English Language Teaching and Linguistics Studies ISSN 2640-9836 (Print) ISSN 2640-9844 (Online) Vol. 5, No. 3, 2023 www.scholink.org/ojs/index.php/eltls 142 Original Paper Exploring the Effects of Corpus-Based Cross-Border E-Commerce Writing Instruction for Vocational College EFL Learners Jing Liu 1* 1 School of International Education, Chengdu Polytechnic, Chengdu, P. R. China * Jing Liu, School of International Education, Chengdu Polytechnic, 83 Tianyi Street, High-tech Zone, Chengdu 610000, P. R. China, E-mail: 48087577@qq.com Received: June 22, 2023 Accepted: July 29, 2023 Online Published: August 03, 2023 doi:10.22158/eltls.v5n3p142 URL: http://dx.doi.org/10.22158/eltls.v5n3p142 Abstract This study proposed and implemented corpus-based writing instruction via the use of a custom-made corpus learning platform for learners of cross-border e-commerce English in the context of vocational colleges. It attested the effects of the proposed DDL pedagogy versus conventional PPP teaching practice on 83 e-commerce majors at a vocational college randomly forming a control group and an experimental group respectively. The experimental group received a three-step inductive pedagogical approach enabled by a tailer-made corpus learning platform while the control group was taught by conventional PPP teaching practice. Results indicated that the experimental group outperformed the control group in writing product description which is a language skill essential to cross-border e-commerce operation. Findings of this research can be interpreted as suggestive that it’s worthwhile to apply custom-made specialized corpora and corpus-based teaching pedagogy to facilitate ESP learning for EFL learners at vocational colleges. Keywords DDL, cross-border e-commerce English, writing proficiency, corpus learning platform, vocational college EFL learners 1. Introduction Data-based learning (henceforth DDL) is the predominant model of utilizing corpora for second and foreign language teaching and learning. Ever since its inception in 1991 (Johns, 1991), there have been a substantial body of researches using DDL for different teaching targets, the overwhelming majority of www.scholink.org/ojs/index.php/eltls English Language Teaching and Linguistics Studies Vol. 5, No. 3, 2023 143 Published by SCHOLINK INC. whom were university learners, examples include Sun and Hu (2020), Jablonkai and Čebron (2021), Bruce et al. (2016), Lin (2021), to name just a few. Albeit prolific publications on DDL targeting tertiary contexts, the use of such methodology with EFL or ESL learners at vocational colleges remains rare. An ongoing but certainly not exhaustive collection of research in the area currently brings up only 13 separate publications explicitly stating that the participants are in vocational college context. Of these, only 5 (Hu &He, 2016; Liu, 2016; Lü, 2015; Ma, 2014; Song, 2014) feature empirical study of DDL. Despite very few studies (e.g., Dong et al., 2017; Ma, 2014; Song, 2014) pointing to the benefits of using DDL for vocational college EFL learners, they adopted DDL as originally proposed by Johns (1991) without any methodological modifications to suit learners’ proficiency levels. DDL is still not widely used in vocational college context for a number of reasons. One of them is that DDL is, according to Boulton (2011, p. 572), “the hands-on use of authentic corpus data (concordances) by advanced, sophisticated foreign or second language learners in higher education for inductive, self-directed language learning of advanced usage”, indicating that DDL might not be appropriate for vocational college learners whose English proficiency and learning autonomy are relatively low. Another reason is that no pedagogical frameworks are readily available for integrating DDL into vocational college EFL courses and curricula. Whilst Carter and McCarthy (1995) outlined DDL approach as “Three Is” (Illustration-Interaction-Induction), such modelling demands high level of learner autonomy and corpus literacy, and the effectiveness of it should take into account different factors such as learning contexts and learners’ EFL proficiency level. For learners at vocational colleges, who tend to be lower-proficiency learners, “Three Is” are not best suited and request adaptation to cater to specific needs and aims of vocational college learners. Thirdly, many available corpora were made by corpus