Page 47 - Proceeding of Atrans Young Researcher's Forum 2019_Neat

“Transportation for A Better Life:
Smart Mobility for Now and Then”

23 August 2019, Bangkok, Thailand

Table 1 Distribution of the travel mode choices Population Population 4 4.348 1.637 0.197 10.647
a, b
among K-12 school students density density 2 (10
persons/km )
Mode Description Counts % a b at Traffic Analysis Zone level
share Adopted from MUCEP (2015)
Non- Walking, 505 41.62 3. Results and Discussion
motorized cycling 3.1 Person product moment correlation
Public Train, Jeepney, 641 52.93 Correlation is a statistical measure used to
bus, tricycle, determine how two random variables vary together.
Pedicap The correlation coefficient may be either positive or
Private Motorcycle, car 65 5.37 negative, and it ranges from -1 to +1. The coefficient
close to -1 and +1 suggests a strong relation, while
The descriptive statistics of the independent the coefficient close to zero implies no correlation.
variables are tabulated in Table 2. The data of The Person product moment correlation coefficient
socioeconomic characteristics are obtained from the was used in our study because it is the appropriate
surveyed respondents, while the home and school measure of similarity only if the two random
address of the respondents are used to merge the variables are ratio scaled. The coefficient is
secondary data of the urban form attributes. The calculated using equation 1 [12]:
distance from home to school was obtained using the
Google Map to approximate the distance, while the = ( 1 , 2 ) (1)
distance from home to the central business district 1 2
(CBD) and the urban train station was calculated
using the CDXDistance2WP function of the where ( , ) is the covariance
2
1
CDXZipStream tool. This method is more accurate between variables and , and and are
1
2
2
1
than the Euclidean and Manhattan methods. the standard deviations of random variables and
1
, respectively.
2
Table 2 Descriptive statistics of the independent Table 3 shows the correlation of the
variables independent variables with the dependent variables.
As apparent from the table, the correlation of some
Variable Description Mean SD Min Max independent variables (i.e., sex, age, family size,
Socioeconomic working adult, K-12 school children, CBD, train
Sex 1 = if male, and 0 0.487 0.500 0.000 1.000 station, line density, and population density) with the
= otherwise
Age Age of student 12.572 3.652 4.000 21.000 dependent variables (i.e., non-motorized mode,
(years) public transport mode, and private mode) were very
Family size No. of household 4.514 1.324 2.000 11.000
members trivial. It was noteworthy that the household income
(persons) had a moderate correlation with the non-motorized
Working adult No. of working 1.717 0.712 0.000 5.000
adults in a family mode (negative correlation) and the private mode
(persons) (positive correlation), but the correlation with the
K-12 school No. of K-12 1.969 0.947 0.000 5.000
children school children in public transport mode was still marginal. This
a family signals that school children originated from higher-
Household Monthly 5.658 5.048 0.250 30.000
income household income income families are more likely to commute to
4
(10 PHP/month) schools by private transport mode, and those of
No. of No. of commuters 1.582 0.754 1.000 4.000
passengers going to school lower-income households have a higher propensity
(persons)
Urban form attributes to travel by non-motorized mode. Students with the
Distance to Distance from 1.824 2.165 0.100 31.100 presence of other passengers (or commuters) are
school home to school more likely to travel by private mode. Distance from
(km)
CBD Distance from 4.849 2.424 0.361 19.088 home to school was found to have a moderate
home to CBD correlation with the dependent variables. The
(km)
Train station Distance from 0.417 0.465 0.093 5.606 negative correlation of the distance with the non-
home to train motorized mode suggests that students residing close
station (km)
Line density a, b Road public 2.416 2.447 0.000 15.422 to schools are most likely to walk or ride bicycles.
transport line On the other hand, students located far from school
density (10
2
km/km ) are likely to travel by private and public transport
modes.

22