WORK WITH ENGLISH CORPORA AS A MEANS OF PROMOTING LEARNER AUTONOMY

 Introduction.Learner autonomy and the ability to determine their own learning paths are becoming the crucial components of graduates’ competitiveness. As Russian students at many higher educational institutions are often brought up in a teacher-oriented learning environment, they are often less adept at autonomous learning. Such skills are especially important nowadays as there tend to be fewer hours of classroom learning. This problem is further compounded by mother tongue interference which is as a rule undesired at advanced levels and also by difficulties of understanding and memorizing foreign language material. Therefore, this paper studies how English corpora and independent work may be used to enhance autonomy.Theaimof the research was to describe the potential of English corpora – alongside other sources – to teach English as a foreign language, and furthermore, to analyse and determine the effects of this innovative approach on learner autonomy and language proficiency.Methodology and research methods.The research method involved at the pre-experiment stage was an extensive literature review of relevant studies. The basis for designing the abovementioned system of teaching techniques incorporated the methodology of corpus linguistics, systematization, analysis and consolidation of the best practices in this field of linguistics. The main tool of the pedagogical experiment whose aim was exploring the potential of corpus-based tasks to promote learner autonomy was the author’s methodology of teaching English to university students.Results and scientific novelty.The author described a number of autonomy forming factors and the pedagogical principles underlying learner autonomy enhancement. A system of corpus-based tasks and activities in conjunction with tasks for intensive and extensive reading and listening for students’ independent work was developed. The designed system was implemented in a five-month experiment which took place at the Institute of Foreign Languages, Ural State Pedagogical University (Yekaterinburg). The research findings based on the pre-experiment and post-experiment data comparison suggest that foreign language teaching is more effective with corpus tasks as they promote learner autonomy. The experimental group (n = 13) became better, i.e. more independent, learners, which indicates their enhanced autonomy, and consequently, a higher level of language proficiency was demonstrated. Thus, corpus-driven language learning coupled with traditional forms of independent work contributes to students’ motivation and sufficient academic progress.Practical significance.The obtained results may provide valuable insights which could be beneficial for foreign language teachers at any educational level. The findings of the study can be used as a framework for further theoretical and empirical research into the effect of the corpus-based approach as well as other forms of independent and classroom learning on enhancing learner autonomy. Another novel contribution of the paper is using the GloWbE corpus that is still relatively unknown in Russia for designing tasks. 