linguists for specialists like themselves and not with teachers and learners in mind, requiring considerable levels of linguistic and technological sophistication (Vyatkina & Boulton, 2017). Nevertheless, since vocational school learners are prone to have a visual or kinaesthetic learning style and prefer graphical presentation of learning material with higher level of interactivity, traditional concordancers are likely to be unappealing. Finally, the corpus examples are often of little or no relevance to vocational college learners because they are often genre-based or discipline-specific rather than representative of workplace scenarios. Learners may get discouraged if too many items in the KWIC context are irrelevant. Consequently, to address the limitations of DDL with lower-proficiency learners at vocational colleges, this study aims to: (1) Explore a DDL pedagogical model targeted at the level and needs of vocational college learners by utilizing a custom corpus learning platform. (2) Conduct empirical studies to explore the effects of the proposed DDL pedagogical framework by measuring the changes in learners’ English writing proficiency. www.scholink.org/ojs/index.php/eltls English Language Teaching and Linguistics Studies Vol. 5, No. 3, 2023 144 Published by SCHOLINK INC. 2. Custom DDL Tools The course Cross Border E-commerce English was offered to E-commerce majors in their fourth semester at Chengdu Polytechnic, a vocational college located at the south-western Chinese city of Chengdu. The time allocation for this course is 4 periods per week which lasts for 20 weeks. Learners are expected to learn the fundamentals of and master the language skills needed for cross-border e-commerce operations. Writing is the primary mode of transmission for cross-border e-commerce communication as e-commerce products can only be displayed in the form of pictures and customer query is mainly addressed via exchange of text messages. According to the cross-border e-commerce workflow, the course was structured into seven modules with each one engaging learners in some kind of writing tasks related to cross-border e-commerce operations as shown in Table 1. Week 1 of the course is a general course introduction not involved in the modules. Table 1. Modular Course Content of Cross Border E-Commerce English Modules Module Description Writing Tasks Weeks 1. cross-border e-commerce platform selection Learners know the development status and characteristics of major cross-border e-commerce platforms Write a report on the main features of major cross-border e-commerce platforms. 2-3 2. cross-border e-commerce platform operation Learners know the processes of cross-border e-commerce business Write a report about the complete business processes of a cross-border e-commerce company. 4-5 3. cross-border e-commerce product selection Learners know how to find suppliers and set the price for the product Contact suppliers for a quote via email. 6-8 4. cross-border e-commerce product descriptions Learners know how to write product descriptions that sell Write a product description. 9-11 5. cross-border e-commerce product marketing Learners know the cross-border e-commerce marketing strategies Write a blog post about a cross-border e-commerce store. 12-14 6. cross-border e-commerce pre-sale service Learners know how to provide customer service in the pre-sale phase Respond to customer enquiries about the product via email. 15-17 www.scholink.org/ojs/index.php/eltls English Language Teaching and Linguistics Studies Vol. 5, No. 3, 2023 145 Published by SCHOLINK INC. 7. cross-border e-commerce after-sale service Learners know how to settle disputes and complaints Respond to a negative customer review via email. 18-20 The lexical items, sentence structures and discourse patterns of cross-border e-commerce English have their own specialties under the banner of ESP, and hence pose writing difficulties to learners as their written proficiency is confined to general English and they barely have any encounter with English in a specialized domain. Therefore, “learners should be exposed to authentic text so that they may have direct contact with input data which reflects genuine communication in the target language” (Breen, 1985, p. 63). The benefits of using authentic materials in ESP course have been acknowledged by many researchers (Rogers & Medley, 1988, p. 475; Widdowson, 