Reference this

Lin, M. (2016). Effects of corpus-aided language learning in the EFL grammar classroom: A case study of students’ learning attitudes and teachers’ perceptions in Taiwan. TESOL Quarterly, 50(4), 871-893.


Numerous studies attempt maximising learners’ English proficiency with the aim of identifying the best practices to increase acquisition and development in second-language learning. One approach that has attracted significant attention in recent years is the corpus-based approach. A corpus is defined by Reppen (2010) as “a large and principled collection of naturally occurring texts (written or spoken) stored electronically” (p. 2). This approach has impacted second-language teaching and acquisition significantly. Strong research suggests that when learning an L2, students have a greater colloquial understanding of a language, or can express their exact meaning more clearly, if they do not learn about grammar rules in isolation, but within the context in which those rules are used. Thus, data-driven learning DDL facilitates learning and teaching either grammar, collocation, vocabulary, or writing and develops autonomy, motivation, and confidence.


Lin’s study was conducted in Taiwan by applying a blended approach, namely a mix between Data-Driven Learning (DDL) and traditional deductive approach (TDA). The aim was to examine how Taiwanese students and teachers in an ELF classroom learn and teach passives and relative clauses. The research question asked whether DDL, compared with TDA, was considered to be more pedagogically suitable and more effective in promoting learning motivation and self-efficacy.

Get Help With Your Essay

If you need assistance with writing your essay, our professional essay writing service is here to help!

Find out more

Lin used a mixed method to collect data. Quantitative method was used to collect data from 52 first-year undergraduate students (aged 18 to 19 years old) who were majoring in the English language. He used three classes; one class applied 100% TDA, the second class applied 40% DDL and 60% TDA, the last class applied 60% DDL and 40% TDA. Qualitative method was used with 14 early-career teachers (ECTs); the ECTs were all postgraduate students, and at the same time, they were teachers of English to undergraduates in Lin’s experiment.

After using a paired-sample t-test, the quantitative results showed that only one out of three classes, which received 60% DDL instruction, had shown improvement in students’ learning attitudes while the qualitative results from the teachers showed that the applying DDL was interesting and innovative experience and could inspire the students to be more independent about learning grammar. Lin recommends applying DDL more widely in teaching and learning grammar.


The article has clear subsections that effectively organise the content. The analysis is clear, and Lin compares his study with references to other researchers in this field by demonstrating the findings of other empirical studies. He supports his analysis and findings via tables and figures that clarify his points.

Structure and Methodology

The article has clear subsections that effectively organise the content. The analysis is clear; and he supports his analysis and findings via tables and figures that clarify his points.

The methodology and procedure of Lin’s experiment, including the number and level of participants, are relatively similar to other studies that I will compare with. All the studies typically have one or two treatment/focus group(s) and one control group. All supporting studies in this paper have compared traditional inductive/deductive teaching approaches with DDL approach. However, Lin’s study is unique in terms of the duration of both the course for the student participants and the training session for teacher participants on how to apply DDL in the classroom (see Table 1). The longer durations affect the validity of Lin’s study negatively as explained in the validity section below.

Table 1: Differences in the duration of the DDL courses between Lin and other researchers.

Study Course Duration
Lin (2016) Students: 3 weeks (9 hours in total)

Teachers: 2 weeks (6 hours)

Akinci and Yildiz (2017) Students: 5 weeks (15 hours)
Barabadi and Khajavi (2017) Students: 7 weeks
Huang (2012) Students: 12 weeks
Oghigian and Chujo (2010) Students: 20 weeks (30 hours)
Farr (2008) Teachers: 24 weeks (84 hours)
Mukherjee (2004) Teachers: 1 year

Data collection

Lin used mixed method research (MMR) to collect data. According to Dorney (2007), MMR helps develop a full understanding of a complex issue by “looking at it from different angles” (p. 164). Lin’s first method was the quantitative method of 16-question questionnaire through using a 5-point Likert scale. Undergraduates completed entry and exit questionnaires. The second method was an open-ended (semi-structured) interview. This qualitative method was directed at the teachers, the postgraduate.

Table 2: Comparison of Lin’s and other researchers’ studies.

Researcher(s) Pre-/post- test


Interview (semi-structure)


Entry/exit Questionnaire

Likert Scales


Lin (2016) MA-Teachers UG-Students
Akinci and Yildiz (2017) UG-Students UG-Students UG-Students
Barabadi and Khajavi (2017) UG-Students
Huang (2012) UG-Students UG-Students
Oghigian and Chujo (2010) UG-Students
Farr (2008) MA-Teachers
*Mukherjee (2004) Experienced Teachers

*Mukherjee used multiple answers and multiple choice in the questionnaire.

Other researchers have either employed different data collection methods or applied the same methods to their participants in different ways comparing with with Lin (see Table 2). The above table shows the following:

  1.  All the studies focused on one sample, either students or teachers.
  2. All undergraduates took a pre-test or/and post-test to confirm the effectiveness of DDL. The importance of conducting these tests was to measure to what degree the students had increased their acquisitions through DDL as a means of learning.
  3. Farr (2008) and Mukherjee (2004) had teachers complete entry and/or exit questionnaire(s) as to whether the teachers perceived DDL to be an effective tool to teach in the English classroom as compared with traditional approaches.

Similar to Lin, as shown in Table 2, Akinci and Yildiz (2017) and Huang (2012) conducted an exit questionnaire on their students using the quantitative method. In contrast, Mukherjee asked the teachers to complete entry and exit questionnaires with different scales (with multiple-answer and multiple-choice items). Using the qualitative method, Akinci and Yildiz used a semi-structured interview (open-ended questions). However, while Akinci and Yildiz interviewed the students, Lin interviewed the teachers.

