USING A FREE CORPUS TOOL FOR TIME-EFFICIENT FEEDBACK ON ENGLISH AS A FOREIGN LANGUAGE REPORTS

English as a Foreign Language (EFL) learners often struggle with some of the elements and features of text composition, such as conventions, socio-cultural aspects like hedging, degree of formality or synonyms. This paper aims to provide, by way of exemplification, the application of a corpus-aided technique that helps teachers determine whether general task completion has been achieved in order to identify learners’ deficiencies in writing. This method was employed to consider lexical range (i.e. synonyms), cohesive devices and hedging strategies, including modal verbs, in the participants’ written outputs, a total of 93, in a higher education EFL classroom. To this end, the tool LexTutor (Cobb n.d.) was used to explore the corpus. The data gathered have been analysed following a quantitative approach. Findings indicate that, on average, learners’ productions met the indications in the instructions. Nevertheless, there was a tendency to use lower-level or simpler structures and words rather than opting for language exploration, thus prioritising accuracy. The present study raises the possibility that EFL teachers can offer general class feedback on students’ productions promptly and efficiently.


Introduction
Writing is considered one of the basic language skills (Graham 2019;Fitria 2020) that students need to master in order to attain communicative competence successfully (e.g. Cabezuelo Vivo and Pavón 2019; Graham 2019), a skill which is harder for non-native speakers (e.g. Hyland 2008;Furtina et al. 2016;Gomez-Laich et al. 2019). Nevertheless, the Spanish education system often overlooks text composition in language teaching modules despite being part of the national curriculum, leading to undergraduate students being untrained in this respect (Solé et al. 2005). Upon reaching tertiary level education, students are required to produce a host of writing types or genres, namely dissertations, reviews, or reports, among others. As Hyland (2006) and Mauranen (1993) pinpoint in their respective work, writings and, thus, genres can be considered a social construct on the grounds that they follow specific conventions and paradigms in which both the reader and the writer are trained for effective communication. A text genre is a type of text, whether written or oral, with a communicative intent (Bhatia 1993).
When the reader is acquainted with the genre (i.e., text type), certain linguistic patterns are expected. This prepares the reader to anticipate the type of content through the formal features of the text (Hyland 2006(Hyland , 2008. In this vein, students are required to master the recurring patterns and constructs in genres common at university level, such as reports (Gardner and Nesi 2013;Wirantaka 2016), and to compose written texts in the academic context with a certain level of proficiency (Coffin et al. 2003;Castelló et al. 2012;Gardner and Nesi 2013). In other words, high-quality writings on a variety of topics and following the different conventional forms of text types are expected in a higher education context (Sarani and Talati-Baghsiahi 2017). These, in turn, are needed in the labour market (Whittaker et al. 2011;Graham 2019) which is becoming increasingly demanding and competitive in relation to advanced written communication skills (Walkinshaw et al. 2017;Gomez-Laich et al. 2019;Ferretti and Graham 2019). In addition, English proficiency standards have become pivotal by virtue of globalisation and other social factors (Dickson 2009;Carrió-Pastor 2016;Dearden 2016;Mcdougald 2019;Aguilar et al. 2020;Sun and Lan 2021). Thus, writing has progressively acquired a relevant role in the academic world to prepare students for the professional world (Lasagabaster et al. 2014;Bellés-Fortuño 2016;Wirantaka 2016;King 2018). Therefore, some direct instruction on formal features might be needed in the classroom to attain higher writing proficiency skills (File and Adams 2010;Sarani and Talati-Baghsiahi 2017;Graham 2019).
Higher education institutions require students to demonstrate linguistic proficiency and acquired knowledge generally via written text (Coffin et al. 2003;M. H. Chen et al. 2015;Marulanda and Martínez García 2017). However, even though highliterate contexts, such as universities, demand fluent command as well as sophisticated and accurate use of the written language, learners are rarely offered the chance to practise writing skills (Castelló et al. 2012). Instead, academic writing is, most of the time, part of an assessed assignment (Applebee and Langer 2011). In fact, writing skills have often been relegated to the language classroom and neglected in other branches of knowledge, thus leading to the scarcity of direct instruction to develop writing competences, even in primary and secondary education (Graham 2019). That is to say, the student is expected to learn how to communicate exclusively in the language classroom, where instruction and practice for this is usually carried out, but not elsewhere. In consequence, there are high expectations in the students' written outputs, although little guidance is provided.
In the Spanish general education system, as a study by Solé et al. (2005) shows, complex text-elaboration tasks are scarce -albeit key to ascertaining students' knowledge acquisition (Solé et al. 2005). Therefore, direct instruction might be needed to build students' writing skills, such as the construction of their identities as writers (Graham 2019), among others. Special attention has traditionally been paid to essays as the academic writing genre par excellence (Coffin et al. 2003) in which learners express their knowledge, becoming at the same time an assessment tool for learners' achievement in the classroom used to measure accuracy, structure, content, and style. Learners' writing skills and their proficiency and fluency in textwriting can only be measured when a certain level of command is demonstrated in composing different genres (Graham 2019), given that the communicative goal of each text-type is different, as are the syntactic structures, vocabulary and degree of formality, among other features.
Report-writing can be challenging for Spanish learners of English as a Foreign Language (EFL) enrolled in a Translation and Interpreting degree course since they are not required to produce this type of writing at lower-level language courses or during the compulsory education stages. Bearing in mind the elements mentioned above, which are key to successful formal and academic writing, this paper focuses on a corpus-based analysis of student report-writing with the purpose of facilitating lecturers' general feedback provision to students regarding task completion and identifying common shortcomings in text composition at B2 level in an EFL formal higher education setting. This type of pedagogical approach has been gaining momentum over the last decade (Poole 2016). This study aims to contribute to the literature in this field by building and employing corpora tools to evaluate task completion in participants' written productions. A corpus has been compiled from students' reports to analyse the quantitative data used to detect shortcomings. The present article does not advocate a focus on grammatical or lexical inconsistencies. Instead, attention is paid to the use of certain lexical items that determine report task completion, such as synonyms, connectors, and hedges. This investigation has focused on these three items for the following reasons. Firstly, students were requested to use their own words as far as possible, avoiding those provided by the task instructions, to show their vocabulary range, namely their ability to resort to synonymy and other paraphrasing techniques. Secondly, a report is a textual genre characterised by clarity and clear organisation; hence, the relevance of using appropriate connectors. Finally, considering that B2 reportwriting tasks require students to provide recommendations or suggestions, measuring the use of hedging strategies to tone down proposed courses of action and achieve a more persuasive tone was appropriate. This study was conducted with students attending a language-based module at a Spanish university during the COVID-19 pandemic period during which a hybrid teaching methodology was implemented. For this investigation, 93 students participated submitting their reports for analysis. This study might be of interest to teachers who seek to find less time-consuming alternatives for obtaining an overview of their students' performance in order to provide general feedback.

