DDL doesn’t necessarily entail improvement of L2 speaking skills. The following research questions have guided the study:  What are the Turkish EFL learners’ perceptions of the use of concordancing in learning epistemic markers used in spoken English?  What are some difficulties that the Turkish learners have in using concordancing to practice speaking skill?  Are there any differences regarding the perceived difficulties in using corpus between proficiency levels?  What are the Turkish learners’ perceptions of the general use of BNCweb? 2. Literature Review Borrowing the notion of “affordances” from the field of perception psychology, Leńko- Szymańska and Boulton (2015) state that corpora have multiple affordances for language teaching that offer areas of applications not envisaged even by the pioneering corpus compilers. Together with the readily accessible corpora made available through the advances in internet technologies, more and more research has gone into various affordances of corpora in language pedagogy (Chambers, 2005; Leńko-Szymańska & Boulton, 2015). Embodied within DDL approach which involves the direct and indirect application of corpus technology in the classroom to help students explore the target language (Johns & King, 1991), these studies have listed the advantages of using corpora in language teaching as follows (for a comprehensive report, see Cobb and Boulton, (2015)). First, corpora applied within DDL approach are powerful tools for language learners as they enable learners to discover facts about the language through the authentic examples of the target language in the corpus data. As Johns (1991) puts forward on the use of DDL in classroom, learners “often notice things that are unknown not only to the teacher, but also to the standard works of reference on the language” (p. 3). Second, DDL facilitates active involvement of the learner with the learning process as learners are required to explore the language on their own through the corpus-based observations of language, which, in turn, boosts autonomous learning. Bernardini (2002) supports DDL approach by defining corpora as “rich sources of autonomous learning activities of a serendipitous kind” (p. 165). EFL learners, particularly, do not have the opportunities of rich target language input to practice and thereby improve their language learning skills out of class. In addition, in-class experiences of language are very likely to be structured according to teacher’s preference of language. Therefore, learners can study different types of texts of both written and spoken language through corpora that are readily available online (Gabrielatos, 2005). Third, corpora use in the classroom may enhance the learners’ motivation and increase their awareness as they could be able to find the general patterns in language on their own. Ultimately, it has been suggested that data- driven learning overlaps the view of language learning that highlights “guided observation on Şahin Kızıl & Savran 378 the part of the learner rather than exposition on the part of the teacher” (Hunston, 2006, p. 246). However, in spite of the advantages indicated above, the applications of corpus-based research in the instructional settings have remained limited. As Römer (2006) states “despite the progress that has been made in the field of corpus linguistics and language teaching, the practice of ELT has so far been largely unaffected by the advances of corpus research” (p.121). De Cock (2010) argues that this is especially true when it comes to spoken learner corpus research and adds that a lot more research has to be conducted on spoken learner corpora so that spoken learner corpus informed teaching materials can be developed. In the same vein, Cobb and Boulton (2015) underline the necessity of conducting research which integrates corpus techniques into speaking instruction in EFL context and regard this area of study as a gap to be filled by future research. Considering the arguments of this type in the literature, the present study sets out to address this gap by employing DDL in speaking activities for EFL learners. This study explored the use of epistemic markers, which indicate the “degree of certainty or evidence towards the content” as one of the crucial parts of spoken language, (Biber & Finegan, 1988, p. 30). Some highly common epistemic markers in spoken language are I don’t know, I think, maybe, of course, etc. Given that corpora present genuine examples of language, McCarthy (1998) argues that the L2 learners should be exposed to authentic spoken data to be fluent speakers of the target language. In the same