A Quantitative Corpus Analysis of Word Frequency And Part of Speech in the English Textbooks Used in Senior High Schools in Taiwan

Of all the components involved in English language education–teaching materials, teaching methodologies and approaches, and students’ learning strategies and motivation-the materials used may actually have the greatest effect on the student’s acquisition of English. That is to say, if the textbook is adequate, teachers can teach their students more effectively and efficiently. As a result, their students may more easily acquire the requisite language skills and linguistic knowledge. Therefore, we need to examine the contents of the textbooks in order to judge whether they are appropriately compiled. The textbooks examined in this study were the six volumes used in all senior high schools in Taiwan. Each of the six textbooks covered one semester of English. The six volumes included in this study were for the 1991 academic year. We treat these six volumes as our corpus of study. A corpus was defined by Aarts (1991) as “a collection of samples of running text”, which “may be in spoken, written or intermediate forms, and the samples may be of any length” (p. 45). And corpus linguistics was described by Aijmer and Alwtherg (1991) as “the study of language on the basis of text corpora” (p. 1). During the last three decades, corpus linguistics has expanded rapidly, due to two important events that took place around 1960. One was Randolph Quirk’s launching of his Survey of English Usage (SEU) project in 1959. The other was W. Nelson Francis’ and Henry Kucera’s compilation of the Brown Corpus, “which was to set the pattern for many computer corpus projects” (Johansson. 1991, p. 305). This study is concerned with quantitative corpus analysis, with our efforts especially concentrated on vocabulary, vocabulary acquisition, and the effects of word frequency and part of speech on vocabulary acquisition. The vocabulary items under examination are the words in the 84 reading texts in these six English textbooks, the words in the first volume, and the test words in the monthly English examinations administered in the 1991 academic year at the Kaohsiung Senior High School in Taiwan. There are at least two concepts for word frequency. One is the frequency with which a word appears in a representative corpus. In this sense, words are usually divided into low-frequency, middle-frequency, and high-frequency words. For example, Thorndike and Lorge (1963) listed the 500 words occurring most frequently in English. And Hofland and Johansson (1982) compared the relative frequencies of the 100 most frequent words in text categories A-J (newspaper, miscellaneous information prose, and learned and scientific English) against those in texts K, L, N, and P (different types of fiction). The other concept is the frequency a target word occurs in the target texts under examination. This study belongs to the latter case. In addition, the former researchers rigorously collected the corpora whose features better represented the parameters of the complete existing corpora than the corpus used in this study, which is specially edited for teaching purpose. There have been many findings that the presence of low frequency vocabulary in a text has a negative effect on comprehension (Freebody & Anderson. 1983; Kameenui, Carnin & Fr chi 1 2 • an• M k D. • • r w Wittr. k 1 4). Also there have been well-established findings of frequency effects that facilitate the acquisition of