Effects of a Blended Deep Learning Module on Deep Reading Ability and English Reading Achievement among Chinese Primary School Pupils: A Quasi-Experimental Study

Abstract

Primary English reading instruction in China remains largely oriented toward examination performance and literal comprehension, leaving pupils with few opportunities to infer, evaluate, and respond creatively to texts. This study examined the effects of a Blended Deep Learning (BDL) module on sixth-grade pupils' deep reading ability and English reading achievement. The module treats deep learning as the pedagogical aim and the blended structure as the delivery mechanism, and it was developed by integrating the ASSURE instructional design model with the Deeper Learning Cycle across three phases: pre-class online preparation, in-class face-to-face learning, and post-class online consolidation. A quasi-experimental nonequivalent control group pretest-posttest design was employed with two intact Grade 6 classes (N = 60) in a public primary school in Jiangsu Province, China. The treatment group learned through the BDL module for eight weeks while the control group received conventional instruction, and both groups were taught by the same teacher. Deep reading ability was measured with the Deep Reading Ability Questionnaire, covering inferential, critical, and creative reading skills, and achievement with an English Reading Test. MANOVA was used to examine post-test deep reading outcomes after baseline equivalence had been established, and ANCOVA was used to examine English reading achievement while controlling for pre-test achievement. The analyses showed statistically significant differences between the groups across the combined deep reading dimensions and, after adjustment for pre-test scores, in English reading achievement. The findings suggest that a purposefully integrated blended design can support higher-order reading skills and reading achievement in primary EFL classrooms, while the single-school quasi-experimental design requires cautious interpretation and further replication across larger and more diverse samples.