Critical thinking and collaboration as predictors of self-reported problem-solving competence in social studies...

This study examined the extent to which critical thinking and collaboration are statistically associated with junior high school students' self-reported problem-solving competence in Social Studies (IPS) learning in Indonesia. Using a quantitative correlational design with cluster random sampling of intact classrooms, data were collected from 328 eighth-grade students across six public schools in four cities (Padang, Jambi, Palembang, and Yogyakarta) via a 32-item self-report questionnaire screened for validity and reliability (α = 0.85-0.94). Most students reported moderate levels on all three constructs (66.8-69.2%). Self-reported critical thinking correlated strongly with self-reported problem-solving competence (r = 0.77, p < .001); self-reported collaboration correlated moderately with it (r = 0.57, p < .001). A multiple regression model indicated that critical thinking and collaboration jointly predicted the self-reported problem-solving score, F(2, 325) = 276.50, p < .001, explaining 63.0% of variance (adjusted R² = 0.628), with critical thinking making the larger unique contribution (β = 0.655, semi-partial R² = 30.0%) relative to collaboration (β = 0.214, semi-partial R² = 3.2%). Because students were sampled within intact classrooms, the model was re-estimated with cluster-robust standard errors and as a two-level model with students nested in classrooms (k = 10 classes; intraclass correlation = 0.065-0.115 across the three constructs); conclusions were unchanged. Bootstrapped 95% confidence intervals and a sensitivity analysis addressing one corrected data-entry value further supported the stability of the estimates. Exploratory factor analysis provided partial support for a three-factor structure (28 of 32 items, 87.5%, loaded most strongly on their theoretically assigned factor), but four problem-solving items cross-loaded with the critical-thinking and collaboration factors, and the corresponding factor correlations (r = 0.44-0.65, oblique rotation) indicate meaningful conceptual and empirical overlap among the constructs, particularly between critical thinking and problem-solving. Harman's single-factor test (33.6% of variance) did not indicate that a single factor dominated the item covariance structure, although this diagnostic cannot rule out common-method variance given that all three constructs were measured through the same self-report instrument at a single time point. The findings are consistent with a cognitive-social model in which analytical reasoning and social interdependence are both statistically associated with students' capacity to represent and address authentic social problems; because the design is cross-sectional and all variables share a common measurement method, these associations should be interpreted as statistical rather than causal. Implications for integrating critical-thinking-oriented and collaborative pedagogies in IPS instruction are discussed alongside the study's methodological limitations.