vein, Efstathiadi (2010)notes that various types of oral or written practice based on concordance can serve learner needs so as to enable students to better understand the semantic differences of epistemic devices and actual use of language. It is a fact that efficacy of innovative practices in education like DDL approach has much to do with the perceptions of involved parties (Römer, 2011). Although the advantages of corpora have been evaluated from the perspectives of teachers and material developers, the learner perspective has received little attention (Yoon & Hirvela, 2004). Mizumoto, Chujo and Yokota (2016) rightly note that learners’ point of view toward DDL is of great significance especially when they have not had prior experiences with corpus techniques in language learning. Research to date has presented mixed results regarding learners’ perceptions. While some research has reported positive learner attitudes to engaging with corpus based-language learning tasks (Chambers, 2005; Yoon & Hirvela, 2004; Geluso & Yamaguchi, 2014), some other research reported that the learners are likely to lose their interest a while after the corpus-based interventions (Cargill &Adams, 2006; Hafner & Candlin, 2007). However, most of the studies focused on writing (Kennedy & Miceli, 2010; Chambers, 2005; Yoon & Hirvela, 2004), vocabulary or grammar instruction (Boulton, 2009; Yoon, 2008). Although DDL based investigations of writing, grammar or vocabulary present valuable findings, research can also make use of the information on what kind of effects DDL activities have on the learner performance in speaking (Geluso & Yamaguchi, 2014). With this respect, there is a need to conduct studies to further analyse student perceptions towards using corpora to promote the teachers’ and the students’ successful implementation process of corpus-based activities in the classroom (Yoon & Hirvela, 2004). 3. Methodology 3.1. Research Setting and Participants The present study was conducted at the Department of English Language and Literature at a university in Turkey in the fall term of 2016-2017 academic year. The department admits students based on the results of a nation-wide university entrance exam a part of which is International Online Journal of Education and Teaching (IOJET) 2018, 5(2), 376-389. 379 measuring English proficiency of the students. The study was carried out within a speaking course. In accordance with the convenience sampling procedures which “involvechoosing the nearest individuals to serve as respondents” (Cohen, Manion, & Morrison, 2000, p.102), participants of the present study are 31 Turkish EFL learners at their first year. 24 of the students are female and 7 of them are male and their ages range from 18 to 22. Except for one student who is Arabic, 30 out of 31 participants are Turkish and all the participants are native speakers of Turkish. A proficiency test, Oxford Quick Placement Test (2004), was administered at the beginning of the study as there was a time span between the proficiency exam they took before they started studying in their department and the time when the study was conducted.The results showed that 6 students were at elementary level (A2), 25 students were at pre-intermediate level (B1). Furthermore, all participants included in the study reported that they had never been to an English-speaking country before. 3.2. Data Collection Instrument Data was collected through a post-instruction perception questionnaire which was adapted from Yoon and Hirvela (2004), who assessed the instrument for internal reliability and found Cronbach’s alpha value to be r=0.96. Divided into two parts, the first part of the survey aimed at obtaining the personal information of the participants, and the second part asked about the participants’ perception towards three domains on (1) the use of corpus in learning speaking, (2) difficulties of using the corpus and (3) general use of the corpus. For each domain, the participants were asked to indicate their degree of agreement on a 7 point Likert scale where 1 stands for strongly disagree, 7 for strongly agree. To prevent any misconception, the questionnaire was translated into Turkish by the first author and verified by another instructor in the research setting. 