1990, p. 162; Peacock, 1997, p. 144; Guariento & Morley, 2001, p. 347). Therefore, the Research Center for Application of Big Data on Higher Vocational English Education affiliated with Chengdu Polytechnic, where the author works, developed a parallel textbook corpus called Cross-border E-commerce Corpus for Loose-leaf Textbook Development (CECLOTED) consisting of a collection of authentic texts such as specialized literature, corporate case studies, technical manuals, and data extracted from major cross-border e-commerce websites. This is novel considering that no specialized corpus exclusively applicable to the domain of cross-border e-commerce has hitherto been made readily available, not to mention in bilingual version. We have also built a corpus-based platform (https://search.ctcfile.com) for storing, managing and querying data of CECLOTED (Liu, 2023, p. 137), the size of which is 271,252 sentence pairs. In contrast to traditional static corpus system, the platform allows for adding data to existing corpora and it features the following four major functionalities: (1) Queries can be performed in both Chinese and English, and the results are shown in terms of concordances, texts and terminologies to suit learners’ needs (see Figure 1). Figure 1. Query Function of the Corpus Learning Platform (2) Bilingual alignment can be achieved for learners to add new text data to the platform (see Figure www.scholink.org/ojs/index.php/eltls English Language Teaching and Linguistics Studies Vol. 5, No. 3, 2023 146 Published by SCHOLINK INC. Figure 2. Alignment Function of the Corpus Learning Platform (3) Students’ assignments or teachers’ written feedback in both bilingual or monolingual version can be uploaded to the platform (see Figure 3). Figure 3. Upload Function of the Corpus Learning Platform www.scholink.org/ojs/index.php/eltls English Language Teaching and Linguistics Studies Vol. 5, No. 3, 2023 147 Published by SCHOLINK INC. (4) Simulated buyer-seller interaction is made possible via the individual chat or group chat in the chat box (see Figure 4). Figure 4. Simulated Interaction Function of the Corpus Learning Platform 3. Course Design The overall course design of the Cross-border E-commerce English follows the data-driven learning model proposed by Tim Johns and his colleagues (Johns, 1991; Johns & King, 1991). The logics inherent in DDL is that learners could become detectives to discover linguistic patterns and rules when immersed in samples of naturally-occurring language. However, it is unfeasible for higher vocational college students to have direct interactions with corpora as their English proficiency is relatively low. As Boulton (2010) suggests, activities with prepared DDL materials are more effective for lower-level learners. In this sense, we adopted teacher-constructed concordance tasks, which is the less autonomous version of DDL proposed by Vincent (2013) and provided a structured, scaffolded framework for inductive learning (Jonathan, 2014). The guided DDL is organized into three phases as shown in Figure 5, below. www.scholink.org/ojs/index.php/eltls English Language Teaching and Linguistics Studies Vol. 5, No. 3, 2023 148 Published by SCHOLINK INC. Figure 5. A Three-Step DDL Enabled by a Tailer-Made Corpus Learning Platform (1) Flipped vocabulary learning before class: the learners learn how to search for relevant information following a manual uploaded by teachers to the platform and finish vocabulary exercises by consulting the parallel corpora embedded in the platform (see an example in Figure 6). Teachers then give written feedback to student work, upon which learners can remedy the errors. Figure 6. Sample Vocabulary Exercise for Flipped Vocabulary Learning www.scholink.org/ojs/index.php/eltls English Language Teaching and Linguistics Studies Vol. 5, No. 3, 2023 149 Published by SCHOLINK INC. (2) Guided data-driven learning in class: the learners observe concordance data and perform DDL activities derived from CECLOTED, both of which have been prepared and preselected by teachers before class. Learners then engage in the language discovery process through discussion. Teachers, if necessary, give hints to help learners make rules for the language feature common in or peculiar to cross-border e-commerce English. (3) Writing practice after class: the learners complete writing tasks using the vocabulary and grammatical patterns learned though the previous two stages. Teachers assess and give corpus-informed feedback which facilitates students’ revision to their first drafts. 