Introduction
Our experience and communication with colleagues from different higher educational institutions show that one of the problems teachers elsewhere in Russia face on a regular basis is limited classroom time. In this time they not only struggle to achieve their teaching objectives, but also try to find effective solutions for the obstacles and difficulties students face outside the classroom. EFL teachers are put in a more vulnerable position as they cannot rely on their students' knowledge of the mother tongue. Moreover, this knowledge often results in L1 interference. Also, the Regulation of the Ministry of Education and Science of the Russian Federation № 1367 [1] must be taken into consideration: according to it there can be up to 25 university students in a group. Thus, the objective of effectively accomplishing specific tasks in the classroom is becoming even more difficult to achieve. It takes effort and time the teacher often cannot afford (for example, in 2015-2017 there were 18-20 language students in our groups, which meant only about 5 minutes of class was allotted to each learner). As a result, teachers can hardly squeeze presentation and practice into a ninety-minute class, and as a rule there is no time left for production. To deal with this problem and at the same time to improve the traditional PPP (Presentation -Practice -Production) structure of a foreign language lesson, the focus must be shifted onto autonomous learning.
The concept "learner autonomy" was brought into focus about 50 years ago and as a result research in this area has a long history and as D. Little formulated, "various sources and wide-ranging implications; thus, it cannot be satisfactorily defined in a few paragraphs" 1 . Learner autonomy is the necessary basis for improving language proficiency, which is understood as students' substantial knowledge, skills related to this knowledge and students' attitude (i.e. the belief that their proficiency in EFL is a valuable resource and willingness to increase the level of language proficiency). This definition to a certain extent correlates with the one in the Common European Framework of Reference for Languages (CEFR): […] 'proficiency' is a term encompassing the ability to perform communicative language activities (can do…), whilst drawing upon both general and communicative language competences (linguistic, sociolinguistic, and pragmatic), and activating appropriate communicative strategies [2, p. 32-33].
Unfortunately, in the framework of conventional teaching students are often made to learn, which turns them into passive receivers of information. The limited experience of independent and self-directed learning impedes students' linguistic progress. Teacher-directed environment does not contribute to enhancing students' capacity to take over responsibility for their own learning. This, according to Henri Holec, is the cornerstone of learner autonomy which means […] to have, and to hold, the responsibility for all the decisions concerning all aspects of this learning, i. e.: • determining the objectives; • defining the contents and progressions; • selecting methods and techniques to be used; • monitoring the procedure of acquisition properly speaking (rhythm, time, place, etc.); • evaluating what has been acquired2. Taking responsibility implies that learners are motivated enough to do so, possess metacognitive skills and have language awareness to identify potential challenges in learning, i.e. autonomy should be conscious, cf. D. Little's opinion in his work.
As learner autonomy can evolve out of an autonomous classroom, L. Legenhausen's language-learning model is of interest. In it, the learner has three roles: • a communicator (using English in authentic situations, learners master communicative skills); • an explorer/experimenter (analyzing English, learners build their language awareness); • an intentional learner (monitoring learning, students develop their learning awareness) [3, p. 37].
Performing these roles students develop language skills and enhance cognitive and metacognitive strategies (strategies of how they learn and strategies about learning). To pursue individual learning, to be able to modify and improve acquired strategies in the course of time, though, one needs intrinsic motivation. The use of computer-aided language learning (CALL) has become one such motivating factor [4]. Corpora (electronic collections of texts created in a certain language and functioning as a reference system) are becoming an important CALL's tool.
On these grounds an experiment was designed and carried out to test the hypothesis that using corpus-based tasks and activities together with extensive, intensive reading and listening compensates for the lack of classroom time, contributes to enhancing learner autonomy, which leads to increasing the level of language proficiency. Supposedly, it is due to the fact that learners are confronted with more real-life English not only in books and/or recordings, but also in English corpora. Exploring the language utilizing corpora tools and methods they gain much more foreign language experience. Quantity is transformed into quality, or improved fluency, accuracy and idiomaticity. It was also found that the corpus-based tasks and activities enabled the learners in the experimental group to work autonomously on a regular basis.

