1 Introduction
The value of corpora in language education has long been acknowledged. Numerous books, articles and conference presentations have promoted a variety of corpus applications: from more accurate corpus-based descriptions of the target and learner language, through the creation of new resources and tools for language teaching and learning, to the use of corpora by learners in the language classroom (cf. Ghadessy, Henry & Roseberry Reference Ghadessy, Henry and Roseberry2001; Sinclair Reference Sinclair2004; Aijmer, Reference Aijmer2009; Campoy-Cubillo, Bellés-Fortuño & Gea-Valor, Reference Campoy-Cubillo, Bellés-Fortuño and Gea-Valor2010; Boulton, Carter-Thomas & Rowley-Jolivet, Reference Boulton, Carter-Thomas and Rowley-Jolivet2012; to name just a few collections of papers on these topics). An international biennial conference series, TaLC (Teaching and Language Corpora), is devoted solely to the theme, and for twenty years now has served as a platform for practitioners in this area to present their work, their findings, their ideas and their skills (Aston, Bernardini & Stewart, Reference Aston, Bernardini and Stewart2004; Hidalgo, Quereda & Santana, Reference Hidalgo, Quereda and Santana2007; Kübler, Reference Kübler2011; Frankenberg-Garcia, Flowerdew & Aston, Reference Frankenberg-Garcia, Flowerdew and Aston2011; Thomas & Boulton, Reference Thomas and Boulton2012).
To an extent, corpora have left the realm of the academic debate and found their way into real-life teaching. They are present in the production of dictionaries (e.g., Longman dictionary of contemporary English 2003) and reference grammars (e.g., Biber, Johansson, Leech, Conrad & Finegan, Reference Biber, Johansson, Leech, Conrad and Finegan1999), as well as in the design of language materials and courses (Mascull, Reference Mascull1995; McCarthy, McCarten & Sandiford, Reference McCarthy, McCarten and Sandiford2005; Barlow & Burdine, Reference Barlow and Burdine2005; Lee & Swales, Reference Lee and Swales2006). Yet despite the enthusiasm of a handful of specialists (Johns, Reference Johns1991; Tribble & Jones, Reference Tribble and Jones1997; Boulton, Reference Boulton2009), they are still rarely used by teachers in language classrooms (Römer, Reference Römer2010; Boulton, Reference Boulton2010).
There are many reasons for the reluctance of language teachers to exploit corpora in their work. Great demands on hardware and computing skills have frequently been quoted as obstacles (Yoon & Hirvela, Reference Yoon and Hirvela2004; Tian, Reference Tian2005; Boulton, Reference Boulton2010), although they have recently come to be less acute as computer labs have become more commonplace in educational institutions, and the level of computer literacy among both teachers and learners has generally increased. In his survey on the use of corpora by language teachers and educators, Tribble (Reference Tribble2012) found that a lack of access to a computer and a lack of confidence in its use are among the reasons least frequently listed by the practitioners who do not use corpora in their teaching. More telling is the dearth of resources which are easily available, free or relatively inexpensive, reliable and stable in terms of access, as well as user-friendly (Römer, Reference Römer2010; Tribble, Reference Tribble2012). In spite of numerous on-going corpus projects in different parts of the world devoted to the creation of new corpus resources and tools, their products are usually available only for in-house use (Aston, Reference Aston2004), and are too complex or not versatile enough to serve the specific needs of language teachers (Boulton, Reference Boulton2010).
However, the problem which is probably at the heart of teachers’ disinclination to exploit corpora in language instruction is their lack of knowledge about the different ways that large linguistic databases can be used in the classroom (Mukherjee, Reference Mukherjee2004; Römer, Reference Römer2009, Reference Römer2010). Graduates of language departments might have heard about or even encountered corpora during their linguistic education; in some institutions, they might have even used corpora regularly in their language or linguistics classes (O’Keefe & Farr, Reference O’Keefe and Farr2003; Götz & Mukherjee, Reference Götz and Mukherjee2006; Amador Moreno, O’Riordan & Chambers, Reference Amador Moreno, O’Riordan and Chambers2006; Chambers, Reference Chambers2005; Farr, Reference Farr2008). However, this experience does not automatically imply that they know how to apply corpora in their teaching (Tribble, Reference Tribble2012). As Breyer (Reference Breyer2009: 156) observes, “recognising that there is a significant difference between learning and teaching with corpora, as well as providing student teachers with the required skills, is of great importance.”
