A cognitive processing model of reading comprehension in English as a foreign language using the linear logistic test model

• Five cognitive processes were postulated to underlie reading comprehension in English as a foreign language.
• The processes were parameterized using a linear logistic test model.
• Inference-making turned out to be the hardest process to use while vocabulary was the easiest.
• The difficulty of the five cognitive processes explained 56% of the variance in item difficulties.

Reading comprehension in a foreign language (FL) is a complex process with many underlying cognitive components. Many second language researchers have tried to explain reading comprehension in terms of taxonomies of subskills and processes. However, the nature of these components is not yet known. Previous research using exploratory and confirmatory factor analysis has yielded contradictory results. The purpose of this study is to investigate the underlying cognitive components and processes of FL reading comprehension using linear logistic test model (LLTM; Fischer, 1973). For the purpose of the present study, the data of 400 applicants taking an advanced high-stakes reading comprehension test were used. The cognitive processes underlying the test were derived. The derived processes were reading for details, making inferences, reading for main idea, syntax, and vocabulary. Linear logistic test model showed that making inferences is the hardest process to employ and vocabulary the easiest. The implications for teaching and testing reading comprehension are discussed.