Presented to the School of Economics

De La Salle University - Manila

Term 3, A.Y. 2022-2023

In partial fulfillment
of the course
In THS1ECO (V25)

“Ride or Drive: A Probability Analysis on the Impact of Socio-economic Factors on Vehicle

Type Ownership in the Philippines using Multinomial Logistic Regression”

Submitted by:
Chua, Jason Bryan V.
Lacerna, Bea M.
Manuel, Joaquin Enrico V.

Submitted to:
Dr. Tereso S. Tullao, Jr.
Dr. Myrna S. Austria
Dr. Winfred M. Villamil
Ms. Ma. Ella Calaor-Oplas
[Date of Submission]
Note to all: Please refer to Guide.pdf before writing your parts.
I. Introduction
1.1 Background of the Study BeaCompleted

As globalization continues to heave, vehicles are becoming a part of people’s necessities.

People would undoubtedly need a vehicle to transport them to work, school, leisure, and other

economic activities. According to the findings of a survey carried out by NEDA, Filipinos have a

strong desire to be mobile (2016). The survey found that most of the respondents, composing 77

percent, gravitate towards having a vehicle of their own, while a mere 23 percent decide to

depend on reliable transportation options. With this, significant increases in the sales of vehicles

in the country are seen in recent years, for both four-wheeled and two-wheeled vehicles.

However, are the country’s roads capable enough of housing the continuously increasing number

of vehicles? The Philippines’ capital city ranked 9th out of 389 worst traffic-congested cities in

2022 (BusinessWorld, 2023). Furthermore, a study conducted by JICA in 2018 states that this

traffic congestion leads to the country losing P3.5 billion in "lost opportunities" every day, and is

predicted to reach P5.4 Billion in 2035 if the government fails to implement necessary actions to

alleviate the crisis (CNN, 2018).

Simultaneously, the sales of motorcycle vehicles in the country have been surging,

recording the highest growth rate of 60.1% in January 2023, making it the highest in Southeast

Asia (Talavera, 2023). In comparison, a report presented by the Chamber of Automotive

Manufacturers of the Philippines (CAMPI) and the Truck Manufacturers Association (TMA)

stated that there had been a 44.8% year-on-year increase in car sales for the first quarter of 2023,

led by Toyota, Mitsubishi, and Ford, respectively (Monzon, 2023).

A survey conducted in 2022 revealed that about fifty percent of the entire Filipino
households owned a motorcycle or tricycle (Statista, 2023). It is further stated that approximately

11% of respondents owned a four-wheeled vehicle, whereas a small number of 0.2% had an

electronic jeepney. With the exponentially growing population, the numbers are still expected to

increase, considering the transportation crisis that is currently being experienced in the country.

The rising vehicle trends in the country could aggravate numerous destructive

occurrences such as increased carbon emissions, traffic congestion, and road accidents. Thus, it

is essential to conduct a study on the likelihood of how socio-economic factors, such as income,

family size, gender, age, employment status, urbanization, and location, affect the purchasing

behavior of Filipino consumers on two-wheeled vehicles, four-wheeled vehicles, or both.

1.2 Statement of the Problem BeaCompleted

One of the most crucial decisions a household makes is the purchase of a vehicle, as

owning a vehicle has a significant impact on mobility and access to a variety of opportunities

like employment and social services (Mohammadian, 2005). However, as the sales of vehicles –

two-wheeled and four-wheeled, continue to surge in the country, this could aggravate

significantly the increase in traffic congestion, road accidents, vehicle dependency, and carbon

emissions in the country. Thus, this study aims to answer the question: “What are the socio-

economic determinants and probabilities of vehicle type ownership in the Philippines?”

1.3 Research Objectives BeaIn ProgressJasonCompleted

The research study focuses on the impact of socio-economic determinants on the

probabilities of vehicle type ownership in the Philippines. The researchers are hopeful in finding

factors that have significant effects on the probabilities of vehicle type ownership in the country.
With this in mind, these are the specific research objectives for this study that can help in

answering the study’s research questions:

a. To identify the socio-economic factors that influence vehicle type ownership in the

Philippines

b. To estimate the impact of these factors on motorcycle and car ownership

c. To compare the probabilities of the factors affecting the ownership of motorcycles, four-

wheeled vehicles, or both

1.4 Significance of the Study JasonCompleted

The significance of the research is found in the potential relationship between different

socio-economic factors and their effects on the probabilities of vehicle ownership in the