There is then an argument that pre-service and in-service teachers should be instructed explicitly in the potential of corpora for language teaching. Unfortunately, there are only a few books which serve as manuals for teachers in this area (Tribble & Jones, Reference Tribble and Jones1997; O’Keeffe, McCarthy & Carter, Reference O’Keeffe, McCarthy and Carter2007; Reppen, Reference Reppen2010; Bennett, Reference Bennett2010; Pérez-Paredes & Díez Bedmar, Reference Pérez-Paredes and Díez Bedmar2010; Flowerdew, Reference Flowerdew2012; CALPER, no date). A need for institutionalized teacher-training courses devoted to or featuring the applications of corpora in language instruction has been voiced by several researchers (Mukherjee, Reference Mukherjee2004; Römer, Reference Römer2010; McCarthy, Reference McCarthy2008; Breyer, Reference Breyer2009; Hüttner, Smit & Mehlmauer-Larcher, Reference Hüttner, Smit and Mehlmauer-Larcher2009).
Two such pre-service teacher-training courses have recently been described in the literature. Breyer (Reference Breyer2009) presents a 22-hour course at the English Department of Duisburg-Essen University, whose aim was to “to create a learning experience for student teachers from two perspectives: as learner and as teacher” (op. cit.: 157). The participants were expected not only to develop a basic understanding of corpus analysis but also to learn about applications of corpus-based activities in the classroom. In every class the students first followed corpus-based language tasks, and then reflected on the experience both as learners and as teachers. They also had to complete a number of small projects including writing a reflective essay, reviewing a concordancing programme, and producing a corpus-based language exercise. In the second account, Hüttner et al. (Reference Hüttner, Smit and Mehlmauer-Larcher2009) present an innovative teacher education project at the English Department of the University of Vienna. This teaching English for special purposes (ESP) module covers four one-semester courses of approximately 28 lessons each, which students are expected to take during their last two years of study. The two core courses of the module are based on a model of “mediated corpus-based genre analysis” (Hüttner et al., Reference Hüttner, Smit and Mehlmauer-Larcher2009) and their aim is to teach students how to apply the techniques of genre analysis and corpus linguistics in order to explore new and diverse ESP settings, and how to implement the results of these analyses in language instruction. Participants are required to complete projects in which they analyse corpora of self-selected genres in view of potential teaching situations, and prepare materials for specific groups of learners. The authors stress that their aim is not the development of clearly predefined competences and skills adequate for dealing with predictable teaching situations, nor the promotion of procedural aspects of teaching (classroom management, lesson planning, giving feedback, etc.). Instead, their objective is to “allow students to develop the necessary competence and capacity as language teaching professionals to act autonomously in familiarising themselves with potentially unknown ESP genres and, at the same time, in operationalising their newly gained insights for teaching purposes” (op. cit.: 108).
In addition to detailed accounts of the courses, an evaluation of their effectiveness is presented in both papers. Hüttner et al. (Reference Hüttner, Smit and Mehlmauer-Larcher2009) analyse a typical student project created within the framework of the module in order to highlight the competencies participants gained during the course. Breyer (Reference Breyer2009) focuses on direct responses from her students, examining data from short reflective essays on one of the course units (teaching some and any), and a questionnaire which recorded trainees’ feedback on the project involving the creation of a language exercise with concordances. In both cases the researchers underline the positive effects of their courses, which go beyond the development of corpus literacy and skills in applying corpora in language teaching. Both courses encouraged students to reflect on language and teaching methodology in a broader perspective, and thus were conducive to raising their language as well as “teaching awareness” (Breyer, Reference Breyer2009: 167).
