日本人英語学習者のための語彙テスト開発
Developing a Vocabulary Size Test for Japanese EFL Learners
Kazumi Aizawa
1998
Zenkoku Eigo Kyoiku Gakkai kiyo
I'he purpose of this study is to analyze Nation's Vc)cabulary Levels Test (V' LT) to check its validity, to develop a vocabulary size test for Japanese EFL learners and to give a portion of it to a sample of learners to compute the reliabiLity of the newly developed test. According to the four surveys of VLT, feur problems were pointed out; too high a ratio of Ioan-words in Japanese, djfficulty in understancling the diefinitions of words, inadequacy of employing word family counting, and the
more »
... of classical vocabulary lists, Based upon these findings, a Japanese version ol vocabu]ar}, s' ize test was developed and given to 817 students at junior high school, senior high schoel and university Ievels. The results of this test showed that forms [1], [2] and [3] have a relatively high reliability coefficicnt, while forms L41 and [5] have some ruom to be improved. Further improvement of this te$t is suggested for future re$earch. 1, lntroduction There has been continuing recognition of the importance ef vocabulary knowledge for second language Iearners. This recognition has brought researchers' and practitioner$' interest in measuring learners' total size of vocabulary. It is often necessary to find out how many words particular learners know for diagnostic and for research purposes. It is useful here to distinguish between breadth tests and clepth tests of v・ocabular]r knowledge. Brcadth tests of vocabulary provide rough cornparative estimates of individuals' overall vocabulary knowledge, while depth tests are defined either in terrns of kinds of knowledge of specific words or in terins of degree of know]edge of such words. Popular examp]es of the breadth tests are Vocabulary Leve]s Test (Nation, 1990) and EFL Vocabulary Tcsts CMeara, 1992). The latter examples are VkJord Association Test (Read, 1993) and Vocabulary Knowledge Scales (Wesche and Paribakht, 1996), However, these tests were designed especially for advanced ESL students who will start their 75 Japan Society of English Language Education NII-Electronic Library Service JapanSociety ofEnglish Language Education academic study at university level in English-speaking countries. Therefore, the levels of the tests are not neeessarily suitable for Japanese students at the secondary school level. Furthermore, because of the Hmjtations of these students' language proficiency, few of those tests can measure efficiently how many words Japanese secondary school students actually know (Tono et al., l997). This study is designed to develop a vocabulary size te$t for Japanese EFL learners and to v・erify its reliability. Since Meara's checklist format does not require that the leaniers provide any independent evidence that the}, know・ the words they have ticked, this study focuses on the worddefinition matching forn)at, which Nation used fer his tests. Method In order to develop a vocabulary test, we need te deal with ene simple theoretical question: what are the criteria for knowing a word? Several studies have tried te answer this question (cf. Richards, 1976; Palmberg, 1987; Nation, 1990), but they have not reached a consensus on this matter. Therefore, in this study knowing a word is defined, for the sake of simplicity and ease, as "having an ability to choose, ainong the Iist, the correspending word with the meaning assigned." The process of this study is as follows: (1) To clarify the problems ef Nation's VLT when jt is given to Japanese learners (2) To compare vocabulary ]ists to choose the most suitable one for Japanese students (3) To develop a vocabulary size test and to give it to students at different levels (4) To analyze the results of the test 3. Nation's Vocabulary Level$ Test (VLT) (I) Overview of VLT A sample section of VLT is shown below: 1. business 2. clock3. horse 4 pencil 5. shoe6 wall part of a house animal with four Iegs something used for writing This test was designecl to assess vocabulary knowledge at five frequency levels: 2,OOO-word, 3,OOO-words, 5,OOO-words, the university word level and 10,OOO・word. At each of the five levels, there are 36 words and 18 definitions, in groups of six and three in each section respectively, All the words in each group are the sarne part of speech, in order to avoid giving any clue to meaning based on form. Although VLT has proved to be a useful diagnostic tool, at least four possible 76
doi:10.20581/arele.9.0_75
fatcat:lgpr35hpovavdfnvjxj7ldtlyu