Developing culturally grounded digital teaching materials based on pedagogical deep learning to strengthen generation Z nationalism in the Indonesia and Malaysia border region of West Kalimantan

In this study, the term ‘pedagogical deep learning’ refers exclusively to the educational approach that emphasises conceptual understanding, critical reflection, and experiential engagement, and is not used in the computational sense of deep learning in artificial intelligence. This development study employs a sequential exploratory mixed-methods design to produce, validate, and evaluate the initial promise of pedagogical deep-learning-based digital teaching materials for border nationalism education in the Indonesia–Malaysia border region of West Kalimantan. Grounded in three complementary theoretical frameworks Social Identity Theory (SIT), Civic Education Theory (CET), and Constructivist Pedagogical Deep Learning Theory (CDL), the study integrated three phases: a qualitative needs analysis (n = 10 participants) using thematic analysis; a cross-sectional PLS-SEM analysis serving as empirical grounding for module design (n = 90); and a single-group preliminary module trial (n = 30). The SEM results are interpreted as exploratory and directional rather than confirmatory, because the model fit indices were mixed (SRMR = 0.136; NFI = 0.458). Within that exploratory frame, popular culture (H2: p = .021; H4: p < .001) and digital literacy (H3: p = .011) were positively associated with nationalism-related outcomes, whereas the Digital Teaching Materials paths were non-significant in the baseline cross-sectional model. All p-values are one-tailed, consistent with the directional specification of H1–H8; two-tailed equivalents are reported in Table 14 . The non-significance of the Digital Teaching Materials paths is consistent with the qualitative finding that no contextually appropriate digital nationalism materials existed prior to this study, although measurement limitations, restricted variance, and limited statistical power cannot be ruled out as alternative explanations. The preliminary trial produced a positive pre-to-post change (pre-test: M = 70.5, SD = 11.5; post-test: M = 80.3, SD = 10.2; mean difference = 9.8, SD of differences = 12.75; t (29) = 4.21, p < .001, 95% CI [5.04, 14.56]; Cohen’s d = 0.85 computed with the pre-test standard deviation as the standardiser, and the paired-samples equivalent is Cohen’s dz = 0.77. Because the trial used a single group without a control condition, these results indicate preliminary promise rather than confirmed causal effectiveness. The study contributes the first SIT–CET–CDL grounded development study for border nationalism education in the Indonesia–Malaysia context, offering a replicable theoretical-pedagogical architecture and a preliminary measurement framework for future research in analogous border education settings.