Literature Review
The literature on the importance of writing skills for EFL undergraduate students has highlighted shortcomings in the education system and deficiencies in the classroom, which directly affect learners' production quality in this competence. Many studies have concluded that work is still needed in this area. According to Maralunda and Martínez García (2017), students are unfamiliar with the use of the conventional features of written language. Furthermore, recent studies have shown that when teaching writing skills in the EFL classroom, the focus has traditionally been placed on the accuracy of students' written output (Hyland 2006;Lahuerta 2018) rather than on text analysis, the process of text production or the organisation of ideas (M. H. Chen et al. 2015). Students' lack of awareness of the ins and outs of text composition, along with the lack of in-classroom practice, results in written productions that are not up to higher-education standards (Solé et al. 2005). Previous studies suggest the need to train students in writing competence as they should be ready to meet given rhetorical demands in tertiary education (Poole 2016).
Writing skills go beyond typing ideas into words on a blank document. In order to successfully master the versatile art of text composition, the writer must acquire specific sub-skills related to text elaboration and master specific features governing the genre before composing a coherent and cohesive written whole (Sarani and Talati-Baghsiahi 2017). Not only does this require a high command of writing conventions and text types, but also a high level of language proficiency (Coffin et al. 2003). In order to achieve an appropriate degree of sophistication in written language, the ability to use synonyms and paraphrasing strategies, on the one hand, is essential. These are indicators of students' high command of language production when used accurately (M. H. Chen et al. 2015). On the other hand, accuracy is key for avoiding miscommunication in the same way that complex syntactic structures are pivotal in student text composition in order to show their grammatical and lexical proficiency. However, these two aspects should not be the only focus of student-written outputs, as errors are a sign of learning and progress taking place (Scrivener 2011). Instead, students should be provided with feedback on producing coherent and cohesive paragraphs that are appropriate to the context (Hyland 2006); that is to say, attention should be paid to content, appropriateness, and organisational aspects, including the use of connectors.

