Kiyomi Chujo

Measuring Vocabulary Levels of English Textbooks and Tests Using a BNC Lemmatised High Frequency Word List

The purpose of this study was to create a means for comparing the vocabulary levels of Japanese junior and senior high school (JSH) texts, Japanese college qualification tests, English proficiency tests, and EGP, ESP and semi-ESP college textbooks in order to determine what the vocabulary levels are, and what additional vocabulary is required for students to understand 95% of these materials. This was done by creating a lemmatised and ranked high frequency word list (BNC HFWL) from the British National Corpus. This study found that although most college students should be prepared to take the TOEIC, and high school students should be able to pass both the Daigaku Center Nyushi and Eiken 2nd grade tests, most college entrance exams contain vocabulary that is significantly above the level of high school graduates. Specialized vocabulary lists can be helpful in bridging vocabulary gaps between JSH and ESP, and between JSH and the TOEFL.

A novel, web-based, parallel concordancer for use in the ESL/EFL classroom

To date, the use of parallel concordancers in the classroom has been a largely unexplored area. As a result, there are few guidelines on the design of these tools or the functions they should offer. Also, the number of fully-developed parallel concordance tools is surprisingly few, and most of these are designed for use by corpus linguists and translators. In this paper, we describe a novel, web-based, parallel concordancer designed specifically for use by beginner-level second and foreign language learners of English. The software is built using standard web application tools, such as Linux, Apache, MySQL, and PHP, and thus can run in most university server environments. It is also designed on a similar architecture to the Google search engine, allowing it to work comfortably on very large corpora of hundreds of millions of words. To enable the smooth processing of both English and non-English texts, the concordancer is built to Unicode standards. Its internal token definition settings also employ Unicode character classes meaning that no cumbersome user-defined settings are necessary. Preliminary results show that the new software is considerably faster and easier to use than standard desktop parallel concordance programs. Also, because the tool is web-based, it can be accessed outside of class time, thus allowing students more time for hypothesis-verification and production activities.

Development of a Scale to Measure Learners' Perceived Preferences and Benefits of Data-Driven Learning.

In spite of researchers’ and practitioners’ increasing attention to data-driven learning (DDL) and increasing numbers of DDL studies, a multi-item scale to measure learners’ attitude toward DDL has not been developed thus far. In the present study, we developed and validated a psychometric scale to measure learners’ perceived preferences and benefits of DDL for research and pedagogical purposes. First, we created an item pool by referring to open-ended responses from learners; second, the items were pilot tested with target-level learners of English as a foreign language; and third, with item analyses and exploratory factor analysis, the revised version of the questionnaire was prepared. Finally, the questionnaire was administered, and its psychometric properties were examined with confirmatory factor analysis and fit indices. The final phase also included a measure of task values to explore the convergent evidence of the construct validity of the proposed scale. The results suggest that the scale is a valid measure of learners’ attitudes toward DDL, with the hypothesized model providing a good fit with the data. We propose that the scale can be used in future studies that utilize the same type of questionnaire research to facilitate further investigation of DDL. Suggestions for further research are also provided.

Teaching Remedial Grammar through Data-Driven Learning Using AntPConc

In most Asian countries, students receive between six and eight years of compulsory English education before they enter university. Despite this massive investment in English education, many students, especially in Japan, continue to show a poor understanding of rudimentary grammar rules. In this paper we report on a unique English course designed specifically to address grammar issues at low (remedial) levels using a Data-Driven Learning (DDL) approach. Applications of DDL are becoming more widely reported, but they are generally at the intermediate or advanced level. One of the challenges of using DDL at the remedial level is the lack of suitably leveled corpora. Another challenge is that most corpus tools used in DDL are designed for researchers or advanced learners and thus can appear overly complex. To address these issues, we have developed a simple English corpus built from standard school texts. We have also created a freeware, parallel corpus tool, AntPConc, that is specially designed to be simple, easy, and intuitive to use by beginner learners. Results from the course show significant gains between pre- and post-tests of grammar understanding for beginner-level EFL university students. We also obtained positive student feedback on the AntPConc software.