Corpora in ESP/EAP Writing Instruction Maggie Charles & Ana Frankenberg-Garcia (Eds.). London/New York: Routledge, 2021. 194 pages. ISBN 9780367432348. Data-Driven Learning is widely practised by instructors in ESP/EAP writing instruction. For example, they may ask students to set up a do-it-yourself corpus, analyse the linguistic conventions and moves, and use the results as a guide to writing. This data-driven constructivist approach allows students to interact with the corpus and abstract the rules by themselves, and is thus more effective than a transmissionist approach. Data-Driven Learning has been a growing area of research in ESP/EAP writing instruction. For instance, learner corpora can be used to diagnose students’ problems in their writing to develop a needs-based syllabus (e.g., Laufer & Waldman, 2011) or to monitor students’ longitudinal development of competence to modify instructional design (e.g., Crosthwaite & Jiang, 2017). Expert corpora can be used to abstract the linguistic and stylistic features of successful writings to formulate learning outcomes or set up measures for writing assessment (e.g., Nesi & Gardner, 2018). By combining learner and expert corpora, instructors can compare differences between student and expert writers and take pedagogical actions accordingly (e.g., Cotos, 2014). Those lines of research are continued by a recent monograph edited by Maggie Charles and Ana Frankenberg-Garcia, Corpora in ESP/EAP Writing Instruction. It is the first book devoted to the use of corpora as language resources in ESP/EAP writing. It showcases how different types of corpora can be used successfully to teach ESP/EAP writing in different contexts. The book begins with an introduction, briefing readers on the parameters that can be used to define corpora in terms of the corpus nature and purposes of use. The bulk is divided into three sections, preparation, exploitation, and analysis, which correspond to the three stages of corpus use in writing instruction. It ends with an afterword in which Lynne Flowerdew critically reviews the foci of the contributions in the light of the current literature. The first section is concerned with corpora use in course development. RESEñAS / BOOK REVIEWS Ibérica 43 (2022): 270-276 ISSN: 1139-7241 / e-ISSN: 2340-2784 270 RESEñAS / BOOK REVIEWS Ibérica 43 (2022): 270-276 271 Chapter one is contributed by Benet Vincent, Hilary Nesi, and Daniel Quinn, who share how the British Academic Written English (BAWE) corpus is used to meet students’ writing needs. This specialized corpus is easy-to-use and time-saving for instructors. It meets students’ writing needs because it can help them improve the idiomaticity of their writings, increase their awareness of output expressions, and engage with corpus output and functionality. As illustrations, three cases are presented to demonstrate how the corpus can provide more output than just concordances so that students can get more helpful feedback. A minor drawback of BAWE is that it may contain errors because the corpus consists of unedited assignments from students with different mother tongues at British universities. Its major contribution is that it can solve the problems encountered in previous corpus consultations: it is time-consuming for instructors to provide high- quality feedback and challenging for students to identify target expressions in a corpus when they have no idea of what to search. Chapter two, authored by Geraint Paul Rees, discusses the preparation of academic vocabulary in EAP writing instruction by proposing a phraseological approach to vocabulary list creation and demonstrating why it is more reliable than a generalist frequency-based approach. The statistics from corpus-based experimental operations and three case studies show that collocational patterns and meanings of most academic words vary from discipline to discipline, indicating that they should be considered in the creation of lexical resources. The chapter’s contribution lies in its proposal of a new approach to replace the traditional generalist approach. In the latter, highly frequent lexical bundles are extracted without considering their collocational behaviours and meanings, and this method fails to meet students’ needs in EAP writing. The second section focuses on integrating corpora consultation into ESP/EAP writing. In chapter three, Hsien-Chin Liou and Szu-Yu Liu report on a longitudinal study to explore how L2 college students’ English writing motivation is related to their conceptions of corrective feedback with corpus consultation, which is an important but ignored area of research. The data from questionnaires and interviews suggest that corpus-based feedback enhances students’ motivation for academic writing and improves students’ writing performance. Also, there exist positive associations among students’ motivation, engagement with corpus use, conceptions of feedback outcome, and successful corrections. As an initial attempt to explore the role of motivation in corpus-based writing instruction, it provides empirical evidence for the effect of incorporating corpus-aided feedback in writing courses and points to new directions for future research. Chapter four, written by Katherine Ackerley, investigates first-year college students’ preference for learning approaches and strategy use in learning genre-specific phrases in corpus-based writing instruction. Questionnaire responses indicate that, for first-year students without prior experiences of corpus use, paper-based tasks are more efficient. It suggests that first-year students need more scaffolding before they can handle computer-based tasks with confidence. As for strategy use, the results indicate that high achieving students use more active strategies compared with their low achieving counterparts. The author suggests that low achieving students should be given more training to develop strategies for corpus-based learning. Students’ involvement in corpus-based writing instruction is rarely explored, and this chapter makes a significant contribution because it demonstrates to instructors how to examine students’ learning processes through classroom- based research and use the results to inform instructional design. Reka Jablonkai and Neva Čebron report in chapter five how undergraduate students react to a do-it-yourself corpus in an ESP course. This study fills in a gap in the literature concerning the systematic design of corpus-based ESP courses to develop students’ linguistic competences to meet disciplinary needs. Drawing on relevant research on corpus-based language teaching and computer-assisted language learning, it presents the essential components of designing a corpus-based ESP course for undergraduates (principles, outcomes, and approaches). It elaborates on the authors’ efforts to empirically validate the effectiveness of such a design. Analysis of students’ written reports and questionnaire responses indicate that corpus use is beneficial in terms of ESP writing, vocabulary expansion, and acquisition of discipline-specific terminologies. The results also suggest