The aims and principles motivating these two studies (Breyer, Reference Breyer2009; Hüttner et al., Reference Hüttner, Smit and Mehlmauer-Larcher2009) were also adopted in a semester-long course on the use of corpora in language education offered to graduate students at the Institute of Applied Linguistics, University of Warsaw. Its objectives further included giving participants an opportunity to experience corpus-based activities both as learners and teachers. In addition, the course aimed at enabling students to develop a general competence to deal autonomously with new language situations and observations, including ESP contexts, and to apply the results of their corpus-based explorations in their teaching. The course is part of the teacher-training programme, and at the moment of writing it is running for the third time. The aim of this paper is to outline the objectives, the design, the syllabus, the progression and the outcomes of the course, as well as to present the results of two questionnaires distributed to the students prior to and after the second edition of the course. The results are modest due to the small number of participants, yet they are indicative of a further course of action to be taken not only by the instructor in the next editions of the course, but also by her colleagues as well as other educators involved in various teacher training programmes, in order to promote the use of corpora by language teachers in their work.
2 Description of the course
2.1 The setting
The course on Corpora in Foreign Language Teaching is offered to graduate students at the Institute of Applied Linguistics, University of Warsaw, in the winter semester of the first year of their MA programme. The Institute’s mission is to train linguists, translators, and foreign language teachers. The students get instruction in two foreign languages, choosing from English, German, French, Spanish and Russian, but for over 90% of them one of these languages is English. The translation and teacher-training tracks are both obligatory, yet few students plan to follow a teaching career in future. The students have already completed their BA and have received training in teaching a foreign language at this level. The course is one of several electives offered within the group of foreign language teaching subjects, thus it is not obligatoryFootnote 1. The course is taught in English and there are sixteen places available. No BA-level courses the students had previously taken were based on corpora or featured corpus use, yet the students may have heard about corpus-based research in their linguistics classes.
2.2 Design criteria
The course is designed to introduce students to the concept of a corpus and its analysis, and to outline various applications of corpora in language education, with special emphasis placed on the in-house preparation of courses, teaching materials and class activities. Students are expected to have a good knowledge of language teaching methodology and be familiar with various language teaching techniques. On the other hand, they are not assumed to have had any prior contact with corpora. Thus, similar to Hüttner et al.’s (Reference Hüttner, Smit and Mehlmauer-Larcher2009) programme, the focus of the course is not on the teaching procedures per se (such as deciding on lesson aims or planning a lesson) but on the presentation and exploration of various corpora, corpus-based resources and tools, and on the demonstration of their potential for language teaching. However, during the classes students are constantly reminded about the importance of placing corpus-based materials and activities in a larger educational context so that they serve particular teaching objectives and learning processes and do not become an end in themselves.
Several criteria were applied in the design of the course. The most important one was a presentation of a whole range of corpora, corpus-based materials and tools. Instead of focusing on one or two corpora and concordancers and centring all class activities around them, the course offered a comprehensive overview of the variety of existing resources. The conditions applied in the selection of materials and software to be presented to the participants were that they had to be free, easily available and stable in terms of access. Paid resources and tools could only be mentioned in the instructors’ presentations. The next criterion for the choice of resources was a relatively simple and user-friendly interface.
Another important idea in the design of the course was the exploration of different functions of the software presented. This was done with the aim of exemplifying various types of corpus analysis (studying differences between genres, retrieving collocations, etc.). The examples of possible corpus explorations were always set in the context of language learning activities rather than pure linguistic analysis. As in the case of Breyer’s (Reference Breyer2009) course, it was assumed that the most effective way of presenting the uses of corpora in language teaching would be to engage students in a range of activities as language learners rather than as teacher trainees. To this end, during the classes students were exposed to a variety of language learning tasks at their level, which they had to complete. However, this course had quite different aims and guiding principles from other corpus-based classes described in the literature (e.g., Cresswell Reference Cresswell2007; Boulton, Reference Boulton2009; Charles, Reference Charles2012). The other courses aimed at teaching language via corpus-based activities, thus the choice of resources and tools depended on their usefulness in reaching this immediate aim. The objective of the course described here was to present students with possible ways of exploiting corpora in language education, with the focus consequently on the presentation of a variety of interesting applications. Also, the class discussions did not concern the language that students had studied but the pedagogical value of the resources and tools covered.
2.3 Course content
The course lasts one semester and encompasses, depending on the length of a semester, from thirteen to fifteen 90-minute classes. It consists of three main thematic modules, each spreading over three or more classes. The first module is an introduction to corpora and corpus tools along with an overview of their applications in language education. The second concentrates on the exploration of large general corpora for teaching different language elements and skills. The third is devoted to the creation and analysis of small specialized corpora and their uses in syllabus design and the production of language teaching materials and activities. Table 1 presents the course syllabus for the winter semester 2012/2013.
