The purpose of this study is to clarify linguistic features of free writing texts by CEFR-J A2 level learners of English and to indicate criteria of features for judging the quality of the texts.Such criteria would be useful for supplementing the CAN-DO descriptors for “Writing” of the CEFR-J A2 level. A multiple regression analysis of 106 English free writing texts by CEFR-J A2 level learners of English was conducted in order to find expository variables indicating linguistic features of the texts. These texts had been evaluated holistically by experts who were native speakers of English and the scores of the evaluation became a dependent variable. The result showed that a model consisted of four expository variables: number of words, Coleman-liau Index (CLI), the Measure of Textual Lexical Diversity (MTLD), and average length of error-free T-unit.Thus, this study indicated four variables of linguistic features of the free writing texts by CEFR-J A2 level learners of English. With regard to these four variables, 106 texts were divided on the basis of the scores of the holistic evaluation into two groups, top (60 texts) and bottom (46 texts), and the average numerical values of each group were calculated. T-test showed that there were significant differences in the average values of each variable between the top group and the bottom group. On the basis of these average values, the criteria on three variables for judging the quality of the texts were also made and verified by χ² test except for MTLD.

抄録全体を表示

This study explores the relationship between students’ uptake and the languages used in a classroom by adopting a corpus approach. Student uptake questionnaires and transcribed utterance from 11 junior high school classes comprised the main body of data. The transcribed data were tagged and the types and tokens for each tagged utterance were counted according to class. Based on this data, the classes were compared and classified into identical groups. For the analysis, the author selected classes featuring the same main activities albeit in a different primary language. Using data gathered through the uptake questionnaire, a Kruskal-Wallis nonparametric test and Mann-Whitney U test were conducted to identify the differences between students’ uptake and the language primarily used in class. The results indicated that students’ uptake varied according to the main language used in class and that the L1 and the L2 play different roles in the uptake process.

抄録全体を表示

The present study examined EFL readers’ understanding of expository texts, with a particular focus on text explicitness (explicit vs. implicit), familiarity (familiar vs. unfamiliar), and readers’ L2 reading proficiency (high vs. low). Fifty-five Japanese university students who varied in L2 reading proficiency participated in an experiment; after reading familiar and unfamiliar expository texts, participants judged whether three types of recognition sentences had appeared in the texts (sentence recognition task; SRT). The SRT results revealed that both proficiency and familiarity influenced understanding of explicit and implicit information, suggesting that comprehension difficulties associated with these factors emerged at the level of explicit information. In addition, recognition confidence ratings indicated that (a) proficiency effects become more pronounced for implicit ideas, (b) proficient readers have potential to learn implicit ideas from familiar texts, and (c) familiarity effects might outweigh proficiency effects. Based on these findings, reading instructions that accommodate differences in students’ reading proficiency and text familiarity are discussed.

抄録全体を表示

Academic writing has been traditionally viewed as impersonal, but recent corpus-based research has revealed that writers of research articles (RAs) in both scientific and social sciences use first-person pronouns in their RAs for constructing authorial identity. This research focuses on the use of first-person pronouns, powerful linguistic resources to convey an author’s voice. Using two corpora consisting of 21 RAs each by Japanese and native English TESOL researchers, we investigate how first-person pronouns are used to represent the authors in text. We also explore the factors that characterize the Japanese writers’ use of first-person pronouns by referring to the results of a survey conducted on 70 Japanese EFL teachers inquiring about the practice and perception of first-person pronoun use in RAs. The results indicate that the Japanese researchers use first-person pronouns less frequently in their RAs. The further analysis suggests that the low frequency of first-person pronouns may not be caused just by the researchers’ view of academic writing but that different factors such as consideration for the conventions practiced in the discourse community or research methods adopted may be related to the choice of first-person pronouns in RAs.

抄録全体を表示

Studies have been accumulated on individual policies related to English language education in Japan or on Japanese society as a determinant of these individual policies. Few studies, however, have ever been done on the discussion in the Japanese Diet, which connects the policies and the society. This is because the society, or the view of the Japanese people, affects what and how to discuss in the Japanese Diet, based on which new policies must be developed. The aim of this study is to investigate thoroughly what and how the actors, who are the participants of the Diet, have discussed English education in the Japanese Diet. For this purpose, we gathered all the statements made in the Japanese Diet about English education from 1947 to 2013, and analyzed the data using the method of quantitative content analysis. We divided the post-war era into four periods based on the number of the conferences mentioning English education and defined five actors. The main findings are: (1) each period has its own prominent topics; (2) there are some words or types of statement more frequently used by each actor. We believe this research will open a new field of study, which would enhance the historical studies of English education.

