建成环境与青少年步行通学的非线性关系——基于极限梯度提升模型的研究

doi:10.18306/dlkxjz.2022.02.006

Abstract

Abstract:

Walking is not only a primitive and convenient transport mode but also an important integrant of physical activity, which is beneficial for the promotion of public health, alleviation of traffic congestion, and mitigation of transportation-induced pollution. In modern China, cities are expanding rapidly, people are enjoying a dramatic improvement in living standards, and the pace of life is accelerating. As a result, urban people, including adolescents, tend to travel in motorized modes increasingly more and walk less. The prevalence of physical inactivity among adolescents has brought about a series of health issues, such as deterioration of physical fitness, obesity, and some non-communicable diseases (for example, diabetes and hypertension). Travel to school is among the most important routine travels for adolescents. Promoting adolescents' propensity of walking to school can effectively help them integrate physical activity into daily life and thus enhance their overall physical activity level. Hence, scholars from diverse disciplines (for example, geography, urban planning, and public health) have been drawn to examine the relationships between the built environment and walking to school. However, the current research is insufficient in the following two aspects. First, the existing research is mainly based on the Western context, whereas few studies have been conducted in China. Second, the majority of existing studies assumed a linear or generalized linear (for example, log-linear) relationship between the built environment and walking to school, and no studies, to the best of our knowledge, have examined the non-linear relationships between them. Therefore, this study, taking Xiamen, China as the case and employing its large-scale travel behavior survey dataset in 2015, explored the non-linear effects of the built environment on adolescents' propensity of walking to school. We applied a state-of-the-art machine learning method, namely extreme gradient boosting method (XGBoost), to fit the model, and interpreted the model with relative importance and partial dependence plots. The results show that: 1) Distance from home to school is the most important factor influencing walking to school, with the relative importance of 39.99%. 2) The built environment, which is characterized by the 5Ds (density, diversity, design, destination accessibility, and distance to transit) model, is an important contributor, and relative contributions of the built environment variables at home and school collectively contributed 36.28% of the model's explanatory power, only second to distance to school, much higher than that of sociodemographic variables (23.73%). Distance to city center and population density around both home and school contribute a great deal. 3) All the built environment variables at both ends of school trips and the key sociodemographic variables have non-linear effects on adolescents' propensity of walking to school, and there exist obvious threshold effects. This study can inform decision makers with nuanced policy insights for promoting adolescents' behavior of walking to school.

Key words: built environment, walking to school, non-linearity, gradient boosting decision tree, machine learning, Xiamen Island, China

LIU Jixiang, XIAO Longzhu, ZHOU Jiangping, GUO Yuanyuan, YANG Linchuan. Non-linear relationships between the built environment and walking to school: Applying extreme gradient boosting method[J].PROGRESS IN GEOGRAPHY, 2022, 41(2): 251-263.