Table 1 Course syllabus for the winter semester 2012/2013
The introductory module features two presentations by the instructor. The first one introduces the concept of a corpus, presents various types of corpora and major corpus projects around the world, and discusses the main functions of concordancing software and other corpus tools. It also presents students with links to relevant corpus-related websites. The other lecture provides an overview of various uses of corpora in language education, starting with the creation of dictionaries, syllabuses and course books, and ending with ‘data-driven learning’, in which language learners have a direct contact with corpus citations and other information derived from corpora. Within this module, participants also have an opportunity to follow a full 90-minute language class on the topic of politics. During the class, students perform several tasks, only a few of which involve corpus exploration. The aim of this session is to demonstrate to teacher trainees in practice how corpora can be integrated into a language lesson and how they can supplement other language teaching techniques.
The classes in the next module present ideas for applying corpora in teaching vocabulary, phraseology, grammar and discourse organization as well as for practicing reading and writing. Participants perform tasks whose aims are three-fold: (1) to make them aware of the kinds of language points that can be addressed with corpus-based materials or activities; (2) to demonstrate how these materials and tasks can be incorporated into a lesson; (3) to introduce new tools or their new functions. Participants are presented with activities which involve the consultation of a corpus or which have been created with the help of a corpus (gap filling, multiple choice, matching). The benefits of corpora for teaching spoken language (pronunciation, listening comprehension and speaking) are covered only marginally with a brief exploration of the Backbone corpus (Kohn, Reference Kohn2012). Even though there are papers and projects advocating the value of corpora in this area (e.g., Szakos & Glavitsch, Reference Szakos and Glavitsch2010; Aston & Rodi, Reference Aston and Rodi2012), these applications have been judged too complex and too time consuming to implement in regular language classrooms.
In the third module, students learn how to compile their own corpora, how to analyse them, and how to create corpus-based materials and activities for teaching languages for special purposes (LSP). The theoretical presentation discusses the characteristics of LSP, following which students are taught how to compile a small corpus and are introduced to AntConc and its features (Anthony, Reference Anthony2006, Reference Anthony2013). The instruction in corpus compilation and in operating the software is based on two comparable 30,000-word corpora created by the instructor containing medical articles on urology. In the last class of this module, students create their own very small 10,000-word corpora of weather forecasts and analyse them by following precise instructions prepared by the course instructor.
The end-of-semester project involves the creation and analysis of a small corpus of specialized language (ca. 30,000 words) on a chosen topic, as well as the preparation of a state-of-the-art language lesson which will include materials and tasks based on this corpus. Students are encouraged to work in pairs and are required to submit their corpora and the results of their analyses (such as lists of terminology and frequent collocations) in order to prove that they have acquired relevant skills in corpus manipulation. They also have to submit a lesson plan and all teaching materials (handouts, exercises, texts) needed for the lesson. In the last one or two sessions (depending on the number of classes in the course), participants have an opportunity to present their projects to their peers, describing how they chose their topics and built and analysed their corpora, and discussing their lesson plans and their corpus-based materials and activities.
The participants who have taken this course so far compiled their corpora on a variety of topics such as furniture catalogues, football news, texts on modern architecture or military issues. An example of a lesson plan on football news prepared by one of the students is presented in Appendix 1.
The end-of-semester project is to be submitted three weeks before the end of the course and is the reason why the three course modules are not presented in consecutive order. Participants need some time to compile and analyse their own corpora and prepare their lesson plans and teaching materials, so after the introductory segment followed by two classes on the exploration of general corpora for teaching vocabulary and phraseology, a three-class section is scheduled on building and exploring DIY corpora.
Classes take place in a computer lab with sixteen student computers connected to the Internet and run from the Moodle platform. Except for four classes featuring the instructor’s presentations (also available on the platform), students complete tasks on Moodle. The tasks require them to study selected language points and include detailed step-by-step instructions on how to operate the software. Exercises such as gap filling or matching are presented as CALL (computer-assisted language learning) activities prepared with the authorable software Hot PotatoesFootnote 2. The instructor walks around the classroom assisting students, who can work individually or in pairs to complete all the tasks and submit them through Moodle. If they do not finish the tasks in class they have a week to complete them at home and submit them before the next class. The last 10-15 minutes of each session are devoted to class discussion of the activities with a focus on their pedagogical benefits.
