Evaluation of Reading Difficulty Based on Factor Analysis and Linear Regression
DOI:
https://doi.org/10.54097/fbem.v10i1.10174Keywords:
Measuring text difficulty, Factor analysis, Linear regression analysis, Comprehensive evaluation model.Abstract
This paper conducts research on how to measure the difficulty of English texts and how to extend it to other languages, such as Chinese. Firstly, text features that can represent reading difficulty are extracted and represented using parameters. Secondly, the extracted indicators are subjected to KMO and Bartlett's tests, and the results show that these indicators are suitable for factor analysis. Then, a comprehensive evaluation model of language difficulty based on factor analysis is constructed, and the difficulty coefficient composite score F of 16 English reading texts is calculated. Additionally, a linear regression equation is constructed to separately measure the correlation between Type-Token Ratio (TTR) and the text difficulty coefficient F. Finally, the conclusions obtained from the model are compared with the officially published difficulty coefficients of English reading in the postgraduate entrance examination, validating the rationality of the results. Furthermore, the above constructed multi-indicator comprehensive evaluation model is extended to other languages such as Chinese, confirming the rationality and universality of the model, effectively solving the problem of "measuring text reading difficulty.".
Downloads
References
S.M.Rao, H Zheng, S.J.Li: A Survey of Leveled Reading. Proceedings of the 20th Chinese National Conference on Computational Linguistics, 2021, p.689-702.
Ar H R, Utomo P, Rajendra M Y: Classification of Lexile Level Reading Load Using the K-Means Clustering and Random Forest Method[J]. Kinetik: Game Technology, Information System, Computer Network, Computing, Electronics, and Control,2020.
M C T, E S E, B K A, et al.: Effectiveness of accelerated reader on children's reading outcomes: A meta-analytic review.[J]. Dyslexia (Chichester, England),2022,29(1).
Eleyan D, Othman A, Eleyan A. Enhancing Software Comments Readability Using Flesch Reading Ease Score. Information. 2020; 11(9):430.
X.J.Zhan: Study on the validity verification of the validity of the reading text in the foreign language test - The perspective based on the corpus. Foreign language test and teaching, 2015 (03): 23-29.
J.T.Lu, M.Z.Yang, D.C.Zhang, et al.: Testing and Evaluation of English Reading Ability based on Rasch Model. Science and Technology Vision, 2019 (29): 76-77-75.
Y Cheng, D.G.Xu, J Dong: Automatic grading study of Chinese the difficulty of reading texts based on the integration of multilingual and deep features, journal Chinese Information, 2020, 34 (04): 101-110.
Y Tao: Improve the corpus and improve the quality of Chinese cultural translation. China Journal of Social Sciences, 2017-09-26 (003).
Y Cheng, D.G.Xu, J Dong: Analysis of key factors and easy-to-read formulas for text reading difficulty grading based on the language textbook corpus, 2020 (01): 132-143.
