This study aimed to develop and validate the Internet-Administered
Learning Style VAK Inventory for Junior High School students in
Indonesia, addressing the need for culturally relevant, technologydriven assessments. Existing learning style inventories often lack
adaptation to specific educational contexts, and there is a growing
demand for accessible, internet-based evaluation tools. This study seeks
to fill this gap by examining the inventory’s construct validity,
reliability, and effectiveness in classifying students’ learning
preferences. The inventory, consisting of 15 items measuring visual,
auditory, and tactile learning styles, was tested on a sample of 1,400
students using online data collection. Confirmatory Factor Analysis
(CFA) was conducted to assess the factor structure, yielding good model
fit indices (CFI = 0.946, RMSEA = 0.024, SRMR = 0.027), supporting the
inventory’s validity. However, internal consistency was moderate
(Cronbach’s Alpha = 0.580), highlighting the need for further
refinement. Item discrimination indices ranged from 0.132 to 0.328,
suggesting revisions for clarity and relevance. Descriptive findings
indicated that tactile learners scored slightly higher (Mean = 21.187)
than visual (Mean = 20.872) and auditory learners (Mean = 19.119),
reflecting individual differences in learning preferences. While the
study confirms the instrument’s structural validity, limitations include
moderate reliability and potential biases due to online administration.
Future research should focus on improving reliability, refining weak
items, and ensuring broader population representation. The findings
suggest that this inventory could assist educators in tailoring
instructional strategies to students' learning preferences, supporting
adaptive curriculum planning and personalized learning approaches in
digital education settings