The Moodle content of the course can be viewed in guest mode at http://moodle.ils.uw.edu.pl/course/view.php?id=101. All the course materials and activities, except for the option of submitting completed tasks, have also been made available to the general public at http://corpora.blog.ils.uw.edu.pl/.
2.4 Tools used
Two main corpora that the students explore during the course are the British National Corpus (BNC) with its variety of interfaces (BNC Online, BYU-BNC, Phrases in English) and the Corpus of Contemporary American English (COCA). The remaining resources – the Michigan Corpus of Academic Spoken English (MICASE) and Backbone – are only briefly studied during the course. The first two editions of the course also featured the Collins Cobuild Corpus Concordance Sampler, but this is no longer available. Students are introduced to AntConc as a tool to handle their own corpora. Other resources and tools presented in class are collocation finders (Just the Word, For Better English, Word and Phrase), Tim Johns’ and MICASE Kibitzers, the Compleat Lexical Tutor, academic word lists and academic word highlighters, web concordancers and electronic dictionaries. A complete list of all the resources and their URLs can be found in Appendix 2.
2.5 Data collection
In order to evaluate the effectiveness of the course, two kinds of data were collected from the students. First, they were asked to fill out one questionnaire before the course, targeting their expectations, and one afterwards for their perceptions of its outcomes. Second, the students’ projects, including their own corpora, their analyses and their teaching tasks, were retained in a database as a living proof of the skills they had actually gained during the course. It is beyond the scope of this article to discuss all the available data collected; the examination below will focus solely on the two questionnaires.
The two questionnaires were distributed to the thirteen students participating in the second edition of the course. The first one was completed at the beginning of the first class in order to (1) tease out the reasons why students had signed up for this elective, and (2) establish how much prior knowledge they had on corpora and corpus analysis. The questionnaire consisted of two closed questions and three open ones. In question 1 the students had to indicate which out of nine possible reasons for choosing the course (including the open-ended ‘other’) explained their decisions. The participants could choose up to three reasons and they were asked to arrange them in order of importance. In question 2, the students had to tick one of five sentences which best described their prior contact with corpora; questions 3–5 required them to write their definitions of three terms: a corpus, a concordance and a concordancer. The other questionnaire was filled out by the same students during the last session to elicit their reactions to the course, its scope and content, as well as to its procedures. The questionnaire did not address directly the students’ perceptions of the place of corpora in language education, but their answers were indicative of their attitudes in this respect. The post-course questionnaire was longer and consisted of nineteen questions, three of which were open-ended. In the closed questions the students rated various aspects of the course such as the number of resources explored, the balance between theory and practice, or the workshop format of the classes. The ranking was done on a 7-point scale ranging from 1 (not at all) through 4 (so-so) to 7 (very much), or from −3 (e.g., too easy), through 0 (e.g., the right level) to +3 (e.g., too difficult). However, after each question the students were invited to give additional comments. The three open questions related to what the participants liked most and least in the course and what they would change in it. The final question asked them to give a global mark to the course using the Polish academic marking scale extending from 2 (fail) through 3 (satisfactory), 3+, 4 (good), 4+ to 5 (very good).
Since the questionnaires were completed by only thirteen students (all the students participating in the course), a purely quantitative analysis would not be very meaningful. The discussion below will therefore concentrate on a qualitative examination of the responses and comments provided by the students, supplemented with the opinions expressed in informal conversations with individual students during the workshops.
3 Results
The results of the first questionnaire are presented in Table 2.
Table 2 Summary of students’ responses to the pre-course questionnaire
The most frequently-checked reason why the students had chosen the course (Q1) was its convenient time, which fitted well in the students’ weekly class schedules (eleven students, six of them giving it as the most important reason). Similarly, five students admitted that their preferred electives, devoted to translation, had already been full. The other frequently indicated motivation was a desire to find out more about corpora and the ways to use them (eleven responses, but only three as the most important reason). Barely half of the students (seven) were interested in language teaching and wanted to take a course related to it, and for only three of these was it the main motive for choosing this course.
