Linguistic Complexity and Text Suitability in Pakistan's CSS and PMS English Examinations: A Computational Analysis

This study examines the linguistic complexity and potential suitability of selected English texts from Pakistan's Central Superior Services (CSS) and Provincial Management Service (PMS) examinations through computational analysis using Coh-Metrix. The corpus comprises 21 texts: 18 CSS texts, including nine précis-writing and nine composition texts from 2016 to 2024, and three PMS précis-writing texts from 2019, 2020, and 2022. A quantitative descriptive corpus-based design was employed to investigate readability, syntactic complexity, lexical diversity, cohesion-related features, and situational-model characteristics. Selected Coh-Metrix measures were also compared descriptively with TASA 11–CCR reference values for Language Arts and Social Studies. The findings indicate variation across the linguistic dimensions, suggesting that the selected texts cannot be adequately characterised through a single readability or complexity measure. The study highlights the value of multidimensional computational analysis for describing the linguistic demands of competitive examination texts. However, because the analysis focuses on textual features rather than candidates' reading performance, the findings do not establish candidates' actual comprehension or examination outcomes.