Synonyms
Attaining writing finesse requires scaffolding techniques, which, as Gomez-Laich et al. (2019) suggest, entails training and guidance. A number of authors have considered synonyms indispensable for successful communication, not least in second or foreign language learning (Edmonds and Hirst 2002;Liu and Zhong 2016;Soto et al. 2017). Synonyms are not only a way to show one's language knowledge, but they also contribute to the coherence and cohesion of a text (Halliday and Hasan 1976). Using synonyms is of paramount importance for adding "variety" (Bailey 2003: 79) and accuracy to the text. Therefore, considering their value as a communication strategy for learners, their use should be encouraged within the EFL classroom. To put it another way, students need to be encouraged to refrain from using broader terms, advocating specificity instead (Yeh et al. 2007). Nevertheless, learners struggle with the use of synonyms as they are "not fully intersubstitutable" (Edmonds and Hirst 2002: 107) but rather laden with nuances (Edmonds and Hirst 2002). Synonyms in language require precision. Such intricacies of language pose a challenge to B2 learners since their language knowledge is often too limited to find the most accurate word for the context (Soto et al. 2017).

Connectors
According to the literature on teaching writing to EFL students, another aspect to take into consideration in text composition is cohesive and organisational patterns, in which connectors may have a key role. Previous research findings have proven this to be a problematic area for EFL learners (Hyland 2008;Carrió-Pastor 2013;Özbay et al. 2019) as it is concomitant with socio-cultural aspects (Mauranen 1993;Hyland 2008), i.e. traditions and patterns inherent to a specific culture and therefore susceptible to change across cultures. For instance, comparative literature suggests that connectors (e.g. "nevertheless", "furthermore", "moreover") are abundant in English texts as opposed to other languages (Hyland 2008). One possible explanation behind this might be that English places the onus of getting the message across on the writer rather than the reader and their ability to interpret or decode the message (Hinds 1987in Hyland 2008. In regard to student text production, previous studies show that the presence of connectors is either insufficient in EFL learners' compositions or inaccurate when an attempt to include them is made. Jiménez Catalán and Ojeda Alba (2014) based their study on the use of connectors in letter writing in order to provide a diagnosis of the difficulties learners encountered using them in this type of text. The task proposed to participants in the study was to write an informal letter to a possible future host family in the UK. They concluded that connectors were either scarce or inexistent, inadequately used or irrelevant (Jiménez Catalán and Ojeda Alba 2014). Furthermore, research by Yang et al. (2012), found that EFL students did not use as many connectors in their essays when compared to native speakers' written outputs, or else overused or underused certain types of connectors.

Hedging
Several authors have turned their gaze towards modal verbs as hedging strategies in English, a socio-cultural aspect of the written language, with which EFL learners tend to struggle (Hyland 1996a;Hinkel 2005;Fraser 2010;Neary-Sundquist 2013;Yagız and Demir 2014;Demir 2018). The presence of hedges, such as modal verbs (e.g. "would", "could"), is of considerable significance in English writing for mitigating arguments (Hyland 1996a;Hinkel 2005;Ge 2015) or sounding persuasive (Hyland 1996a;1996b) and, at the same time, establishing a cordial bond with the recipient of the text (Hyland 1996b;Neary-Sundquist 2013;Ge 2015). Ge's (2015) study shows that non-native writers overuse certain modal verbs, such as "should", and do not use "would" and "may" with the same purpose as native English speakers.
Taking into consideration the difficulties that EFL learners face when producing written outputs, namely synonymy, cohesion, and the use of modal verbs as a hedging strategy, this paper aims to answer three research questions (RQ): 1. What use is made of varied lexical items to provide an appropriate answer to the report-writing task?
2. How frequently did learners use cohesive devices in their written outputs?
3. How frequently were modal verbs used as hedging strategies in students' reports?

