Journal of Geographical Sciences >
Exploring zonal heterogeneities of primary school students’ commute-mode choices through a geographically weighted regression model
Wu Dawei (1999-), PhD Candidate, specialized in transport geography and travel behaviour. E-mail: wdw808@bjtu.edu.cn |
Received date: 2023-03-26
Accepted date: 2023-11-03
Online published: 2024-04-24
Supported by
National Natural Science Foundation of China(71971023)
Beijing Social Science Foundation(21DTR055)
Commuting is an important part of primary school students’ travel behavior, which has been concerned for a long time. We found that the commute-mode choice behavior of primary school students in the context of regional segmentation shows strong characteristics in space, but has not yet been considered in traditional research. To fill this gap, this study summarizes the commute-mode choices of primary school students with different characteristics based on the Beijing School Commute Survey. And the geographically weighted regression (GWR) model is built to analyze the zonal heterogeneity of the impact of personal characteristics, family factors and school factors on the primary school students’ commute- mode choices from a low-carbon perspective. The results demonstrate that the possibility of primary school students choosing low-carbon commuting modes is positively correlated with the grade, commuting time, commuting escort type and housing category, but is inversely related with the commuting distance and the distance from the school to the city center. The coefficient estimates of explanatory variables vary across regions. Finally, we put forward policy suggestions regarding promoting the low-carbon commuting structure, such as developing the urban slow traffic system, which serve as a reference for policymakers.
WU Dawei , MA Lu , YAN Xuedong . Exploring zonal heterogeneities of primary school students’ commute-mode choices through a geographically weighted regression model[J]. Journal of Geographical Sciences, 2024 , 34(4) : 804 -833 . DOI: 10.1007/s11442-024-2228-9
Figure 1 Map of central urban area of Beijing (a) and TAZs (b) |
Figure 2 Spatial distribution of schools and number of samples |
Table 1 Part of the results of the commuting survey among primary school students |
Object ID | Mode | School | Student | Family | ||||||
---|---|---|---|---|---|---|---|---|---|---|
Name | Longitude | Latitude | Region | Grade | Commuting time | Commuting distance | Housing category | Commuting escort type | ||
498** | Walk | B** | 116.36** | 39.9** | Xicheng | Five | AM7:30 | 1.167 km | Self-owned | None |
39** | Bicycle | T** | 116.32** | 40.0** | Haidian | Six | AM7:15 | 2.709 km | Tenancy | None |
304** | Car | T** | 116.32** | 40.0** | Haidian | Six | AM7:30 | 3.454 km | Self-owned | Parents |
Note: To protect the privacy of students, object ID, name, longitude and latitude of the school are not fully displayed in this table. |
Figure 3 Low carbon characteristics of commuting modes of primary school students in different regions |
Figure 4 Spine chart of primary school students’ grade (a), commuting time (b) and the commuting mode choice (Note: 1 to 11 in the vertical coordinate represent walk, bicycle, two-wheel electric bicycle, three-wheel electric bicycle, bus, subway, school bus, shuttle bus, motorcycle, taxi and car, respectively.) |
Figure 5 Violin chart of primary school students’ escort type (a), housing category (b) and the commuting mode choice (Note: 1 to 11 in the vertical coordinate represent walk, bicycle, two-wheel electric bicycle, three-wheel electric bicycle, bus, subway, school bus, shuttle bus, motorcycle, taxi and car, respectively.) |
Figure 6 Distribution of commuting distance among primary school students |
Table 2 Description of the dependent variable and explanatory variable |
Type | Attribution | Name | Definition | Assignment | Category |
---|---|---|---|---|---|
Dependent variable | Classified variable | Commuting mode | Primary school students’ commute-mode choices | 1 | Walk |
2 | Bicycle | ||||
3 | Two-wheel electric bicycle | ||||
4 | Three-wheel electric bicycle | ||||
5 | Bus | ||||
6 | Subway | ||||
7 | School bus | ||||
8 | Shuttle bus | ||||
9 | Motorcycle | ||||
10 | Taxi | ||||
11 | Car | ||||
Explanatory variable | Classified variable | Grade | Grades of primary school students | 4 | Grade four |
5 | Grade five | ||||
6 | Grade six | ||||
Classified variable | Commuting time | Primary school students’ commuting time | 2 | Commuting time from 6:00 to 6:59 | |
3 | Commuting time from 7:00 to 7:59 | ||||
4 | Commuting time from 8:00 to 8:59 | ||||