3.3. Research Procedures This study aimed to integrate a corpus component to the speaking classroom by teaching the learners how to use concordance to improve their speaking abilities. The focus of the teaching activities were epistemic stance markers which are used to expresses ‘the degree of certainty or evidence towards the content’ of the message (Biber & Finegan, 1988, 30). Epistemic markers are considered to be linguistic items which could boost learners’ fluency in order to get closer to a native-like competency in L2 production (Nesselhauf, 2005). In integrating the corpus into speaking classroom, certain steps were followed. Initially, the significance of the use of epistemic markers in speaking was explained to draw the learners’ attention to the focus of the concordancing activities. Then, the BNC corpus (spoken component only) was introduced to the learners and a step by step use of the corpus techniques (e.g. how to conduct searches and interpret the output) was described in detail through teacher-led demonstrations. During this training session, the learners were asked to explore the BNCweb by themselves. That is, the learners conducted searches in the database on their own and tried to discover how they could make use of concordance lines to practice speaking on their own. Each student was provided with special assistance when they experienced any kinds of difficulty in using the program. Following the training session which took 90 minutes, the learners were given a list of 18 epistemic markers of spoken language which were identified by the researchers on the basis of a contrastive analysis of a native spoken corpus, LOCNEC (De Cock, 2004) and a learner corpus, LINDSEI-TR (Kilimci, 2014) along with identification and comparison of overused and underused epistemic markers in spoken English. However, the results of this contrastive study are beyond the scope of this paper. For a full explanation on the selection of epistemic markers Şahin Kızıl & Savran 380 for corpus-integrated speaking instruction, see Savran (2017). The identified markers were taught to the participants in a total of 12 sessions, each of which lasted 45 minutes, integrating DDL activities through following tasks: (a) searching for the target item in the corpus (b) studying the data and writing down self-selected sample sentences, (c) pair or group discussions on the structural and functional properties of the target items under the guidance of the instructor, (d) finding a general pattern in which the target item occurs, (e) producing a dialogue using the target items to practice. Each session ended with the instructor’s summarizing the use of the target item in spoken interaction. At the end of the treatment, questionnaire explained above was administered to the students. 3.4. Data Analysis Responses to the questionnaire items were analysed using descriptive statistics. For ease of interpretation, responses were coded into three main categories as “agree”, “disagree” and “no opinion” by assigning all positive answers (strongly agree, agree and partly agree) into “agree”, all negative responses (strongly disagree, disagree and partly disagree) into “disagree” and the response ‘no opinion’ into ‘no opinion’ categories. Total percentages were calculated by adding up the percentages of the responses assigned under the related category. 4. Findings and discussion The first 8 questions in the questionnaire were related to using concordance lines for learning epistemic markers in speaking. The results are presented in Table 1. Table 1. Perceptions towards the use of corpus for learning epistemic markers in speaking Agree % Disagree % n.o. % p 1. Using the corpus is helpful for learning the meaning of epistemic markers in speaking 83,9 16,2 0,00 .000 2. Using the corpus is helpful for learning the usage of epistemic markers in speaking 87,1 12,9 0,00 .000 3. Using the corpus is helpful for learning the function of epistemic markers in speaking 86,7 13,3 0,00 .000 4. Using the corpus has improved my understanding of certainty in spoken language 87,1 22,6 3,2 .000 5. Using the corpus has improved my understanding of uncertainty in spoken language 90,3 6,5 3,2 .000 6. I believe that I can express my stance appropriately when speaking after this instruction 87,1 9,7 3,2 .000 7. The use of concordance lists challenged me to actively make generalizations about the function of a marker 87,1 12,9 0,00 .000 8. Concordance was useful for learning the epistemic markers in spoken language. 93,6 6,4 0,00 .000 Walsh (2010) proposes that the language learners encounter a lot of problems in speaking and listening and corpus can be very useful for learners to cope with these problems. International Online Journal of Education and Teaching (IOJET) 2018, 5(2), 376-389. 