Literature review
In Russia the teacher is associated with control and direct tuition. Unlike this traditional perception, the teacher's role in student-centered pedagogy is that of a facilitator of learning. The growing interest in fostering autonomy in (language) learning explains the increasing literature on the subject [5][6][7][8][9]. As R. Godwin-Jones points out, "Given the emphasis on student-centered pedagogy and on accommodating student diversity, this is not surprising" [10, p. 4].
The detailed review by Römer [8] shows that corpus linguistics and language teaching have been going hand in hand for almost three decades. According to G. R. Bennett [11, p. 2], corpus linguistics is over one century old and contributes to both linguistic research and language teaching and learning. Today it is an area of computational linguistics. Corpora are "large systematic collections of written and/or spoken language stored on a computer and used in linguistic analysis" [8, p. 112]. Spoken language is presented in the form of transcription of speech. All texts are authentic, i. e. naturally occurring. The language phenomena in corpora can only be analyzed with computers, to be more exact, with a concordancing program. By definition offered in the Macmillan dictionary 'a concordancer' means "a list produced by a computer that shows every example of a particular word that is used in the books, newspapers etc stored on the computer". The concordancer yields quantitative results which are then analyzed qualitatively by the researcher. This interdependence of the two analytical techniques provides an opportunity for data-driven and self-directed learning. The corpus approach allows students to change from passive receivers of information into researchers whose cognitive activity is supported and facilitated by teachers. World Web resources provide access to corpora which are an essential tool a teacher can use to arouse learners' interest, involve them in various activities, generally, urge them on. An English corpus is a tool that opens multiple windows to real English. Bernardini [12] sees corpora as a source of serendipitous autonomous learning activities. Working with corpora, students are active exploring the world of English at their own pace, building new knowledge on what they already know, thus being involved in discovery learning. Consequently, all these factors make corpus-based activities interesting and beneficial for learners. What is more, work with corpora can be seen as a great opportunity for students to source their own language learning materials. Logically, corpus-based approach to language teaching (see for example Laviosa's 1 research) is becoming more and more widespread [6,13].
A number of researchers focused on separate language skills: for example, the effect of using English corpora to improve writing ability is discussed by Koyama et al [14], Levchenko [15], Yoon [16]; the studies by T. McEnery & A. Wilson 2 and Zanettin [17] describe the effect of corpora-based tasks on translating skills; some papers discuss corpus-based approach to vocabulary teaching [18][19]. Although corpora have been the focus of attention in numerous studies explaining which of their pedagogical applications are useful in language teaching, they do not describe a multifaceted approach to promoting learner autonomy in the EFL classroom with the help of corpus tools and methods. What is more, there still is an apparent lack of large-scale rese- arch examining the relation between promoting learner autonomy and corpus linguistics methods used in teaching EFL.
English corpora definitely open up new possibilities for the teacher to foster learner autonomy 1 . This key term of student-centered pedagogy can be used in various ways 2 . In summary, the focus of student-centered pedagogy is learners -their needs (what they want to learn or find interesting and beneficial), motivation (the impetus for learning), learning strategies (methods for acquiring knowledge), and language awareness (students' ability to master languages). English corpora provide all of the information and tools necessary for a productive, successful autonomous learning environment. Exploring a corpus means observing the realities of language in use, which is intellectually engaging and triggers learners' interest, cf. L. van Lier's views on actionbased teaching and learning [20]. This process is equal to linguistic research and develops cognitive and metacognitive skills that make students responsible for their own learning process.
Students in Russia often lack certain cognitive and metacognitive skills and strategies necessary for successful autonomous language learning. A. Wenden argues that "Learning strategies are mental steps or operations that learners use to learn a new language and to regulate their efforts to do so" 3 . But as a rule, students have been guided for too long. They are too used to the teacher-centered environment, are passive and not ready to be in charge of their own education. Moreover, many teachers cannot see themselves in the role of a facilitator. As a result, learners are unable to solve problems for themselves, which is an obstacle to successful self-directed learning.
Skills are necessary to implement strategies. V. Cook speaks about the following cognitive strategies: resourcing (learners refer to dictionaries, grammar books, etc.), translation, note-taking, deduction (learners make a judgement about something based on the information they have), inferencing and others 4 . The strategies (logical reasoning, being able to construct rules for solving problems and offer hypotheses, testing the formed hypotheses, etc.) are transferable, i.e. can be implemented in any situation. What is metacognitive knowledge? Wenden suggests that it "includes all facts learners acquire about their own cognitive processes as they are applied and used to gain knowledge and acquire skills in varied situations" 1 . In other words, learners are able to describe how they learn, identify activities essentially important for their learning, reflect on their own achievements, can monitor the progress they have made or are making, assess themselves, direct their learning. Generally speaking, it is the responsibility they take for their studies.
To conclude, it is possible to say that there are enough data to believe that work with corpora contributes to strengthening various skills and develops language learning strategies. At the same time "The practice of ELT to date, at least, seems to be largely unaffected by the advances of corpus research, and comparatively few teachers and learners know about the availability of useful resources and get their hands on corpus computers or concordancers themselves" [9, p. 18].