Participants
The corpus-based technique detailed below was applied during the second half of the academic year 2020-2021 to an EFL class of students doing a Translation and Interpreting degree at Universitat Jaume I (Spain). The 93 learners enrolled in this second-term course were first-year undergraduate students who had been exposed to English throughout their primary and secondary education stages as well as during the first semester at the university. Thus, an upper-intermediate level of the language was expected in both their production and comprehension skills. In fact, according to the course plan, students must attain a B2 level in order to pass the module. No language level test was performed, as the results from the previous semester served this purpose. The majority of students were at the expected level, with the exception of seven students who had a lower level and four students who had a higher level. The learner-participant body was mostly of Spanish nationality, except for two international students of French origin.
Over the years in which they had been schooled in the compulsory stages of the Spanish education system, the students had been acquainted with the production of different text genres in English, such as essays (Coffin et al. 2003), articles, notes, or letters. Nevertheless, the learners first encountered report-writing in their first year in higher education. In designing this study, I drew from the premise that learners had an intrinsic motivation towards all aspects of language and in developing their language skills, including their writing competence, considering they had enrolled in a language-based degree of their own volition. Given that this study was conducted towards the end of the academic year, students were already -or were expected to be-familiar with the different types of writing established in the course syllabus, namely essay, article, review, letter, and report writing, as presented in the textbook Expert First (Bell et al. 2016), including the style and tone the writer needs to adopt in each of these categories. At this stage of the course, students had already worked with synonymy, hedges, and connectors.

Task Procedure
Considering students had composed just one report thus far on the amenities their city has to offer, further direct instruction and guidance were of paramount importance to aid them in attaining good-quality written outputs in this genre. Therefore, as a first step towards report-writing, language related to the topic was practised in the course materials. Secondly, participants were furnished with a short instructional video on the institutional virtual classroom on writing a report and aspects to bear in mind, emphasising how this genre differed from articles or essays. They were reminded that a report needs to include collected data, for example, from interviews or surveys, and the information needs to be organised in sections with subheadings. Thirdly, syntactic structures and phrases inherent to this text type were taught and reviewed as part of the course materials, including passive voice structures. Learners were expected to include adverbial clauses of reason, cause and result, and passive structures. Furthermore, a sample report was supplied to the learners for analysis and guidance prior to student report production. Fourthly, participants were encouraged to design a plan and a draft of their writing. Finally, after having studied the layout and structure of reports and common linguistic patterns appropriate for this type of formal writing and having organised their ideas on a separate paper, students were asked to write and submit a report on a different topic and with a different informative objective. In the section below, further details on the task learners had to complete are given.
The task students needed to complete consisted in analysing the benefits and drawbacks of welcoming a group of international students to our college, coupled with a final recommendation statement ( Figure 1).
As part of the writing process, and to ensure task completion, students were encouraged to focus on the keywords in the task. Underlining those keywords was deemed necessary so as not to fail to include any of the elements requested; otherwise, they could not attain the highest mark in task completion (see Figure  2), which requires using a "range of vocabulary, including less common lexis". Those keywords or phrases are "group of students from a different country", "report", "advantages", "disadvantages" and "recommend". With these keywords in mind and in sight, brainstorming techniques via the class forum were promoted to furnish learners with possible ideas to include in their written reports, which replaced live interaction in the brick-and-mortar classroom. Moreover, participants were encouraged to include plausible survey results to adopt a factual tone, in line with the textual conventions of this specific genre.
Furthermore, learners were recommended to consider synonyms for those keywords provided by the task to show their "range of vocabulary". With the aim of seeking variety in terms of lexis in learners' written production, students were allowed and invited to use thesaurus and collocations dictionaries. The purpose of this was three-fold. Firstly, as a way to illustrate their understanding of the words given; secondly, to show a wide range of vocabulary; and thirdly, to demonstrate their ability to employ vocabulary-coping strategies in their written productions. These three aims are part of a bigger and broader objective: showing the students' language command and proficiency (Takač 2008;Soto et al. 2017). To that end, synonyms, when used correctly, prove students' ability to produce texts in the target language, as detailed in the introduction of this paper.

