1. Literature review
1.1. Lexical diversity
Lexical diversity (LD) has been defined as ‘the range of vocabulary and avoidance of repetition’ (Malvern et al., Reference Malvern, Richards, Chipere and Durán2004). LD is still perceived as an indicator of second language (L2) writers’ writing abilities (Engber, Reference Engber1995; Vo, Reference Vo2019). As the illustrative indices to assess LD, the measure of textual lexical diversity (MTLD) and vocd-D are often utilised in many research studies. While MTLD is examined ‘as the mean length of sequential word strings in a text that maintains a given [type-token ratio] value’ (McCarthy & Jarvis, Reference McCarthy and Jarvis2010, p. 384), vocd-D is calculated based on a series of randomly selected vocabulary. As LD can be examined in various ways and from various perspectives, researchers need to utilise multiple benchmarks, including MTLD and vocd-D, rather than a single yardstick (McCarthy & Jarvis, Reference McCarthy and Jarvis2010).
1.2. Theme-based instruction
Theme-based instruction (TBI) is one model of content-based instruction that aims to improve L2 learners’ language abilities through the integration of language and the content of the topics relevant to their specialities and interests (Brinton & Snow, Reference Brinton, Snow, Snow and Brinton2017; Lyster, Reference Lyster2018, see Figure 1 in the poster).
The effectiveness of TBI on L2 learners’ writing ability development, including LD development, has not been investigated comprehensively in many settings. For example, Phonhan et al. (Reference Phonhan, Phusawisot and Praphan2016) reported that the participants reduced their writing apprehension and developed their writing abilities while receiving TBI, but that writing abilities developed through TBI were not fully explored. In Shibata's (Reference Shibata, Clements, Krause and Bennett2019) study, Japanese high school students wrote more tokens and provided more examples and reasons as they received TBI. However, neither the complexity of vocabulary utilised in their essays nor the possible factors affecting their writing abilities were fully explored.
1.3. Research questions
In order to address research issues from the review of the literature, the present study aims to address the following research questions:
1. Does TBI affect intermediate English proficiency level Japanese university students’ LD in their academic writing throughout one academic year?
2. Do the participants realise the benefits of sociocultural scaffolding provided by the teacher, their classmates, and materials while receiving TBI?
2. Method
2.1. Participants and context
Forty-five intermediate proficiency level second-year Japanese university students participated in the present mixed-methods research study for nine months. All of them were Japanese native speakers and had no experience of studying and/or living abroad.
In this present study, the participants wrote cause-and-effect essays about the following four themes decided by the researcher (see Essay Themes and the Timing in the poster). These topics were not controlled or considered with respect to LD.
2.3. Data collection and analysis
2.3.1. Students’ essays and drafts
The researcher collected all students’ drafts and essays as digital files were collected in every class and utilised an online text-analysis tool called Text Inspector (2020) to examine their MTLD and vocd-D in their academic essays. After that, Bayesian Friedman tests were performed with JASP (2020) to compare the first drafts with revised drafts in each assigned theme. The reliability of the p-value is currently considered problematic to estimate the probability and alternative statistic benchmarks are required (Wasserstein & Lazar, Reference Wasserstein and Lazar2016). To address this issue, Norouzian et al. (Reference Norouzian, De Miranda and Plonsky2019) maintain that a Bayesian approach can be a useful measurement instead of the p-value. Therefore, in addition to the p-value, the researcher used Bayes factor clarification scales (see Table 1 in the poster).
2.3.2. Students’ learning reflection notes
Participants wrote their learning reflection notes in English and submitted them as digital documents to the researcher after each class. They reported how they sought to revise their academic essays and how they learnt about the target themes to develop their academic essays throughout the academic year. 1,350 reflection notes, 30 per student, were collected. Reflections were coded thematically to detect patterns and categorised into groups. After the researcher coded all qualitative data and checked the reliability and validity with the participants, inter-coder reliability was estimated with Cohen's kappa (Cohen's k = 0.90).
3. Results and discussion
Both descriptive statistics and the Bayesian Friedman test results indicated that the participants developed their LD in writing, regardless of the LD indices and the assigned themes (see Figures 2 & 3 and Table 2 in the poster). In the draft phases of some of the assigned themes, the mean value of LD decreased slightly or remained stable for a while and then rose significantly. These findings might be consistent with Ishikawa's (Reference Ishikawa2015) findings that L2 learners’ LD decreased and then increased as they developed their language abilities. In their learning reflection notes, the participants remarked on the roles of scaffolding by the researcher, their classmates, and materials to deepen their understanding of the assigned themes and learn a wider range of vocabulary to utilise them in their revisions (e.g., Qualitative Data in the poster). These findings can be consistent with Liang's (Reference Liang2007) findings that scaffolding can raise L2 writers’ confidence and abilities to use the target language to express their ideas and opinions and Padmadewi and Artini's (Reference Padmadewi and Artini2019) report that learners’ engagement in their writing activities and their writing quality developed while receiving scaffolding. Consequently, TBI played a significant role in developing the participants’ LD in their writing.
4. Conclusion and limitations
The findings indicated that TBI had noticeable positive effects on their LD in their essays. Through this pedagogy, the participants received scaffolding from the researcher, their classmates, and materials to deepen their comprehension of the assigned themes and improved their writing abilities. From these findings, TBI can be said to contribute to intermediate English proficiency level Japanese university students’ foreign language learning, mainly to utilise a broader range of vocabulary in writing.
Although the positive effects of TBI on students’ writing abilities were illustrated in this mixed methods research study, the following limitations need to be addressed in future investigations:
1. More writing samples are necessary.
2. L2 writers’ topic familiarity need to be fully investigated in order to further explore the roles of topics in TBI to affect learners’ writing skills.
3. The effectiveness of schema development on their LD in writing still need to be investigated.
This research study illustrates that the participants developed their LD in essays, regardless of the assigned themes, while taking TBI and receiving sociocultural scaffolding from the researcher, their classmates, and materials in this specific setting. Therefore, TBI can still be considered as an effective writing instruction to help learners to improve some parts of their writing abilities in other tertiary educational settings.
Supplementary material
To view supplementary material for this article, please visit https://doi.org/10.1017/S0261444821000306
Naoya Shibata is a part-time lecturer at Nagoya University of Foreign Studies and other universities in central Japan. He is also a doctoral candidate at Anaheim University. His research interests include L2 content-based instruction, and teacher development.