Materials and methods
The teacher's task is to create and maintain conditions for promoting learner autonomy. The following pedagogical principles underlying the teaching of EFL with the focus on autonomous learning guided the implementation process. The fundamental principle was to enhance pedagogical practice effectiveness via utilizing pedagogical techniques based on the corpus approach. This methodological decision is closely connected with the next core principle, i. e. the need to motivate students to work autonomously and in a systematic way, which correlates with the motivation of students' cognitive activity. Another important principle was making students conscious of their learning process and able to understand their learning tasks. Finally, assessment criteria (a rubric) were developed to provide effective guidelines for the learners to see the direction of learning and for the teacher to assess their progress at the end of the experiment.
We believe that a carefully designed system of corpus-based tasks and activities can complement the extensive, intensive reading and listening activities traditionally used in teaching EFL. In the CEFR the terms 'activity' and 'task' are defined as follows: Language activities involve the exercise of one's communicative language competence in a specific domain in processing (receptively and/or productively) one or more texts in order to carry out a task. […] A task is defined as any purposeful action considered by an individual as necessary in order to achieve a given result in the context of a problem to be solved, an obligation to fulfil or an objective to be achieved [21, p. 10].
The authors see text as the input provided to students to enhance their foreign language proficiency. In this paper corpus data perform the function of text and at the same time form the basis for systematic analysis and discovery learning. It is worth noting that developing such a system ought to start with perfecting learners' cognitive strategies, the main objective being increased metacognitive awareness. Another result will be a higher level of language proficiency. One of the major guidelines in this work is Gavioli and Aston's investigation [22], followed by Braun's research, showing "how corpora as a pedagogical resource can considerably enrich the learning and teaching environment and support autonomous language learning and teaching" [23, p. 1].
The corpora used in this research were the Corpus of Global Web-Based English, the National British Corpus, and the Corpus of Contemporary American English (see [24]). The Corpus of Global Web-Based English (GloW-bE) is the largest one available, constructed only a few years ago, in 2013. As a result, there are very few papers referring to it [25] and consequently most teachers are unlikely to be aware of its existence. Nevertheless, GloWbE ought to become quite popular as it includes 1.9 billion words (1.8 million web pages from 20 different English-speaking countries).
Approximately 60 percent of the corpus comes from informal blogs, and the rest from a wide range of other genres and text types. Because of its large size, its architecture and interface, the corpus can be used to examine many types of variation among dialects, which might not be possible with other corpora -including variation in lexis, morphology, (medium-and low-frequency) syntactic constructions, variation in meaning, as well as discourse and its relationship to culture. [25, p. 1] The paper is based on a literature review and the analysis of the empirical data derived from the study carried out at the Institute of Foreign Languages, Ural State Pedagogical University (USPU) for a period of one semester (5 months) within the first-year English class and was confined to 13 language students (one out of the four groups of first-year students majoring in English). The aim of the study was to enhance learner autonomy and increase the learners' language proficiency using tasks and activities based on English corpora in addition to extensive, intensive reading and listening activities. The novel Black Swan Green by D. Mitchell chosen for additional intensive reading was supplemented by a system of tasks for independent work outside the classroom; for intensive listening 10 authentic recordings from 2 to 30 minutes in length were selected and tasks for them were developed. The materials were to arouse the students' interest and curiosity, to motivate them to work independently. The learners were free to choose any fiction audio books for extensive listening. The students in the other three (control) groups followed the syllabus without doing any corpus-based tasks and activities. They were not engaged in additional intensive reading and listening activities either. Both cohorts did extensive reading (English original fiction texts) and compulsory intensive, or home, reading (The Case of the Perjured Parrot by E. S. Gardner).
The language proficiency of both control and experimental groups was assessed twice -at the end of the first semester (before the experiment) and at the end of the second semester -through an exam consisting of two parts (written and spoken). According to the system of assessment common in the Russian Federation, students got excellent (A/5), good (B/4), satisfactory (C/3), or unsatisfactory (F/2 = failed) grades. The examination was set with the help of the same materials. It is necessary to note that grades A and B were important for our research as they demonstrated language proficiency above average. All the students majored in English. The proficiency levels (the CEFR guideline) varied within both the experimental and the three control groups, ranging from A2 (4-5 students in each group) to C1 (1-2 students in each group). The average proficiency in the four groups was approximately the same at the end of the first semester.
According to a survey conducted at the beginning of the study, the students in the experimental group were unfamiliar with English corpora and had never used them before. Thus, the first step was to introduce the learners to at least one English corpus (as a rule the Corpus of Web-Based Global English was chosen at this stage due to its accessibility and hugeness). After a short introduction to the corpus' concordancer was given, each student registered on corpus.byu.edu and logged in. A non-researcher is allowed 50 queries a day for free, which is usually enough for our purposes. Generally, the learners found their way around the site easily and once starting, they got used to consulting it daily, which is believed to be an excellent way to learn content and language. Such regular work with the corpus did increase students' exposure to real-life English vocabulary, grammatical structures, phraseological units in different contexts. As some authors point out (for example, see [26-27]), content-based instruction results in "increased contextualization for language learning in comparison to traditional grammar-based or communicative language teaching approaches, leading to comparatively greater gains in student language proficiency" [28, p. 3].
At the next stage ten sets of corpus-based tasks and activities were developed; the learners had to do them on a regular basis (once a week for five months). The tasks were confined to either regular homework (short-term inde-pendent work focusing on vocabulary, grammar, micro writing and other skills) and intensive reading and listening (medium-term independent work) or to extensive reading and listening (long-term independent work). The corpus-based tasks and activities generally fall into four broad categories presented below.

Building writing skills
Learners coming to the course often have poor writing (both micro and macro) skills and a vague idea of how to write in English. Work with English corpora comes in handy as it improves learners' writing and strengthens scanning and reading faster skills. Learners have to compile mini-corpora of materials relevant to spelling, punctuation, mechanics in L2. Another option involves reviewing outlines and essays of peers looking collocations and/or colligations and their frequency up in the corpora accessible online.

