THE INDICES OF EXAMINATION TEXTS COMPLEXITY

The article presents the results of the analysis aimed at quantitative (word length and sentence length) and qualitative (narrativity, syntactic simplicity, word concreteness, referential cohesion and deep cohesion) characteristics used in Flesh Reading Ease and Coh-Metrix to determine complexity/simplicity of FCE and EGE examination texts corresponding to B2 level by Common European Framework of References. The data used in the research include18 examination texts of three language tests analogous in structure and content: in Russian (Test of Russian as a Foreign Language) and in English (Unified (Russia) State Examination in English (EGE) and First Certificate Examination (FCE)). Flesch Reading Ease indices of all the Russian texts analyzed testify to their higher level of complexity than the announced in the corresponding papers (B2). The English texts correspond to the Level B2 by CEFR. The computerized with Coh-Metrix texts utilized in EGE and FCE prove to belong to B2 level, though the EGE texts demonstrate on average higher than FCE texts levels of concreteness but lower levels of syntactic simplicity. Maintaining approximately equal levels of concreteness EGE and FCE texts obtain about the same levels of complexity (B2) due to the differences in the number of verbalized local and global content links. Thus, the Russian test developers model the authentic English texts for examination purposes eliminating the connectives and verbalizing referential and deep cohesion.