中学语文教科书文言选文难易度评量模型检验

Tests of the Evaluation Model for the Difficulty of Classical Chinese in Middle School Chinese Textbooks

文言文难易度的评量问题关系到中学语文课程内容、学习材料的选择和文言阅读能力的测评。张秋玲于2010年构建的文言文难易度评量的数学模型,首次为该问题的解决提供了一条可持续探索的路径。在此基础上,本文采用内容分析法和调查研究法,对2010数学模型的遗留问题及疏漏之处进行了两次检验和修订:第一次是借助读者阅读能力与文本复杂性契合度的实证检验,评判数学模型评定的文本难易度与通过读者阅读能力评定的难易度是否能够拟合;第二次是借助专家干预下的人工检验,重点对模型中词频统计可能存在的分类偏差和疏漏作出了校正。经过两次修订,数学模型的相关系数提高了0.112,提高了数学模型评量文言文难易度的信效度,这也是在为今后实现文言文的自动化分级做充足的理论准备。

Evaluation of the difficulty of classical Chinese is critical for the selection of Chinese curriculum content and learning materials,as well as for the test of classical Chinese reading ability in middle schools.The mathematical model for evaluating the difficulty of classical Chinese constructed by Zhang Qiuling in 2010 provided an exploratory path for problem solving.On this basis,this paper uses the methods of content analysis and investigation to test and revise the remaining problems in the 2010 mathematical model twice.First,with the help of the empirical tests of the consistency between readers’reading ability and text complexity,we judged whether the difficulty of classical Chinese evaluated by the mathematical model matched that assessed by readers’ability.Second,we corrected the possible classification deviations and omissions in word frequency statistics with the help of manual test of expert intervention.After the revisions,the correlation coefficient of the mathematical model was increased by 0.112.The improvement of the reliability and validity of the mathematical model in evaluating the difficulty of classical Chinese made sufficient theoretical preparation for realizing the automatic grading of classical Chinese in the future.