抄録全体を表示

Successful reading requires main idea comprehension by macrorules: selection (distinguishing explicit main ideas from explicit details), generalization, and construction (integrating subordinate words or propositions into an implicit superordinate one). Although past studies used a summary task to elicit readers’ macrorule use, readers might not use the rules without such a task. This study examined L2 readers’ macrorule use adapting a sentence recognition task, which reflects main idea activation by the macrorules. After 73 Japanese EFL university students read each paragraph of two texts, they judged whether the following probes were written in the paragraph: (a) explicit main ideas vs. (b) explicit details (to examine selection use), and (c) implicit main ideas vs. (d) thematically-related but undescribed ideas (to examine generalization and construction use). The participants recognized the explicit main ideas more correctly and quickly than the explicit details. Moreover, whereas they should have rejected the implicit main ideas as these were not explicitly described, they often recognized that the ideas were written in the paragraphs, compared to the thematically-related ideas, and needed longer time to suppress that information. These findings indicate the activation of the explicit and implicit main ideas and suggest that L2 readers could use macrorules without the summary task.

抄録全体を表示

Previous studies have insisted that attention to both the meaning and form of new lexical items is a prerequisite for vocabulary learning through inferencing. However, the kinds of language cues conducive to this process have not yet been thoroughly examined. Therefore, this think-aloud study investigates the relationship between inferential strategies, inference success, and subsequent vocabulary learning, focusing on the properties of two cue types (i.e., the presence of a prefix × the degree of contextual constraint). A total of 10 Japanese undergraduates and graduates read a short passage containing target words of different cue types and inferred their meanings. Thirty minutes after the inference task, a vocabulary test was administered to measure vocabulary learning. The results demonstrated that the degree of inference success was higher for target words in strong constraint context irrespective of a prefix’s presence. As for learning, the retention rate was high for prefixed target words or those in less constraining sentences. However, the pattern of strategy use for words in weakly constraining contexts contradicted the overall tendency for form-focused strategies to correlate with retention; qualitative analysis of some protocols implied that learners attended to forms of unknown words more often for the words in weakly constraining contexts.

抄録全体を表示

The Presentation-Comprehension-Practice-Production (PCPP) approach is often used in Japanese junior high school English classes, with a focus on communicative interaction. Although ample communicative activities are implemented in a PCPP approach, teachers are likely to present grammar and vocabulary in the comprehension stage using a traditional teacher-centered deductive method. In this study, Data-Driven Learning (DDL) was implemented as a means to Focus on Form in a PCPP approach and students explored language data to understand vocabulary and grammar in a learner-centered way. To investigate the effectiveness of this DDL method, we implemented deductive (traditional) and inductive (DDL) instruction at two schools. In School A, DDL was used to review and deepen known vocabulary, and at School B, to learn new grammatical items. Accordingly, four groups totaling 121 students took part in the research. Results show that DDL was as effective as traditional instruction in a posttest given one week after instruction. However, after four weeks, the DDL groups retained the gained knowledge better than the traditional groups at both School A and B. Furthermore DDL students performed a more detailed analysis on finding grammar rules in the posttests. Therefore, we can conclude that the use of DDL was effective for retaining language knowledge and for developing students’ ability to explore language rules.

抄録全体を表示

This paper reports on the opinions of Japanese learners of English at the first author's high school about their English language lessons. In these lessons, English was mainly used as a medium of instruction while English grammar and usage were explained in Japanese. A survey was administered to the learners; it had 25 items, many of them focusing on the usefulness of the English lessons. Next, two analyses, one dealing with all the learners' responses and the other with the learners' categorizations, were conducted. The results of the first analysis indicated that (a) the learners think that the English lessons were more useful for their future effort to improve their English than for preparing for their university entrance examinations, (b) they think that the lessons improved their listening and speaking skills, and (c) they preferred to have their teachers explain English grammar and usage in Japanese. The second analysis, conducted using a decision tree algorithm, specified the types and/or tendencies of the learners. The results of both analyses seem to have implications for the planning of English lessons.

抄録全体を表示