Data Collection
Data were obtained from the participants' reports. Submissions of their respective compositions were to be made in an editable text format, namely a .docx or .odt file, via the virtual classroom for later study and analysis. As a result of the considerably high enrolment rate in the subject, on the grounds that this English Your college has been asked to accept a group of students from a different country for a couple of weeks. Your teacher has asked you to write a report. List the advantages and disadvantages of having these visitors at the school and say whether you recommend it. Write your report in 140-190 words in an appropriate style. course is compulsory to obtain the above-mentioned undergraduate degree, 93 reports were submitted. The aim of collecting these samples was to compile a corpus generated from students' texts for later analysis, which is detailed in the paragraph below. This corpus would be used to reveal the degree of task completion and provide general class feedback on the participants' work.

Corpora Generation
In order to analyse learners' written productions, building a corpus was deemed appropriate. For that reason, all participants' submissions had to be converted into plain text files, i.e. .txt, as this was the only format compatible with the free-access tool used for such purposes, namely LexTutor (Cobb n.d.). Subsequently, all the converted files were uploaded to the aforementioned web-based tool for list generation. As a result, two lists were generated: one for word frequency and another for word concordance. The former was designed to obtain numerical values, whereas the latter provides examples of the words used in context. The section below focuses on analysing and interpreting the numerical values.

Results
This section shows the quantitative analysis of the students' written outcomes concerning content words and lexical variety, connectors, and modal auxiliary verbs used as a hedging strategy. The present study aims to present a less timeconsuming alternative to obtain an overview of the students' performance in order for teachers to provide general feedback. This research article intends to do so by providing a general diagnostic on task completion by means of the compilation of a corpus; thus, focusing on a qualitative approach, the analysis of the data collected will provide a way of obtaining a general overview of students' performance that will help teachers to provide general feedback in a time-efficient manner.
Three tables were constructed from the first list created based on word frequency. First, Table 1 summarises the most relevant findings related to synonymy and content achievement in regard to task completion. Second, Table 2 illustrates how frequently connectors were used in the students' reports at this level, which also serves to analyse learners' likely shortcomings regarding organisational patterns. Third, Table 3 summarises the use of hedging modal verbs in the students' writings. Table 1 below highlights the most frequent content words present in the participants' written reports, more specifically, the 37 most frequently repeated words related to the topic provided in the task, which is detailed in the methodology section of this paper. The table shows a selection of the words that were deemed relevant for the purpose of this study. The numbers related to frequency reveal that the student reports addressed the task provided in the course materials. For instance, if the task required writing a report on visiting "foreign students", the use of the words "foreign", "country", or "international" in the text would show that students indeed stuck to the task. However, if the teacher's instructions stated that students should use their own words as far as possible, high-frequency indicators of the word "foreign" reveal that most students did not resort to synonymy or paraphrasing, but instead used the word provided by the task. Nevertheless, those who use the word "country" or "international" would have followed the teacher's guidelines. The word "students" appears in first position the first table, followed by the word "foreign". Although "foreign" comes in second with 171 appearances, it is far behind the first one, i.e. "student(s)", which is present 528 times. The position of these two could have been predicted, as they take an obvious spot within the list considering the topic. The second one, albeit very common, does not always appear in concordance with the word "students". The adjective "foreign" appears as a modifier accompanying "students" in most cases, 140 times out of 171; however, at other times it collocates with "country", "countries", "culture", "cultures", "group" or "language". Considering that the phrase "group of students from a different country" was provided in the task instructions, students were expected to produce other phrases that fit the context and meet the task requirements such as "visiting students".

