<![CDATA[Critical thinking skills (CTS) are a core competency that must be instilled from elementary school. However, CTS among Indonesian elementary students remain relatively low, particularly in evaluation and inference. This study examines the effect and improvement of students' CTS through a video-assisted problem-based learning model based on socio-scientific issues on deforestation, an integration not yet examined by Facione indicator at the elementary level. A quasi-experimental nonequivalent control group design was used with 50 fifth-grade students from an elementary school in Purwakarta Regency, selected through purposive sampling. The instrument comprised 12 multiple-choice items covering the six Facione CTS indicators, validated by experts and tested for reliability using Anates Version 4. Data were analyzed using the Shapiro-Wilk test, Levene's test, Wilcoxon signed ranks test, Mann-Whitney U test, N-Gain (overall and by indicator), and Cohen's effect size (r). The Mann-Whitney U test confirmed a significant between-group difference (Z = −3.493; p < 0.05) with a moderate effect size (r = 0.494). The experimental class's mean N-Gain reached 87.91% (high category), far exceeding the control class's 33.78% (low category), with all experimental students showing improvement. N-Ganin by indicator analysis revealed the highest gains in self-regulation (94.1%), evaluation (90.0%), and inference (88.9%), while explanation showed the lowest gain (25.0%), indicating this dimension needs more targeted scaffolding. The PBL-SSI-video integration effectively developed five of six Facione CTS dimensions and is recommended as an evidence-based science learning model for elementary schools]]>
