A Study of Readability Measurement
Sukeyori Shiba · The Japanese journal of psychology · 1957
Current methods of readability measurement are generally “element countig”devices. On the contrary, the new method “Cloze produre” measures the degree of the likeness between the pattern a writer used and the patterns the reader is anticipating while he is reading. The basic assumption of this method lies in transitional probability of language and dispositional mechanism in individuals.As W. L. Taylor, the originator of this method, said, Cloze Procedure can be applicable to any language in any culture. In this study, the validity of applying Cloze Procedure to the Japanese language was tested. As Cloze Unit, two kinds of units were used ; letter and word. A letter corresponds to Japanese “ji” containing “kana” and “kanji”, and a word to Japanese “go”.Results :1. Cloze procedure is applicable to the Japanese language. We had better delete the word i. e. “go” as Cloze Unit.2. Cloze Procedure using “go” (word) as Cloze Unit shows more powerful discrimination than using “ji” (letter) as Cloze Unit.3. The value of 10-15% for deleting frequency may be suitable. In this range, thre effect of the deletion is not significant.4. As subjects of Cloze Procedure, 30 persons are enough to be used.Further problems :In Cloze Procedure, readability is measured only relatively, and it also may be able to make a standard scale of readebility. It is also important to study the dospositional mechanism, individual language pattern and readability itself.