When family and school support rose together, that match was associated with higher creative-thinking scores, an analysis of PISA data found. The same broad pattern appeared when student openness matched the openness culture of the school. Creative activity pointed in the opposite direction: equal increases in in-school and out-of-school participation were associated with lower scores.
That does not mean that matching support or openness raises creativity, or that activity harms it. The study was cross-sectional, meaning it compared conditions and scores at one point rather than following students over time. It therefore cannot establish cause and effect.
Three pairs, three samples
The analysis examined correspondence, or how closely two related settings or behaviors matched, across three pairings: family and school environmental support; creative activity in school and out of school; and student openness and school openness culture. The source population contained 499,843 students from 63 countries or economies. The pairings were assessed in partly different samples: 387,147 students from 61 countries and 16,667 schools for environmental support; 326,648 students from 62 countries and 16,753 schools for activity; and 365,361 students from 62 countries and 15,252 schools for openness.
Creative thinking was measured on a 0 to 60 scale using 10 plausible values. The models accounted for mathematics, science, reading and socioeconomic measures. Researchers used response-surface analysis, a way to examine both matched levels and mismatches in paired measures, in two-level random-intercept models with country as the second level. The full cubic specification fit significantly better than additive and interaction specifications for all three pairs.
Support and openness moved with scores
In the environmental-support model, matched increases in family and school support were associated with higher scores, but the positive association weakened at higher matched levels. A family-favoring mismatch was also associated with higher scores than a school-favoring mismatch. The quadratic part of the mismatch pattern was not statistically significant, so the precise shape of that difference remains uncertain.
The openness model showed a similar but not identical pattern. Matched increases in student and school openness were associated with higher creative thinking, again with diminishing returns. When the two were mismatched, higher student openness than school openness was associated with higher scores than the reverse arrangement. The quadratic and cubic parts of this mismatch pattern were not statistically significant.
Activity followed a different pattern
Activity followed a different pattern. Equal increases in in-school and out-of-school creative activity were associated with lower creative-thinking scores. The mismatch was asymmetric too: out-of-school participation exceeding in-school participation was associated with lower scores than the reverse configuration. The model included nonlinear terms, but its cubic mismatch component was not statistically significant.
The added contribution was modest
Across the models, congruence terms accounted for more variation within countries than between countries. The full models explained 55% of within-country variation and 78% to 80% of between-country variation. After achievement and socioeconomic controls were included, the unique contribution of congruence ranged from 3.0% to 6.7%. Those figures describe predictive association in the modeled data, not the share of creativity caused by family, school or activity conditions.
The study's limits matter. Because the data were cross-sectional, they cannot show whether changes in support, openness or activity would lead to changes in creative thinking. Many measures were self-reported and may carry method bias. The authors call for longitudinal and cross-cultural research to test whether the patterns persist or change in other settings.
The main message is not that matching always wins. The relationship changed with what was being matched: support and openness had positive associations, while activity had a negative one. With congruence adding only a modest share after key controls, the findings are better read as patterns to investigate than as a causal recipe for intervention.
Paper data and sources
Original title: The role of congruence: a response surface analysis of family-school alignment in predicting creative thinking
Authors: Tat O
Journal/Repository: Frontiers in psychology
Status: Peer-reviewed
First online: 2026-10-08
DOI: Not available
Original paper · Full text