Figures/Tables 6

Fig.1

Tab.1

Tab.2

Tab.3

Fig.2

Fig.3

References 44

[1]	Zhu X M, Lee C. Correlates of walking to school and implications for public policies: Survey results from parents of elementary school children in Austin, Texas[J]. Journal of Public Health Policy, 2009, 30(S1):S177-S202. doi: 10.1057/jphp.2008.51
[2]	姜玉培, 甄峰, 王文文, 等. 城市建成环境对居民身体活动的影响研究进展与启示[J]. 地理科学进展, 2019, 38(3):357-369. doi: 10.18306/dlkxjz.2019.03.006
	[ Jiang Yupei, Zhen Feng, Wang Wenwen, et al. Influence of urban built environment on residents' physical activity: Review and implications. Progress in Geography, 2019, 38(3):357-369. ]
[3]	Lee I-M, Buchner D M. The importance of walking to public health[J]. Medicine and Science in Sports and Exercise, 2008, 40(S7):S512-S518. doi: 10.1249/MSS.0b013e31817c65d0
[4]	王依茹, 王琛, 曾金迪. 个体与环境交互作用下中国成人超重肥胖情况变化趋势及影响因素研究[J]. 地理科学进展, 2020, 39(1):100-110. doi: 10.18306/dlkxjz.2020.01.010
	[ Wang Yiru, Wang Chen, Zeng Jindi. Study on the trend and influencing factors of overweight and obesity in Chinese adults under interactions of individual and environment. Progress in Geography, 2020, 39(1):100-110. ]
[5]	Guthold R, Stevens G A, Riley L M, et al. Global trends in insufficient physical activity among adolescents: A pooled analysis of 298 population-based surveys with 1.6 million participants[J]. The Lancet Child & Adolescent Health, 2020, 4(1):23-35.
[6]	Dong Y, Lau P W, Dong B, et al. Trends in physical fitness, growth, and nutritional status of Chinese children and adolescents: A retrospective analysis of 1.5 million students from six successive national surveys between 1985 and 2014[J]. The Lancet Child & Adolescent Health, 2019, 3(12):871-880.
[7]	梁璇. 中日青少年体质健康数字“差”的背后 [N/OL]. 中国青年报, 2017-06-26(8) [2021-01-01]. http://zqb.cyol.com/html/2017-06/26/nw.D110000zgqnb_20170626_1-08.htm.
	[ Liang Xuan. Behind the gap of physical fitness between Chinese and Japanese adolescents. China Youth Daily, 2017-06-26(8) [2021-01-01]. http://zqb.cyol.com/html/2017-06/26/nw.D110000zgqnb_20170626_1-08.htm. ]
[8]	王敬东. 中国儿童肥胖筛查共识发布, 明确儿童体重超标与严重超标标准 [EB/OL]. 央视网, 2019-09-23 [2021-01-01]. http://news.cctv.com/2019/09/22/ARTIYm98cMCfYj5bNPmK2K0A190922.shtml.
	[ Wang Jingdong. The consensus on childhood obesity screening in China has been released to clarify the criteria for overweight and severely overweight children. CCTV.com, 2019-09-23 [2021-01-01]. http://news.cctv.com/2019/09/22/ARTIYm98cMCfYj5bNPmK2K0A190922.shtml. ]
[9]	马淑婧, 羊柳, 赵敏, 等. 1991—2015年中国儿童青少年血压水平及高血压检出率的变化趋势[J]. 中华流行病学杂志, 2020, 41(2):178-183.
	[ Ma Shujing, Yang Liu, Zhao Min, et al. Changing trends in the levels of blood pressure and prevalence of hypertension among Chinese children and adolescents from 1991 to 2015. Chinese Journal of Epidemiology, 2020, 41(2):178-183. ]
[10]	刘吉祥, 周江评, 肖龙珠, 等. 建成环境对步行通勤通学的影响: 以中国香港为例[J]. 地理科学进展, 2019, 38(6):807-817. doi: 10.18306/dlkxjz.2019.06.002
	[ Liu Jixiang, Zhou Jiangping, Xiao Longzhu, et al. Effects of the built environment on pedestrian communing to work and school: The Hong Kong case, China. Progress in Geography, 2019, 38(6):807-817. ]
[11]	Ewing R, Cervero R. Travel and the built environment: A meta-analysis[J]. Journal of the American Planning Association, 2010, 76(3):265-294. doi: 10.1080/01944361003766766
[12]	Moran M R, Plaut P, Baron-Epel O. Do children walk where they bike? Exploring built environment correlates of children's walking and bicycling[J]. Journal of Transport and Land Use, 2015, 9(2). doi: 10.5198/jtlu.2015.556. doi: 10.5198/jtlu.2015.556
[13]	Ito K, Reardon T G, Arcaya M C, et al. Built environment and walking to school: Findings from a student travel behavior survey in Massachusetts[J]. Transportation Research Record: Journal of the Transportation Research Board, 2017, 2666(1):78-84.
[14]	Rothman L, To T, Buliung R, et al. Influence of social and built environment features on children walking to school: An observational study[J]. Preventive Medicine, 2014, 60:10-15. doi: 10.1016/j.ypmed.2013.12.005 pmid: 24333604
[15]	Giles-Corti B, Wood G, Pikora T, et al. School site and the potential to walk to school: The impact of street connectivity and traffic exposure in school neighborhoods[J]. Health & Place, 2011, 17(2):545-550.
[16]	Yang Y Y, Lu Y, Yang L C, et al. Urban greenery, active school transport, and body weight among Hong Kong children[J]. Travel Behaviour and Society, 2020, 20:104-113. doi: 10.1016/j.tbs.2020.03.001
[17]	Committee on Environmental Health, Tester J M. The built environment: Designing communities to promote physical activity in children[J]. Pediatrics, 2009, 123(6):1591-1598. doi: 10.1542/peds.2009-0750 pmid: 19482771
[18]	McDonald N C. Children's mode choice for the school trip: The role of distance and school location in walking to school[J]. Transportation, 2008, 35(1):23-35. doi: 10.1007/s11116-007-9135-7
[19]	Rothman L, Macarthur C, To T, et al. Motor vehicle-pedestrian collisions and walking to school: The role of the built environment[J]. Pediatrics, 2014, 133(5):776-784. doi: 10.1542/peds.2013-2317 pmid: 24709929
[20]	Mehdizadeh M, Nordfjaern T, Mamdoohi A R, et al. The role of parental risk judgements, transport safety attitudes, transport priorities and accident experiences on pupils' walking to school[J]. Accident Analysis & Prevention, 2017, 102:60-71. doi: 10.1016/j.aap.2017.02.020