4. Methodology To date, research on the effects of applying corpus-based platform in teaching and learning cross-border e-commerce English, with higher vocational English learners as the subjects, is still in its infancy. There are studies suggesting that a course spanning a short-term span of 10 weeks do not suffice to enable much progress in L2 proficiency (Lennon, 1995; Politzer & McGroarty, 1985; Rifkin, 2005; Storch, 2009). Therefore, this study investigates the long-term progress spanning 20 weeks in L2 writing proficiency by gauging the learners’ writing quality in the pre- and post-tests. 4.1 Participants The subjects of this study are 83 e-commerce majors from 2 classes in their fourth semester at Chengdu Polytechnic, and have a mean age of 21. Chinese is the first language for all of them. As EFL learners, all subjects had at least six years’ English learning experience before being admitted to the college. Prior to taking the cross-border e-commerce English course, they have studied General English for one freshman year as their compulsory course. One class was randomly assigned as the experimental group (EG, n = 41, including 27 females and 14 males) while the other as the control group (CG, n = 42, including 29 females and 13 males). The same teaching materials, textbook, and assessment scheme were applied to both groups. In order to reassure the linguistics homogeneity, a validated test (Practical English Test for Colleges Level A, PRETCO-LA) to assess learners’ writing ability was administered to the participants in the two groups at the beginning of the semester. The result of the Independent Sample t-Test indicated that there was no significant difference between the EG and CG (p = 0.080). Thus, it can be assured that the participants were comparable in terms of their English proficiency. 4.2 Data Collection For both pre-test and post-test, students of CG and EG were asked to write a product description for shoes which includes a headline, a description paragraph introducing the features, a bulleted list of technical details, social proof and a conclusion calling to action. Two vocational English teachers with more than 10 year’s teaching experience assess the written work of all participants in the EG and CG using an analytic scoring rubric developed by Tribble (1996). The average inter-rater reliability coefficient was reported to be 0.79. Since none of the participants has prior knowledge about English specific to cross-border e-commerce, Week 1 was spent on introducing learners to the course. The www.scholink.org/ojs/index.php/eltls English Language Teaching and Linguistics Studies Vol. 5, No. 3, 2023 150 Published by SCHOLINK INC. pre-test was then conducted in Week 2 and the post-test conducted in Week 20. Whereas the EG received guided DDL instruction with the aid of corpus-based platform, the CG was taught by traditional Presentation-Practice-Production (PPP) method. 4.3 Data Analysis Difference in scores between pre-test and post-test within a group was analyzed for both the CG and EG. The results, presented in table 2, showed that both groups have made improvements to their writing performance as the mean score of the CG increased from 78.488 (SD=6.1580) in the pre-test to 82.917 (SD=4.8727) in the post-test while the mean score of the EG increased from 77.061 (SD=6.9572) in the pretest to 85.512 (SD=3.0913) in the post-test. Independent sample t-test compared the pre-test and post-test mean scores between the CG and EG, which showed that there was no difference between the groups (p=0.335) in the pre-test, whereas a significant difference in favor of EG was identified between the groups (p=0.000). The findings indicated that students’ writing proficiency developed to a greater extent with the implementation of DDL-ized cross-border e-commerce English writing instruction than with conventional PPP pedagogy. Table 2. Descriptive Statistics for the Pretest and Posttest Time Group Mean SD N pretest CG 78.488 6.1580 42 EG 77.061 6.9572 41 Total 77.783 6.5640 83 posttest CG 82.917 4.8727 42 EG 85.512 3.0913 41 Total 84.199 4.2705 83 5. Implication and Contribution Since the aim of this study is to explore the feasibility and effectiveness of adopting corpus-based cross-border