381 According to the responses of the participants of the study, the implementation of DDL activities in the classroom have potential to help the students learn to express their stance in speech. In addition, majority of the learners stated that the corpus helped them to understand how the certainty and uncertainty is expressed in spoken communication. These findings are in line with the results of the study by Geluso and Yamaguchi (2014) who found out that students reported quite positive attitudes towards the use of concordances for speaking skill. The next domain in the questionnaire was the difficulties the learners had when using concordance lines. This domain included 9 items in total. Table 2 provides the student responses to the items. Table 2. Perceptions on the difficulties on using corpus Agree % Disagree % n.o. % p 9. I have some difficulty in using the corpus due to time and effort spent on analysing the data 26,6 66,7 6,7 .000 10. I have some difficulty in using the corpus due to unfamiliar vocabulary on concordance/collocate output 51,7 45,1 3,2 .000 11. I have some difficulty in using the corpus due to cut-off sentences in concordance output 43,3 50 6,7 .000 12. I have some difficulty in using the corpus due to too many sentences in concordance output 45,1 51,6 3,2 .000 13. I have some difficulty in using the corpus due to the limited number of sentences in concordance output 19,3 74,3 6,5 .000 14. I have some difficulty in analysing concordance output 41,9 54,9 3,2 .000 15. I have some difficulty in analysing output for epistemic markers in speaking 38,8 58 3,2 .000 16. I have some difficulty in performing the search technique 40,1 56,6 3,3 .000 17. The real texts in the corpus are too difficult to understand 33,3 50 16,7 .000 In this part, the learner responses seem to fall in two sides. It is obvious that for most of the items here, while nearly half of the students reported that they had difficulty in using/ analysing concordance output, the other half reported that they did not find it very difficult to search for an item in the corpus. When the responses to the question “I have some difficulty in using the corpus due to time and effort spent on analysing the data” are examined, it is seen that more than half of the students (66,7%) disagreed that the use of corpus was difficult because of time and effort spent on analysing the data. In addition, the learner responses to the question number 13 showed that 74,3% of the students did not agree that the sentences in the concordance output was limited. Şahin Kızıl & Savran 382 The findings from this domain are consistent with previous studies commenting on the student perceptions towards the difficulties of using corpus in learning grammar structures in English. For instance, Girgin (2011) investigated the effectiveness of using corpus-based tools on grammar learning with lower level Turkish EFL learners and suggested that most of the learners had differing, and uncertain opinions about the difficulty of using concordance lines in learning grammar. Additionally, it was found out in the interviews that the learners needed guidance from the teacher to grasp how to analyse concordance output. In order to reveal if there are any differences between experienced difficulties by different proficiency levels for speaking skill, Table 3 presents a clearer picture of the similarities and differences between elementary and pre-intermediate level students. Table 3. Perceptions on the difficulties in using corpus by proficiency levels Category Lower Level (A2) Upper level (B1) Agree % Disagree % n.o % mean s.d. Agree % Disagree % n.o. % mean s.d. 9. Time and effort spent 50 50 0 3,66 1,16 20,9 70,8 8,3 3,00 1,74 10. unfamiliar vocabulary 66,6 33,3 0 3,66 1,36 48 48 4 3,60 1,55 11.cut-off sentences 66,6 33,4 0 3,66 1,50 37,5 54,2 8,3 3,37 1,71 12.too many sentences 50 50 0 3,83 1,47 44 52 4 3,60 1,73 13.limited number of sentences 33,3 50 16, 7 3,83 1,72 16 80 4 2,60 1,35 14. analysing concordance output 33,3 66,7 0 3,33 1,03 44 52 4 3,36 1,52 15. analysing output for epistemic markers 66,7 33,3 0 4,00 1,67 32 64 4 3,24 1,45 16.search technique 80 20 0 4,20 1,48 32 64 4 3,12 1,61 17.too difficult texts 50 50 0 3,83 1,32 29,1 50 20, 9 4,00 2,04 Note: * Lower level group (n=6), Upper level group (n=25) International Online Journal of Education and Teaching (IOJET) 2018, 5(2), 376-389. 