[21]	Rossen L M, Pollack K M, Curriero F C, et al. Neighborhood incivilities, perceived neighborhood safety, and walking to school among urban-dwelling children[J]. Journal of Physical Activity and Health, 2011, 8(2):262-271. doi: 10.1123/jpah.8.2.262
[22]	Yu C-Y, Zhu X M. Impacts of residential self-selection and built environments on children's walking-to-school behaviors[J]. Environment and Behavior, 2015, 47(3):268-287. doi: 10.1177/0013916513500959
[23]	Yu C-Y, Zhu X M. From attitude to action: What shapes attitude toward walking to/from school and how does it influence actual behaviors?[J]. Preventive Medicine, 2016, 90:72-78. doi: 10.1016/j.ypmed.2016.06.036
[24]	Liu J X, Xiao L Z, Yang L C, et al. A tale of two social groups in Xiamen, China: Trip frequency of migrants and locals and its determinants[J]. Travel Behaviour and Society, 2020, 20:213-224. doi: 10.1016/j.tbs.2020.04.001
[25]	Stevens M R. Does compact development make people drive less?[J]. Journal of the American Planning Association, 2017, 83(1):7-18. doi: 10.1080/01944363.2016.1240044
[26]	Ewing R, Cervero R. "Does compact development make people drive less?" The answer is yes[J]. Journal of the American Planning Association, 2017, 83(1):19-25. doi: 10.1080/01944363.2016.1245112
[27]	Handy S. Thoughts on the meaning of Mark Stevens's meta-analysis[J]. Journal of the American Planning Association, 2017, 83(1):26-28. doi: 10.1080/01944363.2016.1246379
[28]	Ding C, Cao X Y, Wang Y P. Synergistic effects of the built environment and commuting programs on commute mode choice[J]. Transportation Research Part A: Policy and Practice, 2018, 118:104-118. doi: 10.1016/j.tra.2018.08.041
[29]	Yang L C, Ao Y B, Ke J T, et al. To walk or not to walk? Examining non-linear effects of streetscape greenery on walking propensity of older adults[J]. Journal of Transport Geography, 2021, 94:103099. doi: 10.1016/j.jtrangeo.2021.103099. doi: 10.1016/j.jtrangeo.2021.103099
[30]	Cheng L, Chen X W, De Vos J, et al. Applying a random forest method approach to model travel mode choice behavior[J]. Travel Behaviour and Society, 2019, 14:1-10. doi: 10.1016/j.tbs.2018.09.002
[31]	Kim K, Kwon K, Horner M W. Examining the effects of the built environment on travel mode choice across different age groups in Seoul using a random forest method[J]. Transportation Research Record: Journal of the Transportation Research Board, 2021, 2675(8):670-683.
[32]	Parsa A B, Movahedi A, Taghipour H, et al. Toward safer highways, application of XGBoost and SHAP for real-time accident detection and feature analysis[J]. Accident Analysis & Prevention, 2020, 136:105405. doi: 10.1016/j.aap.2019.105405. doi: 10.1016/j.aap.2019.105405
[33]	Cheng L, De Vos J, Zhao P J, et al. Examining non-linear built environment effects on elderly's walking: A random forest approach[J]. Transportation Research Part D: Transport and Environment, 2020, 88:102552. doi: 10.1016/j.trd.2020.102552. doi: 10.1016/j.trd.2020.102552
[34]	Ding C, Cao X Y, Næss P. Applying gradient boosting decision trees to examine non-linear effects of the built environment on driving distance in Oslo[J]. Transportation Research Part A: Policy and Practice, 2018, 110:107-117. doi: 10.1016/j.tra.2018.02.009
[35]	Tao T, Wang J Y, Cao X Y. Exploring the non-linear associations between spatial attributes and walking distance to transit[J]. Journal of Transport Geography, 2020, 82:102560. doi: 10.1016/j.jtrangeo.2019.102560. doi: 10.1016/j.jtrangeo.2019.102560
[36]	Wu X Y, Cao X Y, Ding C. Exploring rider satisfaction with arterial BRT: An application of impact asymmetry analysis[J]. Travel Behaviour and Society, 2020, 19:82-89. doi: 10.1016/j.tbs.2019.12.006
[37]	自然资源保护协会, 清华大学建筑学院. 中国城市步行友好性评价: 基于街道功能促进步行的研究[R]. 北京, 2017.
	[Natural Resources Defense Council, School of Architecture of Tsinghua University. Appraisal of walking friendliness of Chinese cities. Beijing, China. 2017. ]
[38]	Song Y, Merlin L, Rodriguez D. Comparing measures of urban land use mix[J]. Computers, Environment and Urban Systems, 2013, 42:1-13. doi: 10.1016/j.compenvurbsys.2013.08.001
[39]	Chen T Q, Guestrin C. XGBoost: A scalable tree boosting system[C/OL]// KDD '16: Proceedings of the 22nd ACM SIGKDD international conference on knowledge discovery and data mining. San Francisco, USA: Association for Computing Machinery, 2016:785-794. doi: 10.1145/2939672.2939785. doi: 10.1145/2939672.2939785
[40]	Hox J. Multilevel modeling: When and why. Classification, data analysis, and data highways[M]. Berlin, Germany: Springer. 1998: 147-154.
[41]	Friedman J H. Greedy function approximation: A gradient boosting machine[J]. The Annals of Statistics, 2001, 29(5):1189-1232. doi: 10.1214/aos/1013203450
[42]	Su S L, Pi J H, Xie H, et al. Community deprivation, walkability, and public health: Highlighting the social inequalities in land use planning for health promotion[J]. Land Use Policy, 2017, 67:315-326. doi: 10.1016/j.landusepol.2017.06.005
[43]	厦门市统计局. 2020厦门经济特区年鉴[M]. 厦门: 厦门市统计局, 2020.
	[Xiamen Burear of Statistics. Statistic yearbook of Xiamen Special Economic Zone in 2020. Xiamen, China: Xiamen Burear of Statistics, 2020.]
[44]	Lin P F, Weng J C, Brands D K, et al. Analysing the relationship between weather, built environment, and public transport ridership[J]. IET Intelligent Transport Systems, 2020, 14(14):1946-1954. doi: 10.1049/itr2.v14.14