Strengthening vocabulary
Corpus methods were also used to improve teaching vocabulary. The students had to perform corpus analyses of the corpora texts adding more frequent collocations to the list, so teacher-corpus interaction affected the content of teaching materials and led to learner-corpus interaction (see [8, p. 113-114]). Cf. the example task: Study the collocations and add more nouns collocating with the verbs.
• peel (eggs, potatoes, …) • roast (duck, potatoes, …) • roll out (pastry, marzipan, …) Using corpora exposes students and teachers to real, live English which is often different from what is described in dictionaries and textbooks, especially if they are not authentic. As an option, compiling concordances with vocabulary items "most relevant and useful to language learners" followed. The next step can be the compilation of corpus-based collocations dictionaries or glossaries. By doing such tasks, learners get invaluable experience in dictionary work and lexicography.

Integrating teaching methodology and learning methodology (developing professional competence of future teachers)
The students were asked to develop a fragment of a lesson plan for a class using authentic language samples from the abovementioned English corpora. It is worth mentioning that GloWbE has "the option of re-contextualizing", i. e. learners can see full contexts or go to the website to read the whole piece. Thus, a corpus-based approach coupled with a discourse-based ap-proach is sure to contribute to better pedagogical corpus exploitation. Developing their lesson plans the language learners had to familiarize themselves with concordances first, choose suitable patterns, then study fuller contexts, reading for detail, analyze mechanics, collocations, and colligations. This contributed to the learners' receptive skills development. At the next stage students created their own tasks for the peers. Their productive skills improved as they had to write (at home) and speak (in the classroom) teaching their groupmates according to the developed lesson plans. Moreover, being confronted with mechanics, collocations, colligations on a regular basis, language learners developed vocabulary/writing/grammar skills, put to use various teaching techniques they learnt in their English teaching methodology class. Students of English not only got access to authentic texts, but were also involved in research and were provided with an opportunity to be creative. All these factors contributed to professional competence of future teachers.

Strengthening grammar to build research skills
Any natural language is always and forever changing. For example, contrastive analysis of some grammar books and naturally occurring English (corpus research of object clauses introduced by suggest) revealed mismatches. According to literature, should in the object clause is said to be the norm in British English. The research conducted on the British National Corpus (BNC), accompanied by paper-based analysis (17 novels / 5390 pages by contemporary English writers) disproved this point. The received data inspired me to adjust grammar teaching syllabus in order to improve pedagogical practice. Similar tasks given to students enable instructors to teach grammar inductively and implicitly. This way learners act as explorers, which contributes to enhancing their metacognitive skills and language awareness. It is intellectually challenging to analyze the obtained data, consequently, such tasks help learners to acquire research skills.

Results and Discussion
In this research, the correlation between doing corpus-based activities, students' degree of autonomy and achievements (grades in exams, i.e. proficiency levels) was studied. Based on the results of the empirical research, it can be concluded that employing English corpora in conjunction with traditional forms of independent work had an important effect on enhancing learners' autonomy.
At the same time, as the research was conducted on a sample of 13 students (the experimental group), it does not allow us to generalize and refer the results to all educational situations. Nevertheless, the obtained evidence indicates a possibility for enhancing learner autonomy with the help of corpus-based tasks and activities. The findings from the two stages of the experiment were compared and are presented in Table 1. The analysis of the experimental group's results prior to and after the experiment (the assessment was made at the end of the first and second semesters) demonstrates that the level of language proficiency grew: Table 1 Experimental group's results before and after the experiment, grades Таблица 1 Результаты экспериментальной группы до и после эксперимента, баллы

