Validation of a Japanese walkability index using large-scale step count data of Yokohama citizens

Validation of a Japanese walkability index using large-scale step count data of Yokohama citizens
复制标题

使用横滨市民的大规模步数数据验证日本步行适宜性指数

DOI:
10.1016/j.cities.2022.103614
复制
发表时间:
2022
期刊:
影响因子:
6.7
通讯作者:
Shimizu Chihiro
Shimizu Chihiro
中科院分区:
经济学1区
文献类型:
--
作者:
Hino Kimihiro;Baba Hiroki;Kim Hongjik;Shimizu Chihiro

文献摘要

相似文献

在人口老龄化和人口下降的日本,有必要评估和改善城市的步行能力。然而,在亚洲人口高度密集的城市中,使用为美国城市开发的指数可能并不合适。这项研究旨在验证新开发的可步行指数与横滨市民步数(n=54698)之间的关联。该指数量化了各种城市便利设施的可达性。参与者根据性别和年龄被分成四个小组。计算了每个参与者所在社区的新指数和现有指数的分数。使用广义线性混合模型,检验分数与其步数(计数模型)和实现目标的概率(二项模型)之间的关联。总体而言,新的指数预测公民的步数在两个模型中都比现有的指数更好。在计数模型中,新指数与老年人的步数有更强的相关性,在二项模型中,新指数与女性的目标成就率有更强的相关性。这些发现支持了新指数在评估高密度城市的步行能力方面的有效性,以及根据城市背景定制指数的必要性。新指数指的是各种便利设施,而不是密度变量。
In Japan where the population is aging and declining, it is necessary to evaluate and improve the walkability of cities. However, it may not be appropriate to use the indices developed for American cities in highly dense cities in Asia. This study aimed to verify the association of a newly developed walkability index, which quantifies the accessibility of various urban amenities, with step counts of Yokohama citizens (n= 54,698). Participants were categorized into four subgroups according to sex and age. Scores from both the new and an existing index were calculated for each participant's neighborhood. Using generalized linear mixed models, associations of the scores with their step counts (count models) and probability of target achievement (binomial models) were examined. Overall, the new index predicted citizens' step counts better than the existing index in both models. The new index had a stronger association with step count in older adults in the count models and with target achievement in women in the binomial models. The findings support the efficacy of the new index, which indexes various amenities instead of density variables, in assessing the walkability in highly dense cities, as well as the need for indices tailored to the urban context.