Synonymy
Furthermore, the presence of the word "foreign" in the above-mentioned word combinations reveals task achievement as well as the use of synonyms and paraphrasing strategies. In those cases in which the phrase "different country" can be read in the course materials, "foreign" was provided as a synonym. Concerning alternative wording for that phrase, the last word in Table 1 is "international", which also shows learners' ability to rephrase concepts. To put it another way, it demonstrates students' ability to resort to synonymy to show their command of the language and avoid repetition. Nevertheless, the number of times "international" was used is well below "foreign", as the former was only used 13 times and the latter 171.
Other examples worth focusing our attention on are the nouns "college", "experience", "language", "country", "University", "culture", "exchange", "opportunity", or "recommendation" along with some of their respective derivative forms, since they reveal that participants did not deviate from the task. For instance, the word "recommendation", which coupled with its plural or verb forms, shows that learners included their advice in the task, as requested. Nevertheless, a wide range of vocabulary has not been shown in many cases despite the teacher's instructions to avoid repeating the words provided by the task. The high frequency of derivative words from the verb and stem "recommend" is a case in point. Furthermore, learners had been furnished with a variety of structures to express a recommendation, such as "We have no hesitation in recommending" or "It would be advisable for X to", amongst others, which were not as frequently used as had been anticipated. Only four students included the latter in their reports. Instead, participants preferred structures they were confident with, i.e. "recommend + ing/noun/that-clause", thus showing a preference for accuracy rather than variety and complexity (see Appendix 1). Three examples (1-3) are included below by way of illustration: (1) I totally recommend accepting the group for two weeks. (item 11174 1 ) (2) I would recommend admitting the 50 students because in spite of the space… (item 11182) (3) I would recommend that you accept this group because it is very unlikely… (item 11184) Another notable example of synonymy and wide range of vocabulary is the use of words such as "benefits", "positive [aspect(s)]", or the adjective "beneficial" to refer to those "advantages" that the task required participants to include. By the same token, the word "university" replaces "college" or "school" in the instructions to avoid repetition. Nevertheless, it can be observed in Table 1 that the word "college" was still used more often than "university" or other words that could have been employed in this context, namely "faculty" or "institution", to name a couple.
Closer inspection of the lexical items in the table reveals that those words at the top of the list, which are the most frequently used, such as "students", "foreign", "advantages", "disadvantages" or "college", are all provided by the task. This indicates that most students did not show a wide range of vocabulary by replacing those words with synonyms in their written productions. Only a small number of participants employed alternatives like "benefits", "beneficial", or "drawbacks" to express the same concepts. The words "recommend", "recommendation", "recommendations", or "opinion" show that most students included suggestions as requested by the task. However, few used alternative lexical equivalents.
As for report conventions in relation to structure, the fact that the words "aim" and "purpose" are represented among the words in Table 1 reflects that the learners followed the instructions provided in class regarding report organisation, which requires stating the objective of the writing at the outset. Interestingly, even though "main" and "aim" were used with the same frequency, which to a certain extent indicates the correlation between adjective and noun, the adjective "main" also appears accompanying the words "purpose" and "benefits". This reveals learners' lexical awareness of collocations.