变量	定义/描述	均值(标准差)或百分比
社会经济属性
年龄	受访者年龄(岁)	11.09 (8.32)
性别	1=女,0=男	0=48.0%, 1=52.0%
住房面积	家庭住房面积(m²)	79.77 (46.37)
住房性质	分类变量(1=自有, 2=单位住房, 3=租住)	1=81.3%, 2=4.5%, 3=14.2%
家庭规模	家庭成员数量(人)	4.85 (3.22)
户口	1=有厦门户口,2=无厦门户口	0=9.8%, 1=90.2%
拥有小汽车	1=所在家庭拥有至少1辆小汽车,0=无	0=83.7%, 1=16.3%
父/母以步行通勤	1=父母亲至少一人以步行为通勤方式,0=无	0=81.17%, 1=18.83%
有人接送	1=有长辈接送,0=无	0=70.11%, 1=29.89%
出行特征
通学距离	受访者通学实际距离(km)	1.24 (1.68)
建成环境
人口密度	人口数量/TAZ面积(人/km²)	13194.87 (10325.10)
容积率	总建筑面积/TAZ面积	0.88 (0.56)
土地利用混合度	14种主要的土地利用类型(包括居住、工业、教育、行政办公、休闲娱乐、公共开放空间等)的数量及比例,采用改良版的熵值法计算	0.58 (0.16)
交叉口密度	道路交叉口数量/ TAZ面积(个/km²)	82.88 (64.35)
公交密度	公交线路数量/ TAZ面积(条/km²)	69.44 (62.78)
离市中心距离	从TAZ中心到CBD(即中山路)的路网距离(km)	8.78 (4.07)

