Development of a Vocabulary Level Test for EFL Learners : The Relationship between learners' Vocabulary Size and Reading Comprehension Skills

Abstract

This study attempts to develop a vocabulary level test and examine the relationship between EFL learners' vocabulary size and their reading comprehension skills. Three hypotheses were tested: (1) a carefully designed vocabulary test based on the word frequency can determine which vocabulary level learners belong to with ample reliability and validity, (2) learners vocabulary levels are significantly correlated with their reading comprehension skills, and (3) there is a minimal vocabulary size (threshold) which is necessary for learners to make successful reading comprehension. Two tests, a vocabulary level test and a reading comprehension test, were administered to 260 twelfth students. The vocabulary test consists of four sections with eighteen multiple-choice questions, i.e., 2,000, 3,000, 4,000 and University Word List (UWL) levels so that it can classify learners into five vocabulary levels. The original version was invented by Nation(1990) as Vocabulary Levels Test. He selected proper words in terms of their frequency, based on three authorized word lists: A General Service List of English (GSL) by West(1953), The Teachers' Word Book of 30,000 Words by Thorndike and Lorge (1944), A Computational Analysis of Present-day American English by and Kucera and Francis (1967). The original test contains 2,000, 3,000, 5,000, 10,000 and UWL word level sections, but not 4,000 level section. The 4,000 level section was originally developed for this experiment under Prof. Nation's instruction by the present writer, on the basis of the three authorized word lists mentioned above. The reading comprehension test, consisting of 20 questions divided into multiple-choice, explanation, summarization and so forth, aims to gauge how much readers have understood the text. The results show that the first hypothesis and the second one were supported, but not the third This means that the vocabulary test developed for this study worked well in classifying the learners in terms of their vocabulary levels. The significantly high correlation (r=.84, p<.0001) between vocabulary levels and reading comprehension skills implies that about 70% of reading comprehension skills can be explained by readers' vocabulary levels.