Connectors
The analysis of the results regarding the use of discourse markers, summarised in Table 2 below, reveals a higher preference for simpler connectors than complex ones.
The overuse of "because" and "also" is one issue worth highlighting. According to the English Vocabulary Profile (Cambridge English Language Assessment et al. 2012), these two cohesive devices are two basic connectors frequently employed in both formal and informal contexts. While this might be a logical explanation for their overuse, it brings to light the need for in-class instruction and practice to raise awareness of the importance of the variety of cohesive devices in academic discourse. Such overuse raises the question of whether the learners were indeed acquainted with a wide range of linkers. Were that to be the case, this prompts another question: whether their choice lies in their self-confidence or a lack thereof. However, the next group of connectors in the table seems to be far less common in contrast to the first, although "moreover" and "according to" are still amongst the most frequent.
Formal addition connectors, such as "moreover" or "furthermore", seem to be two of the preferred linkers for such purposes, whereas "in addition" and "besides" lag behind on the list. Given that the tone in a report should be formal, the use of "besides" should be discouraged. This might be something to point out in general class feedback. In contrast, "moreover" and "furthermore" reveal that the register in the learners' reports was generally appropriate.
Regarding causal clauses, participants were more inclined to use "since" than "due to" or even "owing to". Therein lies another example of an absence of grammatical ambition, as these are the structures that had been presented in the unit. In other words, students showed a propensity for including those cohesive devices seen at lower levels, such as "since", to express reason in B1 (Cambridge English Language Assessment et al. 2012). Therefore, it is not surprising that those connectors in the class materials, namely "despite", "thus", "consequently", "in spite of", "owing to", or "as a result" were not as frequently used in studentwritten productions as anticipated. Yet, to express contrasting ideas, the participants' favourite structure was "on the one/other hand".
Concerning both task achievement and genre features, the fact that learners chose "according to" shows that they included external sources to expound and back their arguments, for instance via surveys or interviews (see Appendix 2). Sentences 4-6 contain examples of these uses extracted from the corpus: (4) according to most students [,] accepting the foreigners… (item 479) (5) according to the opinions of our students accepting overseas students … (item 491) (6) According to studies conducted … (item 487) Along the same lines, the fact that "such as", "for example", and "for instance" figure in the list demonstrates that learners resorted to exemplification in some cases. In analysing the use of cohesive devices, a need for genre-based further practice to build on learners' self-confidence in the language comes to light. The evaluation of the results obtained from the compiled lists, as illustrated in Table 2, reveals that students aim for accuracy when making their choices by opting for most familiar words or structures rather than newly learnt ones. In doing so, simpler cohesive devices are used. Lackadaisicalness or half-heartedness in writing might result in the opposite effect since their outputs might not conform to the level of the course syllabus. Learners should be encouraged to use a wider variety of connectors, alerting them to "discourse marker overkill", as Milton and Tsang (1993) put it. Learners should be guided to acquire the ability to express arguments logically, for which connectors are necessary. They are meant to facilitate communication, but their overuse -or underuse-might hamper this. To ultimately master this skill, students need to find the right balance in the use and choice of cohesive devices (C. W. Chen 2006), which is conditioned by the text type they are writing.

Hedging: Modal Auxiliary Verbs
Further analysis of the corpus, which was conducted manually, showed the presence of modal auxiliary verbs used as hedging strategies in the participants' written productions. Hedges express uncertainty or inconclusiveness, sometimes deliberately as a means to sound cautious and show politeness (Hyland 1996a).
Considering that a report is usually written for a superior, modal verbs to express deliberate tentativeness are expected. Nevertheless, some modal auxiliary verbs convey a high degree of certainty, namely "must" or "cannot/can't", used to express strong opinions or high probability. In Table 3 below, a low-frequency rate is reported for these two, which indicates that students generally meet the genre conventions. Along the same lines, a high-frequency rate of "would" and "could" reveals that participants employed a cautious tone to achieve the formal style required (see Appendix 3). Sentence examples 7 and 8 below extracted from students' productions illustrate the use participants made of "would" and "could": (7) This report aims to explain all the long term benefits that the university could achieve if a group of fifty students from different countries… (item 3510) (8) the students as it would be a memorable experience for them and would also favor tourism in our city and its surroundings (item 17398) Participants seemed to be less confident in the use in their reports of "can", "may" or "might", in that order. The difference in the frequency of use between "can" and "might" is considerable. The reason behind this might lie in the level of familiarity learners have with the language, as occurred in the case of the other elements analysed. "Can" used to express possibility falls within the A1 level contents in the CEFR, as compiled in the English Vocabulary Profile (Cambridge English Language Assessment et al. 2012), while "may" and "might" are labelled as B1.