383 When the mean scores of student perceptions according to two different proficiency levels are examined, it can be seen that the mean scores centeraround 3,00 – 4,00. Depending on the mean scores of student reactions, it can be claimed that neither lower nor upper level students found corpus very difficult or very easy. Even if the focus of this study is the speaking instruction, relatively similar results are reported by Yoon and Hirvela (2004) who investigated intermediate and advanced level learner reactions to corpus use in writing instruction. However, when the learner reactions to item 15 and 16 are analysed, we observe that lower level learners had more difficulties in analysing the output for epistemic markers in spoken English and in performing the search technique while pre-intermediate level students had fewer problems. Regarding this issue, it could be possible that learning how to use a corpus and trying to focus on learning spoken features of language at the same time stood as a highly demanding task for lower level learners. Concerning the percentages of reactions to the items 10 and 11, it is observed that the differences in experienced difficulties due to unfamiliar vocabulary and cut-off sentences in a spoken corpus resulted from the proficiency levels of learners. Although the mean scores do not display a huge difference between two groups, the percentages show that lower level learners were confronted with more problems in using a spoken corpus in classroom. Considering that the texts in a spoken corpus include pauses, incomplete sentences and a lot of hesitations, it is not surprising that the texts challenged both groups to understand the epistemic markers in general through concordance output as they were not familiar with corpora beforehand. When all these findings are taken into consideration, it is important to note that corpus training stands as an important factor in successful implementation of DDL activities in the classroom. However, the corpus can be helpful to motivate students in terms of speaking English as Walsh (2010) asserts “when we look at a corpus, we find that native speakers also hesitate a lot, are not always coherent, frequently use shorter turns, and may use a fairly narrow range of vocabulary” (p. 336). The last item of this domain showed that half of the elementary and pre-intermediate level students disagreed that the texts in the corpus are too difficult to understand. According to the responses of the learners to this item, it was maintained that the lower level learners could effectively use and benefit from corpus-based activities in the classroom, which provides a contrast to the idea that the corpus-based instructional sources are most useful for learners of English at advanced level (Boulton, 2009). The last domain of the questionnaire was on attitudes towards the general nature of using corpora. Table 4 displays the learner responses to the items in this domain. This part included 13 items in total. Table 4. Perceptions towards the general Use of BNCweb Agree % Disagree % n.o. % p 18. The corpus is more helpful than a dictionary for my English speaking ability 70,9 25,9 3,2 .000 19. The searching technique was easy to learn to use as a reference when I practice speaking 74,2 22,6 3,2 .000 20. I understand the purpose of using the corpus in this treatment 90,3 3,2 6,5 .000 21. When I need to get prepared for a spoken performance, I search for help in the corpus 80 16,7 3,3 .000 22. When I search for information in the corpus, I usually get the information that I 80,6 9,7 9,7 .000 Şahin Kızıl & Savran 384 need 23. I use the corpus when practicing for other courses 41,9 45,2 12,9 .000 24. As I have learned more about the corpus, I have come to like it more 80,7 9,6 9,7 .000 25. I will use the corpus for my English speaking skill in the future 70,8 16,1 12,9 .000 26. Learning about the corpus has increased my confidence about speaking in English 77,4 12,9 9,7 .000 27. If I had used the corpus earlier, I would have had a better performance on speaking 87,1 9,7 3,2 .000 28. Overall, the corpus is a very useful resource for my English speaking 80,7 16,1 3,2 .000 29. The corpus should be introduced in all EFL departments 90,4 6,4 3,2 .000 30. I will recommend the corpus to other students at Fırat University or elsewhere. 80,6 9,7 9,7 .000 When the percentages of learner responses to the general use of corpus in language learning and especially for speaking skill are examined, it is observed that the participants reacted highly positively to the use of concordance lines in the classroom. Interestingly, for the item 23, it looks like students are not decisive about whether they would use corpus for their other courses. While nearly half of the students (41,9%) reported that they would consult corpus for their other courses, the other half stated they wouldn’t use the corpus for practice in other courses and nearly 13% of them stated no opinion. Although they found corpus-based activities useful, they may not have felt confident in how to consult corpus for