As far as the students’ prior knowledge is concerned (Q2), eight indicated that they had a rough idea about what a corpus was, and four had heard the term before but admitted no understanding of the concept. Only one student had worked extensively with corpora for her BA dissertation. The open questions revealed that even though the majority of the students admitted some familiarity with corpus linguistics, they in fact had a very poor grasp of its basic concepts. All thirteen students attempted to write a definition of a corpus (Q3), seven of the term concordance (Q4), and five of the term concordancer (Q5); however, with the exception of the student who had done corpus-based BA research, all the answers were inaccurate, vague or even meaningless. For example, one student conceptualized a corpus as a “database of words, phrases, collocations and sentences taken from texts”, another defined a concordance as “a form of congruency”, and still another described a concordancer as “a means of matching factors.” These answers imply that at least some of the participants had heard or seen these terms mentioned before, but corpora did not feature prominently in their previous classes or any other academic work. The picture emerging from the initial questionnaire is thus that the students had had no prior experience with corpora, and no understanding of even the basic concepts in corpus linguistics. Although they were eager to find out about the field as such, they had no particular interest in the use of corpora in language teaching.
The results of the post-course questionnaire are summarized in Table 3. On the whole the students liked the course, evaluating it as good or very good (Q19). The course also met their expectations (Q17), and they reported that after the course they had a good understanding of what a corpus is and what it can be used for (Q1 and Q2). The number of tools presented seemed more or less sufficient, only one participant claimed that far too many tools had been presented (Q3). The students also reported that the course had given them a good opportunity to become acquainted with the tools and materials (Q4). However, the additional comments revealed that they found it difficult to remember various functions of the software, and indicated that they would like to have had more time to explore them.
Table 3 Summary of students’ responses to the post-course questionnaire
The students seemed satisfied with the format of the course. They found both the theoretical presentations clear and easy to follow (Q6) as well as useful and informative (Q7). They enjoyed working on Moodle in the workshop format (Q10) and found the instructor helpful during the hands-on activities (Q11), though they were not entirely satisfied with the amount of feedback they had received on the activities completed (Q12). The only complaint that emerged from the responses and additional comments was dissatisfaction with too big a workload.
In the question referring to the balance between the presentations and hands-on activities (Q5), only four students found the balance right: for six there were slightly too many tasks, and two found the practical tasks far too numerous. For four students the level of difficulty of the hands-on activities (Q8) had been just right, five evaluated the tasks as a little too difficult, and three found them very difficult; only for one student had the activities been too easy. The students evaluated the hands-on activities as useful, although here the responses were less enthusiastic than for the instructor’s presentations (Q9). The general complaint arising from the comments was that the activities had been too time-consuming, the procedures rather difficult, and the students had had to spend too much time at home completing the tasks. This in fact shows the inconsistency of the students’ responses – on the one hand they felt that they had not explored the tools and functions enough, but on the other hand they thought there had been too many activities. The students positively evaluated the relevance of the end-of-semester project (Q13).
The students believed that the course had demonstrated well how to use corpora for language teaching (Q14), though the additional comments to this and other questions indicated that they would need more time to feel confident in this respect. Here again they reported they needed to get to know the tools better, but also called for more guidance about how to create their own materials and activities.
They particularly liked the activities related to the end-of-semester project, creating and analysing their own corpora, and reported they had learned a lot of new language from this experience (Q15). They also enjoyed the applications of general corpora for learning vocabulary and phraseology. In informal conversations during workshops they admitted that they had started using corpora to answer their own linguistic queries surfacing in other courses, particularly in their translation classes. The students least liked the class devoted to the applications of corpora for teaching grammar, and did not find data-driven activities a useful method for teaching this language element (Q16).
The answers to the open question related to recommended changes in the course (Q18) repeated the comments which had already surfaced in the previous responses: more time to explore the resources presented, more tools and more functions, more examples of classroom activities, but at the same time less homework.
The results of the post-course questionnaire demonstrated that the students had positive reactions to corpora on the whole and saw their benefits in language education. In individual conversations they admitted that they found corpora useful and relevant resources for their own learning, and claimed that they wanted to find out more about the different possibilities for language exploration. At the same time, they found corpus analysis daunting. However, the difficulties related to the manipulation of corpora were not a reason for their rejection as a possible learning aid, the students just felt they needed more time and practice in order to take full advantage of the approach.