e-commerce English writing instruction for lower-efficiency EFL learners at vocational colleges, which is an under-explored area of research, the findings of this study will enrich both the theory and practice of applying DDL in L2 teaching and learning. The current study outlined, implemented and evaluated a pedagogical model suited to the level and needs of vocational college ESP learners, and hence it could provide conceptual recommendations and methodological solutions for those who wish to enhance English language education in the vocational college context by corpus-based means. A major problem for vocational school EFL learners whose English proficiency is at a lower level, is that they lack the cognitive abilities necessary for sophisticated language work, and yet most conventional ESP learning materials are compiled without in-depth investigation into how authentic language is being used in the original context. As such, the corpus-based pedagogy proposed www.scholink.org/ojs/index.php/eltls English Language Teaching and Linguistics Studies Vol. 5, No. 3, 2023 151 Published by SCHOLINK INC. here allows teachers and researchers to take a further step towards a paradigm shift where spoon-fed learners become “language detectives” (Johns, 1997, p. 101). Also, as scant attention has been paid to employing corpus-assisted teaching in the context of vocational colleges, this study will be conducive to raising awareness of and excitement in utilizing corpora and corpus tools to facilitate EFL learning for vocational college learners. 6. Conclusion The present study explored the effectiveness of corpus-based cross-border e-commerce English writing instruction enabled by a custom-made corpus-based learning platform, which focused on the learning needs of less-proficient learners at vocational colleges. Between-groups comparison results showed that the experimental group outperformed learners in the control group which render evidence that it is feasible to adopt DDL in learning cross-border e-commerce English and such findings presumably can be generalizable to other ESP contexts. Albeit the positive findings, the study is not without limitations. First, this study did not investigate the age and gender of participants as potential mediating variables, and hence it is advisable for future studies to explore their probable influence on the effects of corpus-based writing instruction. Second, this study compared the effects of corpus-based pedagogy with conventional PPP practice while much is yet to be known regarding to what extent learners benefit from the DDLized writing instruction facilitated by specialized corpora and accordingly a custom corpus platform as opposed to DDL assisted by other publicly available corpora and corpus platforms. So, future studies are suggested to devise two experimental groups, one working with custom corpora and corpus tools while the other with publicly available mainstream ones. Acknowledgement This research is supported by the “2021 Research Project of Foreign Language Education Reform in Vocational Schools” and the grant reference is [WYJZW-2021-2006], granted by the Advisory Committee of Foreign Language Teaching in Vocational Education, Ministry of Education, P. R. China. References Boulton, A. (2010). Data-driven learning: Taking the computer out of the equation. Language Learning, 60(3), 534-572. Boulton, A. (2011). Data-driven learning: The perpetual enigma. In S. GozdzRoszkowski (Ed.), Explorations across languages and corpora (pp. 563-580). Frankfurt: Peter Lang. Breen, M. P. (1985). Authenticity in the language classroom. Applied Linguistics, 6(1), 60-70. Bruce, M. L., Coffer, P. K., Rees, S., & Robson, J. M. (2016). Write on the edge: Using a chemistry corpus to develop academic writing skills resources for undergraduate chemists. Chemistry Education Research and Practice, 17(3), 580-589. https://doi.org/10.1039/c6rp00005c www.scholink.org/ojs/index.php/eltls English Language Teaching and Linguistics Studies Vol. 5, No. 3, 2023 152 Published by SCHOLINK INC. Dong, P. N., Wu, H. C., & Zhang, Y. F. (2017). The application model of indirect DDL model in English Vocabulary teaching of higher vocational colleges. Journal of Tianjin Vocational Institutes, 19(1), 94-97. Guariento, W., & Morley, J. (2001). ‘Text and task authenticity in the EFL classroom’. ELT Journal, 55(4), 347-353. Hu, Z. H., & He, L. (2016). Corpus-based