A new measurement method of Chinese texts’ difficulty based on the digital analysis of two-character continuations

Abstract

Abstract A new measurement method of Chinese texts’ difficulty based on the digital analysis of two-character continuations is proposed, which is ‘i + 1’ measurement model of language difficulty. The new method shifts the focus of the measurement of text difficulty from language units and grammatical complexity to combinations of language units, which breaks the limitations of existent methods of the measurement of Chinese texts’ difficulty. First, the effectiveness of the two-character continuations in the differentiation of texts difficulties is verified. Second, for the application of the ‘i + 1’ model in measuring text difficulty, the pedagogy and language acquisition theories are analyzed to check its rationality. Third, a corpus of two-character continuations is built up, with which the empirical values of ‘1’ for two-character continuations with various difficulty levels in the ‘i + 1’ measurement model are obtained. Subsequently, the testing texts, the two-character continuations corpus, and the empirical values of ‘1’ are integrated to test the validity of the ‘i + 1’ model in the measurement of text difficulty. The research shows that, compared to those methods based on the complex features of language, the ‘i + 1’ measurement model based on two-character continuations, with fewer indicators and more convenience in the measurement, demonstrates better measurement result of text difficulty.