that the compilation and consultation of do-it-yourself corpora improve students’ understanding and interpretation of corpus use and output. Corpus-assisted analysis of writing is the focus of the third section. In chapter six, Ji-Young Shin attempts to explore to what extent genre types, revision stages, writer characteristics (language proficiency and L1 background), and essay quality (lexical diversity and complexity) are related to the use of stance markers in first-year L2 academic writing. This topic is significant because, although stance use is challenging for novice EAP writers, it is rarely explored in the pedagogical context of EAP writing. The RESEñAS / BOOK REVIEWS Ibérica 43 (2022): 270-276272 RESEñAS / BOOK REVIEWS Ibérica 43 (2022): 270-276 273 results indicate that genre and essay quality (lexical complexity) are significantly related to stance use and that first-year college students have limited ability to diversify stance markers in EAP writing. This study has important implications for EAP writing instruction: since students’ limited knowledge on stance use restricts their ability to effectively and appropriately convey emotions, attitudes and evaluations in academic writing, instructors should explicitly design corpus-based activities to raise their awareness of stance markers and facilitate effective use of stance markers. In chapter seven, Paula Tavares Pinto, Geraint Paul Rees, and Ana Frankenberg-Garcia focus on the use of academic English collocations among Brazilian researchers who write in English for publication. Since non- native speakers’ familiarity with collocational conventions of English academic discourses is limited, the readability and overall quality of their academic writings are affected. Responding to the paucity of research in this regard, the authors investigate whether scholarly articles published in English journals in Brazil are different from those published in international English journals in terms of collocations. The results suggest that common academic English collocations in articles written by Brazilian researchers are limited in number and less diversified compared with those produced by their international counterparts. They imply that Brazilian researchers should expand their repertoire of academic English collocations to increase the readability of their scholarly outputs. The monograph has two minor limitations. One is concerned with how quantitative analysis is reported. Reporting quantitative results is common in corpus-based studies. In this monograph, descriptive statistics are used more often than inferential statistics. In one chapter where inferential statistics are used, the p values are reported while the effect sizes are missing. P values indicate if the correlation or difference is statistically significant between the variables under investigation and suggest if researchers have adequate evidence to reject the null hypothesis that there is no difference or correlation between the variables compared. What p values cannot tell is the strength of correlations or magnitude of differences. Since p values are sensitive to sample size, a large sample size may lead to a p value smaller than .05, even if there is only minor or trivial differences or correlations and the result is not practically meaningful or important (Barry et al., 2016). Therefore, it is necessary for corpus-based educational research to report the effect sizes. They not only indicate to what extent the variables under investigation are different from or correlated with one another, but also allow colleagues to compare the results of similar studies conducted in different contexts across the published literature, as in the meta-analysis of the effectiveness of corpus use in language learning by Boulton and Cobb (2017). Another one is related to the scope. ESP/EAP writing courses aim at scaffolding students to acquire abilities to produce particular target genres. This includes not only micro-level abilities (appropriate use of words, collocations, stances, pronouns, tenses, hedges, and boosters) but also macro-level abilities (discourse-level rhetorical conventions, e.g., rhetorical moves). It seems that most chapters are concerned with the micro-level of writing. The macro-level in ESP/EAP writing is ignored. Since rhetorical conventions are culture- and language-specific, for students speaking English as a second or foreign language, macro-level abilities are equally if not more important because of the interference of their L1s which have different rhetorical conventions at the discourse level. Therefore, how to develop students’ macro-level competence through corpus-based writing instruction deserves attention, as is the case in two recent studies (Casal & Kessler, 2020; Maher & Milligan, 2019). Despite these minor limitations, this volume is a valuable reference for ESP/EAP writing instructors and researchers. It covers both instructors’ use of corpus in developing materials for ESP/EAP writing instruction and students’ responses, attitudes and strategies in consulting corpora in their writing process. By using different types of corpora, namely, learner or expert, specialized or general, or do-it-yourself or ready-made, the seven chapters approach ESP/EAP writing instruction in different contexts from different perspectives. They inform teachers of up-to-date principles, practices, and problems in ESP/EAP writing instruction, contribute to new insights on corpus consultation in ESP/EAP writing classrooms, and point to new directions for this dynamic and growing area of research. It should be essential reading for everyone concerned with ESP/EAP writing instruction. Received 20 February 2022 Accepted 04 May 2022 Reviewed by Xiangdong Li Xi’an International Studies University, Xi’an (China) xiangdong813@gmail.com RESEñAS / BOOK REVIEWS Ibérica 43 (2022): 270-276274 RESEñAS / BOOK REVIEWS Ibérica 43 (2022): 270-276 275 References Barry, A. E., Szucs, L. E., Reyes, J. V., Ji, Q., Wilson, K. L., & Thompson, B. (2016). Failure to report effect sizes: The handling of quantitative results in published health education and behavior research. Health Education & Behavior, 43(5), 518-527. Boulton, A., & Cobb, T. (2017). Corpus use in language learning: A meta-analysis. Language Learning, 67(2), 348-393. Casal, J. E., & Kessler, M. (2020). Form and rhetorical function of phrase-frames in promotional writing: A corpus- and genre-based analysis. System, 95, 102370. Cotos, E. (2014). Enhancing writing pedagogy with learner corpus data. ReCALL, 26(2), 202-224. Crosthwaite, P., & Jiang, K. (2017). Does EAP affect written L2 academic stance? System, 69, 92-107. Laufer, B., & Waldman, T. (2011). Verb-noun collocations in second language writing. Language Learning, 61(2), 647-672. Maher, P., & Milligan, S. (2019). Teaching master thesis writing to engineers: Insights from corpus and genre analysis of introductions. English for Specific Purposes, 55, 40-55. Nesi, H., & Gardner, S. (2018). The BAWE corpus and genre families classification of assessed student writing. Assessing Writing, 38, 51-55.