other courses. However, broadly speaking, it can be argued that the participants adopted quite positive attitudes towards using corpora for language learning purposes. Moreover, in item number 20, a vast majority of them (90%) stated that they understood why the corpus was exploited in this research. It is vital for the language learners to understand the purpose and meaning of the tasks they need to complete. O’Keeffe, McCarthy, and Carter, (2007) highlight the usefulness of corpus-informed materials as in the following words: Successful learning is all about motivation. Corpus-informed materials motivate because teachers and learners can be sure that the language they are practicing is modern, used in everyday situations, targeted to situations they are likely to find themselves in, and corresponds to what they will hear and see in real conversations, movies, radio and TV shows, newspapers, books, Internet texts, and magazines. It is not artificial or invented language, but consists of the most widely used words, phrases, and grammar. (p. 17) Therefore, it can be maintained that especially for speaking skill, which is an area in language that the learners have all types of difficulty, providing students with corpus- informed materials can help them become more efficient in oral communication. All in all, by examining the findings from the intervention and the questionnaire, it is observed that students have benefited from corpus-based learning activities, and also hold pretty positive attitudes towards using corpus in the classroom. However, it should be noted that the language learners need assistance in learning how to exploit corpus not only when learning epistemic stance markers in speaking, but also other forms of language. International Online Journal of Education and Teaching (IOJET) 2018, 5(2), 376-389. 385 5. Concluding Remarks and Pedagogical Implications The questionnaire results for the first domain showed that the learners had quite positive attitudes towards the use of concordance lines in learning the meaning, usage and function of the epistemic stance markers and they reported that the corpus was useful to improve their understanding of the (un)certainty in spoken English. The second domain concerned the difficulties the learners had in using concordance lines. The learner responses reflected somehow conflicting opinions. While approximately one half of the students reported that they had difficulties when using the corpus and analysing the concordance output, the other half stated they didn’t have any trouble. Therefore, it was suggested that the learners were indecisive in whether the corpus was user-friendly or not. It was discussed that learners’ unfamiliarity with the corpora may have challenged them to conduct searches in the corpus and analyse the output. However, lower-level learners are likely to benefit from the DDL activities in the classroom if they are provided with enough training and assistance. The last domain of the questionnaire was related to general use of BNC. The learners reacted positively to the items in this domain. They stated that they would consult corpus to practice speaking in the future, recommend it to their peers, and they would have had a better performance on speaking if they had known about corpora before. Surprisingly, the learners stated conflicting responses to the item questioning if they would consult corpus in order to practice for their other courses. The learners of this study used corpus only for a speaking course and they may have felt they would not be able to figure out how to use it for other courses. This can account for the learners’ conflicting responses. All in all, it was found that the learners’ perception towards using concordance lines to learn epistemic markers in speech was quite positive. Therefore, it can be claimed that the use of corpus-informed materials in the classroom may be a motivating resource for the learners as what they come across in corpus is not an invented language, but rather modern and widely used in real life (McCarhty, 2004). Gabrielatos (2005) suggests that the learners ‘have to be guided away from the “single correct answer” concept and the notion of fixed rules and exceptions, towards the recognition of patterns and alternatives, and the importance of context’ (p.18). Particularly for a spoken corpus, the fact that learners see the hesitations, pauses and a narrow range of vocabulary in native spoken language is very likely to raise their motivation by helping them restructure their language learning perceptions. The findings of the study summarized above have a number