模型	操作平台	最佳迭代数	训练误差	测试误差	预测准确率/%	AUC值
XGBoost	R, “XGBoost”	9931	0.1258	0.1713	82.88	0.892
GBDT	R, “caret”	6554	0.2188	0.2149	78.51	0.816
RF	R, “randomForest”	500	0.1561	0.1789	82.11	0.848
LightGBM	Python, “lightGBM”	19821	0.1610	0.1937	80.63	0.803
Adaboost	R, “adabag”	500	0.1686	0.1923	80.77	0.798
Logistic	R, “glm”	—	0.2243	0.2396	76.04	0.743

变量	相对重要性/%	排名
出行特征
通学距离	39.99	1
社会经济属性
住房面积	7.84	2
年龄	7.13	3
有人接送	2.52	15
家庭规模	1.42	17
性别	1.14	18
拥有小汽车	1.13	19
父/母以步行通勤	0.97	20
户口	0.81	21
住房性质	0.38	22
小计	23.73
出发地(家)建成环境
土地利用混合度	3.67	4
离市中心距离	3.59	5
道路交叉口密度	3.50	6
容积率	3.28	8
公交站点密度	2.90	9
人口密度	2.88	10
小计	19.82
目的地(校)建成环境
道路交叉口密度	3.45	7
离市中心距离	2.82	11
公交站点密度	2.70	12
容积率	2.69	13
土地利用混合度	2.56	14
人口密度	2.24	16
小计	16.46

Non-linear relationships between the built environment and walking to school: Applying extreme gradient boosting method

RichHTML

PDF (PC)