Discussion
In light of the results detailed in the previous section, the findings indicate that, on average, students attained successful task completion in line with the guidelines and assessment rubrics provided, as they generally followed the instructions provided by the teacher and included relevant keywords related to the topic. Yet, even though the participants occasionally resorted to synonyms to show a wide range of language, synonymy was not as evident as expected or instructed (RQ1). Therefore, additional guidance is needed in this respect. Furthermore, whilst the students' written works did include connectors to link and arrange their ideas in a logical manner, the range was limited in so far as simple linking words were preferred (RQ2); thus, the variety of connectors seen in the course materials were not included, which would have been more in line with the proficiency level expected and with this type of formal report-writing. This might be due to the limited length of the task or the text type students were expected to compose. It can be observed that these findings in relation to the use of cohesive devices tally with previous studies, which reveal learners' difficulty in employing them in EFL contexts (C. W. Chen 2006; Lahuerta 2018), namely in formal writings such as essays (Hamed 2014). In line with results reported by Milton and Tsang (1993), our findings conclude that emphasis needs to be placed on the role of connectors in writing. The importance of connectors goes beyond the language classroom. Attention to this aspect of learners' communicative competence should be paid in all fields of knowledge to enhance learners' argumentation ability (C. W. Chen 2006) and to meet the requirements of high-literacy contexts (Mauranen 1993;Solé et al. 2005;M. H. Chen et al. 2015;Marulanda and Martínez García 2017;Hyland and Jiang 2017). To answer RQ3, the fact that the participants' written outputs contained modal verbs and hedging strategies indicates that the degree of formality and the use of tentativeness was generally correct.
All in all, these results seem to be consistent with those obtained in other research studies, which found that writing in the EFL classroom tends to be an area that requires special attention (Hyland 2008;Gomez-Laich et al. 2019), especially in regard to (a) the relevance of lexical richness (Edmonds and Hirst 2002;Liu and Zhong 2016;Soto et al. 2017;Yeh et al. 2007) and (b) cohesive argumentation and presentation of ideas (Hyland 2008;Carrió-Pastor 2013;Jiménez Catalán and Ojeda Alba 2014). Contrary to expectations and the findings of other studies in the field indicating that non-native speakers of English find hedging structures harder (Hyland 1996a;1996b;Hinkel 2005;Neary-Sundquist 2013;Demir 2018), this study found that many students attempted to use them (see Appendix 3). Nevertheless, further analysis of the lists compiled should be conducted to determine whether more modal verbs could have been used and whether they were used to good effect. Another finding that should be reported, as a preliminary hypothesis stemming from this study, is that learners underperform in their writings in their search for accuracy at the expense of greater sophistication.

Conclusion
This paper set out to evaluate task completion taking into consideration the participants' ability to use synonyms, connectors and hedging strategies. The study contributes to the field of corpus linguistics applied to pedagogical practices by applying a corpus-based method to verify task completion and achievement by first-year undergraduate students when writing reports at the B2 level.
The methodology employed in this study and its results revealed that by analysing instances of students' performances in a report-writing corpus, lecturers could improve their provision of general feedback, and students' writing skills could be enhanced. The corpus-assisted technique detailed in this paper endows teachers with a labour-saving and efficient method, which is a time-saver as it offers a general view of students' performance. This, in turn, would benefit learners, as teachers' feedback and guidelines to improve their students' writings would be furnished in a timely manner. Taken together, the results of this research support the idea that instruction on writing techniques is necessary in the EFL classroom at all levels, not least in higher education, in order to meet the particular lexicogrammatical demands of each genre (Carrió-Pastor 2013; Marulanda and Martínez García 2017; Riaz and Akhtar 2019).
This study, however, is not without its limitations. Firstly, the most significant limitation lies in the paucity of written outputs collected. A corpus of 93 writings is not considered sufficient to determine whether the results obtained apply to all B2 EFL learners within the Spanish education context, or whether they are limited to summarising the performance of the specific group of students participating in the study. Further work needs to be done in this respect.
Nevertheless, the study suggests that the methodology described can be relevant for language teachers with a large number of students in class. Secondly, it is unfortunate that the study did not consider students' report production after the teacher's comments and feedback. Future studies should focus on the students' engagement with the feedback and compare the first submission and the second after feedback provision as well as the grades obtained.
Given these findings and limitations, a natural progression of this work would be to analyse the participants' subsequent reports, evaluating to what extent the feedback and guidelines provided by the teacher are followed. Such further research should also explore whether the second written outcomes are of higher rhetorical quality and in line with tertiary education standards. The findings above provide the hypothesis for future research that EFL learners prioritise accuracy over complexity, sophistication, and exploration of a broader linguistic repertoire in their written performances, thus favouring simpler structures and more common words.
Notes 1. All examples provided were taken from the corpus compiled from students' texts.