Translanguaging in Reading Comprehension Assessment: Implications on Assessing Literal, Inferential, and Evaluative Comprehension among ESL Elementary Students in Taiwan

Abstract

In an attempt to increase the accuracy of L2 reading assessments on the true knowledge of ESL 
learners, this study explores the three levels of reading comprehension (literal, inferential, and 
evaluative) and the notion of translanguaging for the possibility of an alternate testing method. Data 
were collected through a self-designed reading comprehension assessment among 123 sixth-graders 
in Taiwan. Two versions of the assessment were created: one provided reading text and questions in L2, 
and the other provided just the questions, translated into L1. Results indicate that student performance 
on all three comprehension levels were better with L1 questions and that all levels demonstrated high 
practical significance. Performance on evaluative open-ended questions was significantly better with
L1 questions and L1-allowed responses. Correlations were found between students’ L2 proficiency and 
comprehension, and on the inferential and evaluative levels, L2 proficiency had higher correlations with 
questions in L1 than in L2.