4 Discussion and conclusion
Even though the results of the first questionnaire may seem rather discouraging, the post-course questionnaire looks more promising, at least on the surface. The students’ reactions to the course were positive and the course met their expectations. However, it has to be borne in mind that the major substantive expectation the students expressed prior to the course was to become acquainted with corpora and their various uses, rather than to explore specific applications of corpora in language education. After the course they reported having a good grasp of the concepts related to corpus linguistics, but they expressed their lack of confidence in operating software for corpus analysis themselves. Even though they maintained that the course had familiarized them well with various uses of corpora in language teaching, some admitted that they needed more guidance on how to select appropriate language problems to address with corpus-based materials and activities in the classroom, or how to design such materials and activities.
In order to ensure that corpora find their way into language teaching on a larger scale, two criteria seem to be necessary. First, teachers need to be fairly competent corpus users who are well acquainted with the resources and tools available and can handle them adeptly. Second, teachers also need to have pedagogic skills related to implementing corpora in the teaching process by creating suitable and pedagogically sound corpus-based materials and data-driven tasks, combining them well with other teaching techniques and incorporating them in the instructional context. As regards the first criterion, it has long been acknowledged that manipulating corpora and analysing the output of these manipulations are complex skills which require special training (Yoon, Reference Yoon2008; Boulton, Reference Boulton2009; Römer, Reference Römer2010). The questionnaires pointed to the fact that these skills take time to develop, and even a semester-long course with a large component of hands-on activities may not give students enough confidence in this respect. Because the focus of the course had been on various ways of handling the software (criterion 1), the students’ responses suggested that it had not addressed the relevant pedagogical skills sufficiently (criterion 2), even though the corpus manipulation activities had always been set in the language instruction context.
A deeper analysis of the students’ responses leads to the realization that Römer’s (Reference Römer2010) call is too modest for “‘spreading the word’ about corpora” among teachers by universities which could “organise ‘open days’ and offer lectures and workshops on issues that directly relate to teachers’ problems and needs” (op. cit.: 95). The questionnaire demonstrated that even a 22-hour course devoted solely to these issues, a luxury many teacher-training programmes cannot accommodate, is not sufficient.
It must be acknowledged, however, that the objectives of the course were quite ambitious. The guiding principle in the design of the course was to present the students with many different resources and kinds of analyses. In consequence, they may not have had time to become confident in operating all of them. Their reactions might have been different if the course had focused on a smaller number of resources and the opportunity to get to know them better. This can be viewed as a solution to the problem of the reported lack of confidence in handling corpus-based tools. Yet such a solution also has its shortcomings. Turning the course into a tutorial on using a selected tool will make teacher trainees resource-dependant. In contrast, presenting would-be teachers with a comprehensive overview of various applications has the advantage of enabling them to assess the needs of their own students in the future, to decide on appropriate corpus-based and data-driven techniques, and to choose suitable resources which can meet these demands. It will also make teacher trainees better able to deal with new tools which will be developed and made available after they complete the course.
Thus the conclusion from the study is that corpus exploration cannot be left to one course within a teacher training programme. Until corpora enter mainstream education in language departments and teacher-training institutions on a large scale, which will allow future teachers to encounter various resources, tools and methods of corpus analysis in a variety of language and linguistics classes, we cannot expect that corpora will find their way into language classrooms in other instructional settings. Until teacher trainees gain confidence in handling corpus-processing software and in interpreting the output of their analyses, we cannot hope for educators other than highly-specialized experts such as lexicographers and coursebook writers to take advantage of corpora in their work. It seems that we first have to persuade our colleagues working in language and linguistics departments and on teacher-training programmes to use corpora in their teaching before we can expect that the word will spread further. This process has already started. Tribble (Reference Tribble2012) reports that almost 80% of the respondents to his questionnaire, who admitted using corpora in their teaching, worked in tertiary education. Only extensive exposure to corpora by future teachers coupled with suitable teacher training in the applications of corpora in language education may bring a substantial change in the scope of corpus use in language classrooms.
Appendix 1. An example of a lesson prepared by a student as part of the project
Topic: Football
Lesson Plan
Stages:
1. Warm-up (10 minutes, whole class). The teacher asks the students about the latest football news they heard about, if they watched the recent game on TV, who won, if it was exciting, if anything unusual happened etc.
