Assessment of street space quality and subjective well-being mismatch and its impact, using multi-source big data

This study makes initial efforts by delineating the distribution map of the mismatch between street space quality and SWB in central Qingdao through machine learning approaches, then creatively combines ordered logistic regression and restrictive cubic spline to examine the nonlinear influence of urban variables on the mismatch based on multi-source big data. The study primarily found that low-quality spaces are concentrated in the old city area; The SWB scores of the internal space in central Qingdao are generally good and evenly distributed, while the SWB scores of the peripheral space have significant differences; Road network accessibility, green space, living convenience, and housing prices are positively correlated with SWB significantly higher than street space quality, however, land mixed use, night lighting index, and population density are negatively correlated with it. When the green space agglomeration value reaches 2.9 or exceeds 7.8, the living convenience value exceeds 12.2, and the housing price value reaches 26.6 thousand yuan/m2, improving the street space quality is most likely to enhance residents' SWB. These findings link urban spatial quality with SWB and provide support for urban further planning and regeneration to improve public SWB through targeted interventions.