of pedagogical implications for foreign or second language teaching. First of all, the study shows DDL activities may have a positive effect on learning language, therefore, consulting corpus-based activities in the classroom can be a motivating and useful resource to learners. Corpus use in the educational settings could stand as a possibility for both the teachers and the students to move away from classroom routine. Second, the study has provided evidence that a corpus- informed approach had a significant impact on learning spoken features of language by lower-level students, which provides a contrast to the idea that the corpus-based instructional sources are most useful for learners of English at advanced level who received a lot of training (Boulton, 2009). Taking into consideration the fact that both the upper and lower- level students hold a very positive attitude towards the use of corpus-based activities in the classroom, it can be argued that this type of activities can motivate the students to confidently participate in speaking activities in the classroom. Regarding the second domain of the scale, it can be suggested that since the EFL learners are not familiar with using corpus-based tools in the classroom, it is very likely that the learners experience difficulties in understanding, using, analysing the concordance lines, which is particularly true for a spoken corpus. Therefore, corpus-based activities should be implemented with the guidance of the teacher Şahin Kızıl & Savran 386 and the learners should be provided with assistance whenever they need it. In this vein, choosing a user-friendly corpus comes as a crucial point to take into account. As the study was conducted only with 31 learners at elementary and pre-intermediate levels of English, further research needs to be conducted with a larger number of learners and with learners from a wider range of proficiency levels from beginner level to advanced level to be able to increase the generalizability of the findings. In addition, future research could examine whether the learners show positive attitudes towards learning another aspect of spoken English. International Online Journal of Education and Teaching (IOJET) 2018, 5(2), 376-389. 387 References Akkoyunlu, A. N., & Kilimci, A. (2017). Application of corpus to translation teaching: Practice and perceptions. International Online Journal of Education and Teaching, 4(4), 369–396. Bernardini, S. (2002) Exploring new directions for discovery learning. In B. Kettemann & G. Marko (Eds.), Teaching and learning by doing corpus analysis (pp. 165-182). Amsterdam: Rodopi. Biber, D., & Finegan, E. (1988). Adverbial stance types in English. Discourse Processes, 11(1), 1–34. Boulton, A. (2009). Testing the limits of data-driven learning: Language proficiency and training. ReCALL, 21(1), 37-51. Brezina, V. (2009). “We only say we are certain when we are not”: A corpus-based study of epistemic stance. In J. Levická & R. Garabík (Eds.), NLP, corpus linguistics, corpus based grammar research (pp. 41–53). Brno: Tribun. Can, C. (2009). İkinci dil edinimi çalışmalarında bilgisayar destekli bir Türk öğrenici İngilizcesi derlemi: ICLE’nin bir altderlemi olarak TICLE. Dil Dergisi, 144(Nisan- Mayıs-Haziran), 16–34. Cargill, M. & Adams, R. (2005). Learning discipline-specific research English for a world stage: A self-access concordancing tool? In Higher education in a changing world. Proceedings HERDSA Sydney 2005. http://conference.herdsa.org.au/2005/pdf/refereed/paper_202.pdf Chambers, A. (2005). Integrating corpus consultation in language studies. Language Learning & Technology, 9(2), 111-125. Cobb, T., & Boulton, A. (2015). Classroom applications of corpus analysis. In D. Biber & R. Reppen (Eds.), The Cambridge handbook of English corpus linguistics (pp. 478-497). Cambridge, UK: Cambridge University Press. Cohen, L., Manion, L., & Morrison K. (2000). Research methods in education (5th Edition). London: Routledge Falmer. De Cock, S. (2010). Spoken learner corpora and EFL teaching. In M. C. Campoy, B. Bellés- Fortuño and M. L. Gea-Valor (Eds.), Corpus-based Approaches to English Language Teaching (pp. 123-137). London: Continuum. De Cock, S. 2004. Preferred sequences of words in NS and NNS speech. Belgian Journal of English Language and Literatures (BELL), New Series 2, 225–246. Efstathiadi, L. (2010). The use of epistemic markers as a means of hedging and boosting in the discourse of L1 and L2 speakers of Modern Greek: A corpus-based study in informal letter-writing. Themes in Science and Technology Education: Special Issue on ICT in Language Learning. Computer-Aided