2. Practicing vocabulary (10 minutes; pair work and whole class; handout). The teacher asks the students to do Activity 1 from the handout. Students work in pairs matching the definitions with football vocabulary. The teacher checks the answers with the whole class.
3. Introducing new vocabulary, working with the articles. (15 minutes; pair work and whole class, handout). The teacher asks students to complete Activity 2 from the handout. They are supposed to match the definitions to words they find in extracts.
4. Studying grammatical features (10 minutes, pair work, whole class, computer-corpus). The teacher asks students to do the exercise 3 in the handout. They are supposed to study the sentences with the names of teams and to look for any interesting features. Then the teacher asks the whole class about the results. The main aim is to show students that the verb occurring after the name of the team is in a plural form.
5. Studying the collocates. (25 minutes, pair work, whole class, computer-corpus). The teacher asks students to do the exercise 4 in the handout. They are supposed to look for the collocates of the word “goal” in with the football-related meaning. Then the teacher asks students about the results of their research. Students read aloud the collocates their found, teacher writes them down on the whiteboard.
6. Introducing new vocabulary (15 minutes, pair work, whole class, computer-dictionaries). The teacher asks students to do exercise 5 from the handout. They are supposed to look for the Polish equivalents of given words in Internet dictionaries. Then the teacher asks the students to read the equivalents they found. If there are any problems or doubts, the students are supposed to look for the equivalents they could not find, on the Polish Webpages with the football articles
7. Assigning homework (5 minutes). Students are supposed to complete a gap-filling exercise with the vocabulary they learned during the lesson.
Topic: Football
Handout 1
Activity 1
Match the definition to the term
Activity 2
Study the extracts from football articles and find the words matching the definitions below.
• Wolves manager Mick McCarthy had punched the air in delight at the final whistle - which came after six minutes of stoppage time during which Arsenal bombarded the visitorsʼ penalty area.
• Wolves then survived the second-half dismissal of midfielder Nenad Milijas as inspired goalkeeper Wayne Hennessey denied their opponents time and again.
• “He can head the ball, heʼs as strong as an ox, he can run, dribble and shoot. Most important of all, heʼs a smashing lad.”
• His hat-trick against Inter Milan in the Champions League in October 2010 thrust the Cardiff-born left-sided player into the world spotlight, but Redknapp reiterated that any approach from England or abroad would be rebuffed.
• Things got even tougher for McCarthyʼs side when Milijas was sent off for a foul on Arteta on the edge of the box.
• There was still time for Vermaelen to have a volley saved by Hennessey, but Wolves, thanks mainly to the efforts of their goalkeeper, held on.
• He pounced on a poor header by Britton following Kennyʼs long clearance - the midfielder directed the ball dangerously towards his own goal - and the Scotland international brushed off the challenge of Ashley Williams to fire past Vorm.
• Luiz, making his first appearance of the season in place of captain John Terry, struck in the 67th minute to reward an improved second-half display from the hosts.
The sound announcing the end of a game:
A linking position between the defenders and strikers:
To move the ball along with you by short kicks:
The achievement of scoring 3 goals during one game:
Made to leave the pitch by the referee:
A synonym for penalty area:
A kick, when a player kicks the ball before it touches the ground:
An occasion when a player kicks the ball away from him or her:
A team which provides a place for a football game:
Activity 3
Use the BNC corpus to study sentences concerned with the football teams (e.g., Arsenal, Chelsea, etc.). Can you see anything unusual? Find several sentences and comment on the use of verbs in them.
Activity 4
Use the BNC corpus to study collocates of the word “goal” with football-related meanings.
Activity 5
Use the Internet dictionaries, such as http://ling.pl/ or http://www.e-dict.pl/plen, to find the Polish equivalents of the words:
• tackle
• offside trap
• free kick
• Champions League
• Header
• goal line
• cross bar
• to sit on the bench
• a straight red card
If you have a problem with finding equivalents in dictionaries, look for parallel texts on Webpages, such as: http://pilkanozna.pl/
(Two homework activities are not included due to the lack of space. They involved: (1) completing a gap-filling exercise with the words studied during the class; and (2) reading a text and answering comprehension questions.)
Appendix 2. Summary of resources used in class