Grades
Before the experiment After the experiment Excellent/A 2 students 5 students Good/B 5 students 2 students Satisfactory/C 6 students (4 of the students failed the exam the first time and had to retake it) 6 students (2 of the students failed the exam the first time and had to retake it) The students of the experimental group became better learners. They were able to do more homework compared to the three control groups (as it has already been mentioned, corpus-based tasks and activities were combined with extensive, intensive reading and listening), demonstrated good quality of preparation (which was reflected in their grades during the second semester) and in most cases showed significant improvement on classroom performance as Table 2 indicates. The data reflect the effectiveness of the utilized approach and show a correlation between the results and the students' proficiency levels demonstrated in the examination at the end of the study (cf., Table 1). Additionally, the experimental group made two creative projects in English, the latter of which won the first prize at the annual Creative Project Competition held at the Institute of Foreign Languages (USPU). Table 2 Results of formative assessment in the first and second semesters, average grade Таблица 2 Результаты формативного оценивания в 1-м и 2-м семестрах, средний балл Average Grade Assessment period S1 S2 S3 S4 S5 S6 S7 S8 S9 S10 S11 S12 S13 The results of the experimental group contrasted against those of control groups 1-3 (the end of the second semester) can confirm the aforementioned conclusions and also stress the necessity of incorporating corpus-based approaches into students' independent and classroom work. It is evident from Table 3 that the students in the experimental group made significant progress: seven out of 13 participants got grades above average (≈54%). The unsatisfactory progress in the control groups is fairly obvious as the majority of students got grade C (43 out of 55 students in control groups 1-3). Thus, the experimental results support the hypothesis of the study and justify the practical usefulness of utilising corpus-based tasks and activities in teaching English as a foreign language.  Table 3 demonstrates the basic statistical parameters of the students' results in the experimental and control groups in the final knowledge assessment, including the number of students in each group, their grades, the arithmetic means, standard deviations and variances. The average value (mean) difference convincingly shows the experimental group's more substantial progress versus the academic achievements of their peers in the control groups. The calculated standard deviations and variances indicate the greater diversity in the experimental group's results compared to Term I (SD = 0.6; V = 0.7), which can be explained by more excellent/A grades received by the learners. At the same time, though the diversity of grades is lower in the control groups (SD = 0.4), most students got satisfactory grades, which does not show enough progress in English.
It is worth mentioning that the tool used to assess the students' answers was an analytic rubric with detailed descriptors. This scoring instrument allows a teacher to assess individual aspects of a student's performance in the examination. After that the scores for the individual parts are summed, and the total score is obtained [29]. The grades for extensive reading, portfolio, and writing were given to the students prior to the examination. Limitations of space prevent an exhaustive description of each of the six criteria (extensive reading, student portfolio, writing, reading and retelling, sentence translation, dialogic interaction with the teacher and a peer). A sample is provided in Table 4. Corpus-based tasks and activities provided observation of real-life language, which resulted in profound linguistic reflection. Regular independent work contributed to honing not only various language skills but also cognitive skills necessary for developing and implementing metacognitive strategies. The impact of the utilized approach on the process of education in general and the level of learner autonomy, in particular, is definitely favorable.
This hands-on experience shows that today corpus applications in language teaching are more confined to teachers, not learners, as teachers are much more active users of English corpora. What is more, our communication with colleagues across Russia shows that the percentage of such teachers is still low. Because our aim is to enhance learner autonomy, we should bring students up to make more extensive use of language corpora, facilitating learner-centered activities, encourage language learners to access online corpora, do corpus-based research projects to help them become linguistic researchers, as advantages of using corpus methods and tools are numerous.

Conclusion
In conclusion, it should be noted that corpus analysis should be used in materials development as corpora are sources of extensive data and provide opportunities for innovation in language learning and teaching. Thus, data-driven learning can become an alternative to the traditional study of collocations, grammatical structures, spelling rules outside and in the classroom.
In addition, homework done with the help of corpus tools and methods becomes a less time-consuming, more confidence-building and horizon-broadening activity; dictionary work, for example, takes much more time. Corpora also expose learners to real-world language and its variation, motivating and engulfing them. Corpus-based activities create conditions for learner autonomy as students implement a number of cognitive and metacognitive strategies (for instance, contextualization, resourcing, translation, inferencing, selfmonitoring, self-management, directed attention, selective attention), get motivated and acquire knowledge about language learning. This kind of information is crucial for promoting learner autonomy, as only those students who have mastered the aforementioned strategies can be responsible for their learning and consciously take control over enhancing their receptive and productive skills. Finally, there is the teacher whose function is to develop instructional materials and guide the learner along the way.
The experiment results indicated that doing additional reading, listening, corpus-based tasks and activities, using English corpora as a look-up and learning facility to support writing, reading, vocabulary and grammar acquisition on a day-to-day basis, students made noticeable progress. Corpora broadened the learners' horizons and raised awareness of what real-life English is like, thus fostering autonomous, self-directed learning. At the same time, it is still unclear to what exactly extent using corpora maximizes learner autonomy. A larger research sample would be desirable. Another relevant question is the role of the teacher's personality, i. e. whether the result depends on the amount of effort taken by the teacher, his or her experience and involvement. That is why a more detailed study is needed.