Language Analysis, Teaching and Learning: Αpproaches, Perspectives and Applications, 3(1–2), 181–206. Gabrielatos, C. (2005). Corpora and language teaching: Just a fling or wedding bells? TESL- EJ, 8(4), 1-37. Geluso, J., & Yamaguchi, A. (2014). Discovering formulaic language through data-driven learning: Student attitudes and efficacy. ReCALL, 26(2), 225–242. http://doi.org/10.1017/S0958344014000044 http://conference.herdsa.org.au/2005/pdf/refereed/paper_202.pdf http://doi.org/10.1017/S0958344014000044 Şahin Kızıl & Savran 388 Girgin, U. (2011). Corpus-based activities at lower levels of EFL proficiency: the effectiveness of using concordance lines on grammar learning. Unpublished masters’ thesis, Bilkent University, Ankara, Turkey. Hafner, C. A. & Candlin, C. N (2007). Corpus tools as an affordance to learning in professional legal education. Journal of English for Academic Purposes, 6(4), 303-318. Hunston, S. (2006). Corpus linguistics. Encyclopedia of Language and Linguistics, 234–248. Johns, T., & King, P., (Eds.). (1991). Classroom concordancing. Birmingham: University of Birmingham. Johns, T. (1991). Should you be persuaded: Two samples of data-driven learning materials. ELR Journal, 4, 1–16. Kärkkäinen, E. (2006). Stance taking in conversation: From subjectivity to intersubjectivity. Text and Talk, 26(6), 699–731. http://doi.org/10.1515/TEXT.2006.029 Kennedy, C. and Miceli, T. (2010) Corpus-assisted creative writing: Introducing intermediate Italian learners to a corpus as a reference resource. Language Learning & Technology, 14(1), 28–44. Kilimci, A. (2017). Learner Perspectives Towards Corpus Use in Vocabulary Learning. International Journal of Language Academy, 5(19), 343–359. http://doi.org/10.18033/ijla.3765 Kilimci, A. (2014). LINDSEI-TR: A new spoken corpus of advanced learners of English. International Journal of Social Sciences and Education, 4(2), 401–410. Retrieved from http://www.ijsse.com/sites/default/files/issues/2013/v4i2/Paper-10.pdf Leńko-Szymańska, A., & Boulton, A. (2015). Data-driven learning in language pedagogy. In A. Leńko-Szymańska & A. Boulton (Eds.), Multiple Affordances of Language Corpora for Datadriven Learning (pp. 1–15). Amsterdam: John Benjamins Publishing Company. Leńko-Szymańska, A., & Boulton, A. (2015). Multiple affordances of language corpora for data-driven learning. Amsterdam, Netherlands: John Benjamins. McCarthy, M. (2004). From corpus to course book. Touchstone from corpus to course book. Cambridge: Cambridge University Press. McCarthy, M. J. (1998). Spoken language and applied linguistics. Cambridge: Cambridge University Press. Mizumoto, A., Chujo, K., & Yokota, K. (2016). Development of a scale to measure learners’ perceived preferences and benefits of data-driven learning. ReCALL, 28(2), 227–246. http://doi.org/10.1017/S0958344015000208 Nesselhauf, N. (2005). Corpus linguistics: A practical introduction. Retrieved from http://www.as.uni-heidelberg.de/personen/Nesselhauf. O’Keeffe, A., McCarthy M. J., & Carter R. A. (2007). From corpus to classroom. Cambridge: Cambridge University Press. Römer, U. (2011). Corpus research applications in second language teaching. Annual Review of Applied Linguistics, 31, 205–225. http://doi.org/10.1017/S0267190511000055 Römer, U. (2006). Pedagogical applications of corpora: Some reflections on the current scope and a wish list for future developments. Zeitschrift für Anglistik und Amerikanistik, 54(2), 121-134. http://doi.org/10.1515/TEXT.2006.029 http://doi.org/10.18033/ijla.3765 http://www.ijsse.com/sites/default/files/issues/2013/v4i2/Paper-10.pdf http://doi.org/10.1017/S0958344015000208 http://www.as.uni-heidelberg.de/personen/Nesselhauf http://doi.org/10.1017/S0267190511000055 International Online Journal of Education and Teaching (IOJET) 2018, 5(2), 376-389. 389 Savran, Z. (2017). A corpus-informed study based on a contrastive analysis of stance markers in learner English: from corpus to classroom. Unpublished master’s thesis, Fırat University, Elazığ, Turkey. Walsh, S. (2010). What features of spoken and written corpora can be exploited in creating language teaching materials and syllabuses? In A. O’Keeffe and M. McCarthy (Eds.), The Routledge handbook of corpus linguistics (pp. 333-344). Abingdon: Routledge. Yoon, H. (2008). More than a linguistic reference: The influence of corpus technology on L2 academic writing. Language Learning & Technology, 12(2), 31–48.
 Yoon, H., & Hirvela, A. (2004). ESL student attitudes toward corpus use in L2 writing. Journal of Second Language Writing, 13(4), 257–283. http://doi.org/10.1016/j.jslw.2004.06.002 http://doi.org/10.1016/j.jslw.2004.06.002