Page 48 - 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 3 Correlation between the independent variables, but the coefficients cannot determine how
variables and the dependent variables the independent variables affect the probability of
choosing travel mode. Correspondingly, the
Variable Non- Public Private conditional probability was used in the subsection
motorized below.
Sex 0.06 -0.05 -0.03
Age -0.09 0.09 -0.01
Family size 0.01 0.01 -0.02 3.2 Conditional Probability
Working adult -0.04 0.05 -0.02 The three independent variables (i.e.,
K-12 school 0.04 -0.04 0.01 household income, number of passengers, and
children distance from home to school) were found to have
Household income -0.26 0.10 0.30 moderate correlations with the commute mode
No. of passengers -0.07 -0.15 0.48 choices. These three independent variables were
Distance to school -0.40 0.29 0.20 selected to study conditional probabilities of the
CBD 0.02 -0.05 0.05 travel mode choices. The conditional probability is
Train station -0.06 0.06 -0.01 calculated using equation 2 [12]:
Line density -0.03 0.02 0.01
Population density 0.09 -0.02 -0.14
( , )
( | ) = ( ) (2)

Table 4 presents the correlation among the
independent variables. Evident from the table, the where j and i are the indices representing
correlation among the independent variables are dependent variables (i.e., non-motorized, public, and
very slight, other than the correlation between the K- private) and independent variables (i.e., household
12 school children and family size and the income, a number of passengers, and distance from
correlation between the K-12 school children and the home to school), respectively. ( | ) is the

number of passengers. probability of the dependent variable conditional

Table 4 Correlation among the explanatory on the independent variable , (, ) is the joint

variables probability, and ( ) is the probability of the

independent variable .

The probability of commute mode choice
Sex
Age
Independ Family size Working adult Train station Line density conditional on monthly household income is
CBD
ent K-12 school children Household income No. of passengers Distance to school Population density
variable illustrated in Fig. 2. Students with household income
less than 40,000 PHP/month were most likely to
Sex
1 travel by non-motorized mode, while students with
Age 0.0 1 household income more than 40,000 Php/month
7
Family size 0.0 0.0 1 were probably to commute by public transport mode
0 0
Working 0.0 0.1 0.2 (roughly 62.06%). The private mode choice
1
adult 0 6 8 accounts for the small percentage share for the entire
K-12 school - -
0.0 0.6
children 0.0 0.0 1 household income. On the other hand, an increase in
2 9
4 1
Household - 0.1 - 0.0 - household income by 20,000 Php/month on average
income 0.0 2 0.0 9 0.1 1
1 2 3 could reduce non-motorized mode choice by 8.20%
No. of 0.0 - 0.2 0.0 0.4 0.1 and increase the public transport mode and private
passengers 0.1 1
2 6 1 3 2
4 mode by 4.52% and 3.68%, respectively. The
Distance to - 0.1 - - - 0.2 0.0
school 0.0 3 0.0 0.0 0.0 9 6 1 household income in Metro Manila increased by
4 2 1 6
CBD - - - 2.15% per annum (adapted from [10]), while the
0.0 0.0 0.0 0.0 0.1
0.0 0.0 0.0 1
2 3 4 0 0 percentage share of private mode choice has a
2 3 1
Train station - - - -
0.0 0.0 0.0 0.0 0.0 positive correlation with household income, as
0.1 0.1 0.0 0.1 1
0 5 6 3 4
2 2 9 8 evident from Fig. 2. This implies that the private
Line density - - - - -
0.0 0.0 0.1 0.0 0.0
0.0 0.0 0.0 0.0 0.2 1 mode choice is expected to increase if there is no
4 1 6 2 2
5 5 4 3 7
Population 0.0 - 0.1 - 0.0 - - - - - - governmental intervention.
density 0.0 0.0 0.2 0.2 0.1 0.0 0.0 0.1 1
2 4 8
3 1 1 8 0 8 9 4

The correlation coefficients can indicate how the
independent variables are associated with dependent

23