Abstract
This investigation by computer analysis is concerned with the frequency and register of the words, sentences and sentence patterns that occur in ETW (English Technical Writing) Test and STEP (the Society for testing English Proficiency) Test, both of which are authorized by the Japanese Ministry of Education, Science and Culture.
The purpose of this study is to find some characteristic aspects of EST quantitatively by making clear some distinctions between EST and GE as to token-type ratio, validity of basic vocabulary, frequency of words grouped into parts of speech, sentence length, frequency of sentence patterns, etc.