5 | Commuting time after 9:00 | ||||
Classified variable | Commuting distance | Primary school students’ commuting distance | [1, 20) | Commuting distance range from 1 to 20 km | |
Classified variable | Distance from the city center | Distance from the city center to school | [1.6, 53.9] | Distance from the city center to school range from 1.6 to 53.9 km | |
Classified variable | Commuting escort type | Primary school students’ commuting escort type | 1 | Parents | |
2 | None | ||||
3 | Elderly | ||||
Classified variable | Housing category | Housing categories of primary school students’ families | 1 | Self-owned | |
2 | Tenancy | ||||
3 | Lodge |
Note: The center of Beijing is Tiananmen Square at coordinates (39.9087N, 116.3974E). |
Table 3 Comparison of fitting effect between OLS and GWR |
AICc | R2 | |
---|---|---|
OLS | 112456.9 | 0.2733 |
GWR | 112234.8 | 0.4933 |
Note: R2 of the GWR is the optimal value. |
Figure 7 Spatial distribution of standard residuals for the GWR model |
Table 4 Estimated Value of the GWR Model for the commuting mode of primary school students |
Variable | MEAN | MIN | LQ | MED | UQ | MAX |
---|---|---|---|---|---|---|
Grade | -0.0767 | -0.2053 | -0.0781 | -0.0732 | -0.0697 | -0.0026 |
Distance from the city center | 0.0423 | -0.0409 | 0.0355 | 0.0431 | 0.0504 | 0.0820 |
Commuting time | -1.4454 | -2.2218 | -1.4709 | -1.4489 | -1.4062 | -1.1439 |
Commuting escort type | -1.5022 | -1.7208 | -1.5189 | -1.4931 | -1.4809 | -0.8113 |
Commuting distance | 0.3081 | 0.2225 | 0.3057 | 0.3061 | 0.3076 | 0.3336 |
Housing category | -1.0908 | -1.2200 | -1.1242 | -1.0991 | -1.0803 | -0.0617 |
Figure 8 Spatial distribution of average coefficients of the grade (a) and commuting distance (b) for the GWR model |
Figure 9 Spatial distribution of average coefficients of commuting time (a) and the distance from the city center (b) for the GWR model |
Figure 10 Spatial distribution of average estimation coefficients of commuting escort types (a) and housing categories (b) for the GWR model |
Table 5 The division and assignment of commuting modes based on the sample size |
Commuting mode | Category | Assignment |
---|---|---|
Walk | Strong low carbon | 1 |
Bicycle | Slightly strong low carbon | 2 |
Two-wheel electric bicycle | ||
Three-wheel electric bicycle | Slightly weak low carbon | 3 |
Bus | ||
Subway | ||
School bus | ||
Motorcycle | ||
Shuttle bus | ||
Taxi | ||
Car | Weak low carbon | 4 |
Figure 11 Spatial distribution of average estimation coefficients of commuting escort types in classification methods based on the sample type (a) and the sample size (b) |
Table A1 VIF values of explanatory variables of OLS model |
Variable | Coef. | P-value | VIF |
---|---|---|---|
Grade | -0.057 | 0.047 | 1.021 |
Distance from the city center | 0.034 | 0.000 | 1.024 |
Commuting time | -1.311 | 0.000 | 1.135 |
Commuting escort type | -1.542 | 0.000 | 1.039 |
Commuting distance | 0.317 | 0.000 | 1.159 |
Housing category | -1.181 | 0.000 | 1.037 |
Table A2 Results of spatial autocorrelation test on explanatory variables |
Variable | Moran’s I | Z-score | P-value |
---|---|---|---|
Grade | 0.094 | 103.306 | 0.000 |
Distance from the city center | 0.651 | 2520.812 | 0.000 |
Commuting time | 0.152 | 170.284 | 0.000 |
Commuting escort type | 0.044 | 49.300 | 0.000 |
Commuting distance | 0.177 | 197.302 | 0.000 |
Housing category | 0.287 | 320.557 | 0.000 |
Table A3 The legend names and descriptions of figures |
Figure | Legend name | Description |
---|---|---|
Figure 1 | Map of the central urban area of Beijing and TAZs | Location of the central urban area of Beijing and the division results of TAZs |
Figure 2 | The sample size | The sample size contained at each dot |
Figure 3 | The commuting mode coefficients | The commuting mode coefficients of primary school students in different regions |
Figure 7 | The standard residuals | The standard residuals for the GWR model |
Figure 8 | The coefficients for the model | The coefficients of the grade and commuting distance for the model |
Figure 9 | The coefficients for the model | The coefficients of the distance from the city center and commuting time for the model |
Figure 10 | The coefficients for the model | The coefficients of commuting escort types and housing categories for the model |
Figure 11 | The coefficients for the model | The coefficients of commuting escort types for the model |
[1] |
|
[2] |
Beijing Municipal Education Commission, 2022. Regulations of Beijing on Safety Management of Primary and Middle School Kindergarten for trial implementation. http://jw.beijing.gov.cn/jyzx/ztzl/bjjyxx/jyxxzcfg/202207/t20220706_2764772.html.
