<![CDATA[Reading comprehension in English as a Foreign Language (EFL) remains a significant challenge for Indonesian students, particularly at the inferential level. The background of this study highlights a national literacy crisis where students face a severe "vocabulary bottleneck" and heavily rely on digital translation tools, leading to a cognitive "short-circuit" that disrupts reading flow and deep text comprehension. To address this issue, the present study integrates Technological Pedagogical Content Knowledge (TPACK) as a multimodal digital scaffold to evaluate the effect of the Contextual Vocabulary Guessing Strategy via Google Sites on inferential reading comprehension. This study used a quasi-experimental, non-equivalent control group pretest-posttest design with 63 eleventh-grade students from SMAN 1 Kandat, selected for homogeneity in prior proficiency scores. The experimental group (n = 30) received an intervention focusing on four context clues (definition, contrast, example, and logic) embedded in interactive texts concerning Gen Z health issues, while the control group (n = 33) followed conventional teacher-led translation methods. Data were collected using a validated and reliable 20-item multiple-choice test mapped to Barrett's Taxonomy. The Analysis of Covariance (ANCOVA) results revealed a significant main effect of the instructional strategy (F(1, 60) = 10.34, p = 0.002, partial eta squared = 0.147) after controlling for prior reading ability. The findings indicate that the digital contextual guessing intervention significantly improved students' inferential reading comprehension compared with conventional instruction.]]>
