Linguistic Features for Development of Arabic Text Readability Formula in Malaysia: A Preliminary Study

Abstract

Studies of Arabic text readability in Malaysia began to grow in recent years. This spurred on the need to reassess the quality of learning Arabic literature. But the constraints are lack of objective measurement tools besides cloze test. Unlike the other languages, especially English language has developed various formulas for this purpose. Therefore, this study is the initial exploration in identify potential linguistic characteristics as predictors of difficulty to the development of the Arabic text readability formula among non-native-speakers in Malaysia. For that purpose this study was used High Arabic text book (BAT) of Form Four as a sample. Cloze test was used to measure the readability of reading text in this textbook. A total sample of 390 students of SMKA from all over Malaysia was used for the test. Linguistic features involved words, sentences and contents density were analysed on 15 reading texts in the BAT textbook. Data were analysed using mean, percentage, standard deviation and Pearson correlation. Results indicated that the level of readability of reading texts in the BAT textbooks are at frustration level. While the linguistic features that include the three categories have imbalance distribution of consumption. Five linguistic features were most consistent of consumption in reading texts is from the category of sentences and contents density. The tendencies of sentences and contents features in predicting the difficulty of 15 reading texts of Form Four BAT textbooks are consistently high. Therefore, to develop the Arabic text readability formulas for non-native-speakers in Malaysia, this feature has the potential to be the basis of measurement.