|
[3] |
Beijing Xiaoshengchu Network BXN, 2022. Basic information on the number of primary school graduates and the number of students in all grades in Beijing in the 2021-2022 academic year. https://www.xschu.com/xiaoshengchu/17/44263.html.
|
[4] |
Beijing Youshengxiao Network BYN, 2022. Number of primary schools and classes in each grade in Beijing in the 2021-2022 school year. https://www.ysxiao.cn/c/202203/51422.html.
|
[5] |
|
[6] |
|
[7] |
|
[8] |
|
[9] |
|
[10] |
|
[11] |
|
[12] |
|
[13] |
|
[14] |
|
[15] |
|
[16] |
|
[17] |
|
[18] |
|
[19] |
|
[20] |
|
[21] |
|
[22] |
|
[23] |
|
[24] |
|
[25] |
|
[26] |
|
[27] |
|
[28] |
|
[29] |
|
[30] |
|
[31] |
|
[32] |
|
[33] |
|
[34] |
|
[35] |
|
[36] |
|
[37] |
|
[38] |
|
[39] |
|
[40] |
|
[41] |
|
[42] |
|
[43] |
|
[44] |
|
[45] |
|
[46] |
|
[47] |
|
[48] |
|
[49] |
|
[50] |
|
[51] |
|
[52] |
|
[53] |
|
[54] |
|
[55] |
|
[56] |
|
[57] |
|
[58] |
|
[59] |
|
[60] |
|
[61] |
|
[62] |
|
[63] |
|
[64] |
|
[65] |
|
[66] |
|
[67] |
|
[68] |
|
[69] |
|
[70] |
|
[71] |
|
[72] |
Netease, 2022a. Haidian builds the urban slow traffic system, and the first bicycle dedicated road will be developed in the south. https://www.163.com/dy/article/HB7KKBQS0552UKSD.html.
|
[73] |
Netease, 2022b. Ranking of top 100 cities in China’s GDP in 2021. https://www.163.com/dy/article/ H1QK7JUM0552NMG9.html.
|
[74] |
|
[75] |
|
[76] |
|
[77] |
|
[78] |
People’s Government of Beijing Municipality PGBM, 2022. Main data of the seventh national population census in Beijing. http://www.beijing.gov.cn/gongkai/shuju/sjjd/202105/t20210519_2392877.html.
|
[79] |
|
[80] |
|
[81] |
|
[82] |
|
[83] |
|
[84] |
|
[85] |
|
[86] |
|
[87] |
|
[88] |
|
[89] |
|
[90] |
|
[91] |
|
[92] |
|
[93] |
|
[94] |
Tencent, 2022. What are the high-quality schools in Chaoyang District—Elementary School. https://new. qq.com/rain/a/20220107A05Q5L00.
|
[95] |
|
[96] |
|
[97] |
|
[98] |
|
[99] |
|
[100] |
|
[101] |
|
[102] |
|
[103] |
|
[104] |
Wikipedia, 2022. Beijing Haidian Experimental Primary School. https://zh.wikipedia.org/wiki/%E5%8C%97% E4%BA%AC%E5%B8%82%E6%B5%B7%E6%B7%80%E5%8C%BA%E5%AE%9E%E9%AA%8C%E5%B0%8F%E5%AD%A6.
|
[105] |
|
[106] |
|
[107] |
|
[108] |
|
[109] |
YNET, 2022. Dongcheng district of Beijing promotes the reconstruction of the old city: Old hutongs usher in a new life. https://t.ynet.cn/baijia/32753188.html.
|
[110] |
|
[111] |
|
[112] |
|
[113] |
|
[114] |
|
[115] |
|
[116] |
|
/
〈 |
|
〉 |