efficient teaching of English writing in higher vocational college – With the architectural cost major as the study case. Journal of Wuhan Institute of Shipbuilding Technology, 3, 79-82. Jablonkai, R. R., & Čebron, N. (2021). Undergraduate students’ responses to a corpusbased ESP course with DIY corpora. Corpora in ESP/EAP Writing Instruction, 100-120. https://doi.org/10.4324/9781003001966-5-8 Johns, T. (1991). From printout to handout: grammar and vocabulary teaching in the context of data-driven learning. In Johns, T., & King, P (Eds.), Classroom concordancing. English Language Research Journal (Vol. 4, pp. 27-45). Johns, T. (1991). Should you be persuaded: Two samples of data-driven learning materials. In Johns, T., & King, P. (Eds.), Classroom Concordancing (English Language Research Journal 4). ELR, Birmingham (pp. 1-16). Johns, T. (1997). Contexts: the background, development and trialling of a concordancebased CALL program. In A. Wichmann, S. Fligelstone, T. McEnery and G. Knowles (Eds), Teaching and Language Corpora, London & New York: Longman (pp. 100-115). Jonathan, S. (2014). The role of guided induction in paper-based data-driven learning. RECALL, 26(2), 184-201. https://doi.org/10.1017/S0958344014000081 Lennon, P. (1995). Assessing short-term change in advanced oral proficiency: Problems of reliability and validity in four case studies. ITL Review of Applied Linguistics, 109-110, 75-109. Lin, M. H. (2021). Effects of data-driven learning on college students of different grammar proficiencies: A preliminary empirical assessment in EFL classes. SAGE Open, 11(3), 1-15. https://doi.org/10.1177/21582440211029936 Liu, J. (2023). Design and implementation of the development of a corpus-based loose-leaf textbook for higher vocational English learners: Using the cross-border e-commerce operations English as an example. International Journal on Social and Education Sciences (IJonSES), 5(1), 132-152. https://doi.org/10.46328/ijonses.535 Liu, J. J. (2016). A study on the application of data-driven learning for the construction of a knowledge system of higher vocational maritime English. Straits Science, 9, 74-76. Lü, X. Q. (2015). Applying DDL Approach to the Teaching of Polysemous Phrasal Verbs to Vocational College Students. Journal of Suzhou College of Education, 32(1), 94-96. Ma, Z. Y. (2014). An empirical study on corpus-driven vocational college English teaching. Guangdong Vocational Technical Education and Research, 3, 86-88. https://doi.org/10.46328/ijonses.535 www.scholink.org/ojs/index.php/eltls English Language Teaching and Linguistics Studies Vol. 5, No. 3, 2023 153 Published by SCHOLINK INC. Peacock, M. (1997). The effect of authentic materials on the motivation of EFL learners. ELT Journal: English Language Teachers Journal, 51(2), 144-156. Politzer, R. L., & McGroarty, M. (1985). An exploratory study of learning behaviours and their relationship to gains in linguistic and communicative competence. TESEOL Quarterly, 19(1), 103-123. Rifkin, B. (2005). A ceiling effect in traditional classroom foreign language instruction. Data from Russian. Modern Language Journal, 89(1), 3-18. Rogers, C. V., & Medley, F. W. (1988). Language with a purpose: Using authentic materials in the foreign language classroom. Foreign Language Annals, 21(5), 467-478. Song, Y. J. (2014). The application research of corpus data-driven English writing teaching mode in vocational colleges. Journal of Henan Mechanical and Electrical Engineering College, 22(3), 88-91. Storch, N. (2009). The nature of pair interaction. Learners’ interaction in an ESL class: Its nature and impact on grammatical development. Saarbrucken, Germany: VDM Verlag. Sun, X., & Hu, G. (2020). Direct and indirect data-driven learning: An experimental study of hedging in an EFL writing class. Language Teaching Research, 1-29. https://doi.org/10.1177/1362168820954459 Tribble, C. (1996). Writing. Oxford: Oxford University Press. Vincent, B. (2013). Investigating Academic Phraseology through Combinations of Very Frequent Words: A Methodological Exploration. Journal of English for Academic Purposes, 12(1), 44-56. Vyatkina, N., & Boulton, A. (2017). Corpora in language learning and teaching. Language Learning & Technology, 21(3), 1-8. Retrieved from http://llt.msu.edu/issues/october2017/commentary.pdf Widdowson, H. G. (1990). Aspects of language teaching. Oxford: Oxford University Press.