Sun Guibo, Webster Chris, Ni Michael Y, Zhang Xiaohu
Healthy High Density Cities Lab, Faculty of Architecture, University of Hong Kong.
Geospat Health. 2018 May 7;13(1):653. doi: 10.4081/gh.2018.653.
Uncertainty with respect to built environment (BE) data collection, measure conceptualization and spatial scales is evident in urban health research, but most findings are from relatively lowdensity contexts. We selected Hong Kong, an iconic high-density city, as the study area as limited research has been conducted on uncertainty in such areas. We used geocoded home addresses (n=5732) from a large population-based cohort in Hong Kong to extract BE measures for the participants' place of residence based on an internationally recognized BE framework. Variability of the measures was mapped and Spearman's rank correlation calculated to assess how well the relationships among indicators are preserved across variables and spatial scales. We found extreme variations and uncertainties for the 180 measures collected using comprehensive data and advanced geographic information systems modelling techniques. We highlight the implications of methodological selection and spatial scales of the measures. The results suggest that more robust information regarding urban health research in high-density city would emerge if greater consideration were given to BE data, design methods and spatial scales of the BE measures.
在城市健康研究中，建筑环境（BE）数据收集、测量概念化和空间尺度方面存在不确定性是显而易见的，但大多数研究结果来自密度相对较低的环境。我们选择香港这个标志性的高密度城市作为研究区域，因为针对此类地区不确定性的研究较少。我们使用了香港一个大型人群队列中的地理编码家庭住址（n = 5732），根据国际认可的建筑环境框架为参与者的居住地提取建筑环境测量指标。绘制了测量指标的变异性，并计算了斯皮尔曼等级相关性，以评估指标之间的关系在不同变量和空间尺度上的保持程度。我们发现，使用综合数据和先进地理信息系统建模技术收集的180项测量指标存在极大的变异性和不确定性。我们强调了方法选择和测量指标空间尺度的影响。结果表明，如果更多地考虑建筑环境数据、设计方法和建筑环境测量指标的空间尺度，将能获得关于高密度城市中城市健康研究更可靠的信息。