Data analysis

Lin analysed his data applying a paired-sample t-test to the questionnaire data. First, he used multivariate analysis of variance (MANOVA) for the entry questionnaire. Then he used multivariate analysis of covariance (MANCOVA) for the exit questionnaire, including the scores of the entry questionnaire as covariance.

Find out how can help you!

Our academic experts are ready and waiting to assist with any writing project you may have. From simple essay plans, through to full dissertations, you can guarantee we have a service perfectly matched to your needs.

View our services

In contrast, Akinci and Yildiz and Barabadi and Khajavi (2017) used an ANOVA one-way independent paired-sample t-test for each pre/post-test to have separate results. Huang used (ANOVA) two-way t-test. She concentrated on the independent variable within one group (pre-test and post-test) and between groups (focused and controlled group). Farr and Huang used percentages to present the results of their questionnaire; Oghigian and Chujo (2010) applied a percentage to their pre-test and post-test to present their results.


The findings of the other researchers aligned with Lin’s findings. DDL as a learning tool had a positive effect compared with traditional teaching approaches. The DDL treatment groups outperformed controlled groups taught traditionally with textbooks and dictionaries.


Lin conducted a pilot study with 68 participants after revising the questionnaire for validity and reliability. Cohen et al. (2017, p. 496) stressed the importance of applying a pilot study “to increase the reliability, validity, and practicality of the questionnaire”. However, some validity issues remain in Lin’s study compared with the other studies.

  1. Lin did not apply a pre-test and post-test to measure students’ knowledge and grammar acquisition or to what degree DDL was an effective approach. Other researchers, as shown in Table 2, did conduct both tests. Mackey and Gass (2005) noted that pre-tests “ensure comparability of the participant group prior to their treatment” whereas post-tests “measure the effect of treatment” (p. 148, 149).
  2. The duration of the programme was insufficient to apply DDL appropriately for both students and teachers. Other studies, as indicated in Table 1, had a longer period.
  3. Lin allowed early-career teachers to teach the undergraduates. Farr’s and Mukherjee’s studies help to highlight why this choice is problematic:
  1. Lin noted that his teacher participants complained about the added workload of this study, which had to be completed in a very short period of time, created for them. The teachers were already completing their master’s degree. Farr’s and Mukherjee’s programmes were long enough for a teacher to practise using corpora.
  2. Lin asked the teachers to design DDL practicums before the teachers had adequate experience in designing ELT materials. The teachers reported technical issues due to inexperience with corpora in the interviews. According to Akinci and Yildiz, DDL is still a new method to many teachers and learners, so corpora can initially seem ambiguous. Recognising their participants’ lack of experience, neither Farr nor Mukherjee asked teachers to design a practicum. Instead, their studies tested an introductory course and can, therefore, be beneficial for many teaching and learning purposes.


Lin identified some of the limitations of his study. Although he notes that his sample was small, the samples in the other studies are only slightly bigger or are even smaller than Lin’s. More notably, he notes that his questionnaire failed to ask students their learning preferences (either DDL, TDA, or both), and his interviews failed to ask teachers if they would use DDL to teach complex or simple grammar rules. The study also produced no evidence as to whether DDL helps to increase grammar acquisition. Lin advises that future research use a pedagogical corpus, which is easier to use (concerning finding suitable concordance lines for students) and more appropriate for teaching and learning purposes than regular corpora.

Number of the words: 1570


  • Akinci, M., & Yildiz, S. (2017). Effectiveness of corpus consultation in teaching verb+noun collocations to advanced ELT students. Eurasian Journal of Applied Linguistics, 3(1), 91-109. Retrieved from [Accessed December 12, 2018].
  • Barabadi, E., & Khajavi, Y. (2017). The effect of data-driven approach to teaching vocabulary on Iranian students’ learning of English vocabulary. Cogent Education4(1), 1-13. Retrieved from [AccessedDecember 9, 2018].
  • Cohen, L., Manion, L., & Morrison, K. (2017). Research methods in education. London, England: Routledge/Taylor & Francis Group.
  • Dorney, Z. (2007). Research methods in applied linguistics. Oxford, England: Oxford University Press.
  • Farr, F. (2008). Instruction in a language teacher education context: Perspectives from the users. Language Awareness, 17(1), 25-43. Retrieved from [Accessed December 12, 2018].
  • Huang, L. (2012). The effectiveness of a corpus-based instruction in deepening EFL learners’ knowledge of periphrastic causatives. TESOL Journal, 6, 83-108. Retrieved from [Accessed December 13, 2018].
  • Mackey, A., & Gass, S. (2005). Second language research: Methodology and design. Mahwah, New Jersey: Lawrence Erlbaum Associates, Inc.
  • Mukherjee, J. (2004). Bridging the gap between applied corpus linguistics and the reality of English language teaching in Germany. Applied Corpus Linguistics52, 239-250. Retrieved from [Accessed December 11, 2018].
  • Reppen, R. (2010). Using corpora in the language classroom. Cambridge, United Kingdom: Cambridge University Press. Retrieved from [Access 9th December 2018].
  • Oghigian, K., & Chujo, K. (2010). An effective way to use corpus exercises to learn grammar basics in English. Language Education in Asia1, 200-214. Retrieved from [Accessed December 13, 2018].


Leave a Comment