1. Introduction

This chapter introduces the concept of data-driven learning (DDL) and its effectiveness in second language learning. It discusses the lack of research on how learners interact with concordance-based materials or corpus-based tools in DDL and the need for further examination of learning processes. Additionally, it highlights the limited focus on non-Indo-European languages such as Chinese in DDL studies and the specific focus of this study on the learning of Chinese resultative constructions using concordance-based materials and think-aloud protocols.

2. Data-Driven Learning

This chapter discusses the teaching and learning approach of Data-Driven Learning (DDL). It highlights the growing enthusiasm and empirical studies in the field. However, it also points out some obstacles to adopting this approach, including learners' challenges and the need for further evidence on its effectiveness for learners of different proficiency levels. The chapter also emphasizes the neglected aspects in the existing DDL literature, such as the learning process and the focus on languages other than English, which the present study aims to address.

3. Chinese Resultative Constructions

This chapter discusses Chinese resultative constructions, including their form and function, and their unique characteristics compared to other languages. It also outlines previous research on the difficulties learners face in acquiring resultative constructions in Chinese. The chapter introduces the study's use of data-driven learning (DDL) approach to investigate learners’ cognitive processes when studying resultative constructions and presents two research questions to be addressed in the study.

4. Research design

The study used a quasi-experimental research design and the think-aloud protocol to investigate learners' use of strategies during the DDL activities. Six non-Chinese first-year undergraduate students took part in the study, and they were asked to complete pre-tests and post-tests to measure their knowledge of Chinese resultative constructions. The method used to capture learners' cognitive processes was think-aloud, and a coding scheme was drawn from previous studies to categorize cognitive strategies into metacognitive, cognitive, and affective strategies. The research procedure consisted of three stages: preparation, training and treatment, and after-treatment testing, which lasted for four weeks. During the training and treatment process, participants were provided with concordance-based sentences and encouraged to read and interpret them, while the whole process was audio-recorded. After completing the treatment session, participants were administered the same test used in the pre-treatment preparation.

5. Results

- Learners of Chinese mostly employed cognitive strategies, with Emma and Lily employing all three categories of strategies more frequently than the other learners.

- When decoding the concordance lines, learners typically summarised the key elements and meaning of the sentences, and some also used the grouping strategy to identify similar patterns.

- Only Emma and Lily successfully inferred regularities in Chinese resultative constructions, with Lily guess the meaning of certain constructions.

- Half of the participants also used metacognitive strategies, such as monitoring and planning, while affective strategies were the least used.

- In general, learners made a slight improvement in post-test scores, with Lily and Emma showing learning gains possibly related to their greater use and variety of strategies.

- Mike, who made little improvement, employed strategies with the lowest frequency and limited variety, leading to unsuccessful uptake of the structure.

6. Discussion

The results show that six learners of Chinese in this study employed different strategies, particularly cognitive strategies, with summarising, grouping, and inferring hypotheses being the most frequently used ones. Cognitive strategies were used more often than other strategies, such as metacognitive and affective strategies. Higher-proficiency learners tend to use more various metacognitive strategies than their lower-proficiency counterparts. DDL can be useful for some learners to acquire a target linguistic item or at least to become more aware of it. There is a strong relationship between DDL-focused strategies and the success of corpus-based L2 vocabulary acquisition. Learners’ use of strategies influenced their learning outcomes, as learners who used a larger number of cognitive, meta-cognitive, and affective strategies showed better post-test scores. Unfamiliarity with inductive learning strategies tends to make the exploration of linguistic patterns a struggle. DDL may be more effective if students can be exposed to the method for a relatively longer period of time. Two learners successfully improved their knowledge of Chinese resultatives, confirming the value of concordance lines in raising learners’ awareness towards the item. However, future studies with a larger group of participants will be helpful to further explore the correlation between learners' use of strategies and their test performance.

7. Conclusion and Future Direction

This chapter discusses the study's contributions to the DDL literature, including the use of DDL in learning Chinese resultative constructions and learners' strategy use. The study suggests implications for Chinese language teachers and the inclusion of teacher intervention in DDL. It also acknowledges the limitations of the study, such as the need for a larger sample size and a control group in future studies. The study's findings offer new insights for DDL practitioners. The authors express gratitude to the reviewers and Dr. Shuo Yu for their valuable contributions and feedback.