On Readability Evaluation of News Texts from Russian Multimodal University Websites

Multimodal university websites have drawn researchers’ attention since their appearance at the turn of the century. This chapter reviews the foci of salient studies on university websites, noting that very few of them deal with the readability of materials on university websites. Those few researchers who conducted readability evaluation of university websites used freely available software based on classic readability indices. None of them evaluated the readability of Russian university websites. To bridge this gap, we started a project aimed at evaluating the readability of leading Russian university websites. In the first stage of the project, a corpus of 1000+ texts was built by parsing the selected university websites. The readability of the texts was evaluated both automatically with Python script based on readability indices and by 132 human assessors who evaluated a representative sample of 340 texts of the corpus according to three groups (linguistic, structural and logical, and appealing) of 10 criteria completing the questionnaire. The correlation analysis showed a negative correlation between the automatic and respondents’ estimates for easier and more difficult texts. This chapter proposes a possible explanation for this divergence and outlines further research. The most logical explanation capitalizes on the fact that readability is a function of two groups of mega-parameters: objective parameters of text comprising quantitative and qualitative characteristics and individual characteristics of the reader. Probably, readability indices do not take into account the contributions of text’s qualitative characteristics and readers’ characteristics in full.