مؤسسة الشرق الأوسط للنشر العلمي
عادةً ما يتم الرد في غضون خمس دقائق
Abstract
This exploratory ecological study examines associations
between school-site distribution, school-building quality, population per
school building, and urban quality-of-life (QoL) rank standing across eight
Saudi cities. The cross-sectional analysis used 4,637 unique surveyed school
buildings and 4,085 valid geocoded sites. School-site pattern was assessed
using a Clark–Evans-style nearest-neighbor ratio (NNR), while mean building
quality and population per declared public-school building were examined as additional
city-level indicators. Exact Spearman permutation tests, Holm adjustment for
the three pre-specified hypotheses, OLS sensitivity models, influence
diagnostics, and leave-one-city-out cross-validation were used. NNR showed a
nominal positive association with QoL standing (ρ=0.738, p=0.0458), but this
was not supported after Holm correction (p=0.1375). Building quality (ρ=0.119,
p=0.793) and population per school (ρ=0.405, p=0.3268) were not supported. The
joint model had poor cross-validated performance (Q²=−0.549). NNR was also
entangled with site density and convex-hull extent, and the Oxford functional-urban-area
boundaries do not necessarily coincide with the administrative/project
boundaries of the school data. An illustrative association-based model contrast
of +2.70 standing points had a 95% CI of −2.45 to +7.84. The findings are
hypothesis-generating rather than causal or predictive and require validation
using consistent geographic boundaries, edge-corrected spatial measures,
school-age demand, and effective capacity.