Like

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 6

References 44

Related Articles 15

Metrics

Comments

Recommended 0

[1]	LONG Dongping, XU Ming'en, LIU Lin. Spatio-temporal evolution and regional types of China’s sales-based counterfeit drug crimes [J]. PROGRESS IN GEOGRAPHY, 2023, 42(5): 944-959.
[2]	ZHANG Xuezhen, HE Qingwen, HUANG Jixia. Spatio-temporal variability of forest pest disasters in China from 1985 to 2018 by Meta-analysis and influencing factors [J]. PROGRESS IN GEOGRAPHY, 2023, 42(5): 960-970.
[3]	LUO Wanlu, WANG Wulin, LIN Zhen, ZHOU Weijian. Spatiotemporal evolution and driving factors of urban-rural integration in China [J]. PROGRESS IN GEOGRAPHY, 2023, 42(4): 629-643.
[4]	WANG Zimeng, LIU Yanfang, LUO Xuan, TONG Zhaomin, AN Rui. Nonlinear relationship between urban vitality and the built environment based on multi-source data: A case study of the main urban area of Wuhan City at the weekend [J]. PROGRESS IN GEOGRAPHY, 2023, 42(4): 716-729.
[5]	YANG Hong, SHEN Lifan, HU Yiwen, ZHANG Yuanhang, PENG Jiandong. Spatial and temporal characteristics of elderly people’s metro travel behavior and its non-linear relationship with the built environment: A case study of Wuhan City [J]. PROGRESS IN GEOGRAPHY, 2023, 42(3): 491-504.
[6]	PENG Baoyu, ZHANG Bo, SUN Wei. Evolution and prospect of financial inclusion research in China from the perspective of financial geography [J]. PROGRESS IN GEOGRAPHY, 2023, 42(2): 406-416.
[7]	ZHANG Pei, WANG Jiao'e, XIAO Fan. Spatial evolution and determinants of new infrastructure development in China [J]. PROGRESS IN GEOGRAPHY, 2023, 42(2): 209-220.
[8]	CHU Jun, TONG De, GU Hengyu, CAO Zhiqiang, PAN Xiangxiang. Nonlinear characteristics of distance-decay parameter in intercity links: Based on human mobility big data [J]. PROGRESS IN GEOGRAPHY, 2023, 42(1): 1-11.
[9]	LI Xianzheng, ZHAO Zhenbin, LIU Yang, ZHANG Dazhao, ZHANG Jian, ZHANG Yuqian. Spatial measurement of commercial homogeneity of ancient town destination based on street view images: A case study of Dali ancient city [J]. PROGRESS IN GEOGRAPHY, 2023, 42(1): 104-115.
[10]	LIU Hua, ZHOU Tianyuan, ZHANG Guike, LIU Xuejun, DENG Shaohui. Influence of the built environment on residents’ migration intensity from the life cycle perspective：Based on multi-source data of Wuhan City [J]. PROGRESS IN GEOGRAPHY, 2023, 42(1): 66-78.
[11]	WANG Chenggang, WANG Bo, WANG Qizhi, LEI Yaqin. Nonlinear associations between urban vitality and built environment factors and threshold effects: A case study of central Guangzhou City [J]. PROGRESS IN GEOGRAPHY, 2023, 42(1): 79-88.
[12]	WEI Yue, YANG Dongfeng. Impact of neighborhood destination built environments on daily activity opportunities of the elderly population: Based on perception and participation [J]. PROGRESS IN GEOGRAPHY, 2023, 42(1): 89-103.
[13]	GONG Qinlin, ZOU Donghan, ZHOU Yi, ZHU Shengjun. Influence of high-level scenic spots on tourism development and its spatial effect [J]. PROGRESS IN GEOGRAPHY, 2022, 41(8): 1364-1377.
[14]	SUN Pingjun, ZHANG Keqiu, HE Tian. Shrinkage effect of urban-rural integration on shrinking cities in the three provinces of Northeast China and mechanism [J]. PROGRESS IN GEOGRAPHY, 2022, 41(7): 1213-1225.
[15]	ZHANG Fan, ZHANG Yongyong, CHEN Junxu, ZHAI Xiaoyan, HU Qingfang. Performance of multiple machine learning model simulation of process characteristic indicators of different flood types [J]. PROGRESS IN GEOGRAPHY, 2022, 41(7): 1239-1250.