Philippines. The insights gathered from the study can help economists better understand the

vehicle ownership behavior of households, as well as looking into private vehicle dependence in

the country. Government officials may find interest in the topic in an effort to understand

transportation and traffic problems, with an inclination to provide solutions through regulations

and policies. As urbanization continues to trend upwards, this can become a tool for researchers

who are focusing on urban planning projects, aided by an increased understanding of household

behavior and better management of growing numbers of vehicle owners in metropolitan areas. In

summary, the research aims to study the effect of different factors on the probability of vehicle

ownership, and how the new information provided can help in understanding consumer behavior,

aid in government policymaking, and provide data towards organizations aiming to make

progress in urbanization.

1.5 Scope and Limitations JasonCompleted

 The scope of this thesis is to identify socio-economic factors and their influence on the

probabilities of owning different vehicle types – two-wheeled and four-wheeled, in the

Philippines, more specifically a quantitative analysis of the relationship between the factors and

their effect on the probabilities of vehicle ownership. Specifically, the factors that will be

considered are income, location, family size, marital status, age, gender, and occupation.

However, this study is limited in that it will only use quantitative data obtained from the merged

dataset of the Family Income and Expenditure Survey and Labor Force Survey of 2018. In effect,

all possible perspectives for households may not be taken into consideration. Additionally, the

study will focus only on the specified factors in which data is available and cover only household

data from the Philippines. The study may also be limited by self-reported data, as survey

respondents may inadvertently report inaccurate, and thus unreliable, data for some factors.

Finally, the study may be affected by limitations of the research methods used, such as sample

size, data collection, and analysis techniques.

 II. Review of Related Literature

Figure 1. Literature Map of Socio-Economic Determinants of Vehicle Type Ownership

2.1 The Socio-economic Determinants of Choosing Different Vehicle Types

2.1.1 Family Size JasonCompleted
One study conducted by Ha et al. (2019) found that household size was a

significant factor, with larger households having vehicle ownership. Other factors also

include number of male family members, total trip-length, household workers and total

trips made which were evidenced as vehicle ownership determinants. Ritter and Vance

(2013) also found out that household size was one of the factors that limits owning

vehicles in Germany, thus concluding that household size had different influences on

vehicle ownership. The study implies that there is a positive correlation between
family size and vehicle ownership. Larger households may have more difficulty

accessing public transportation because they have more members who need to travel.

2.1.2 Age JoaquinCompleted

In the years following the financial collapse and economic downturn in 2007,

there have been studies (Fry, 2013) and press articles (Thompson, 2015) which have

recorded a drastic decrease in the willingness of young adults' to own vehicles,. For

example, According to a study by Kurz, Li, and Vine (2016), a drop in the

proportion of new automobiles bought by young adults (18 to 34 years old) indicates

that their financial constraints have increased and their level of interest to drive has

diminished. Young adults have been observed to think that driving isn’t as thrilling or

exciting as it was before. These changes are frequently attributed to the increasing

prominence of the Internet and social media. These reduce the necessity to travel

before being able to see with one’s own senses how a potential location in the world

would be like. Furthermore, new alternative means of transportation, such as ride-

sharing, public transportation, and biking, further reduce the need of owning a vehicle. In

recent years, the rise in the average age of vehicle consumers has mirrored changes

in car buying rates among consumers of various ages (Kurz, Li, and Vine, 2016).

Although a portion of the increase in the average age of vehicle buyers is

attributable to a reduction in the frequency at which young individuals get new

automobiles, the aging of the population at large and a decline in the buying rate

among 35- to 50-year-olds seem to be the most significant age-related factors

affecting vehicle procurement.

2.1.3 Sex Bea Completed

 Studies have found a significant relationship between sex and vehicle type

preference, focusing on two- and four-wheeled vehicles. This factor is further impacted

by other factors such as cultural effects, safety issues, and mobility patterns. In a study by

Yang et al. (2013), women's use of two-wheel vehicles such as bicycles in Suzhou is

significantly higher than men's, having 44.8% versus 35.7%, whereas men's four-wheel

allocation rate is significantly higher than women's at 17.6% versus 25.6%. This result

where men have a higher probability to travel by car is impacted by the mobility pattern

between varying gender, where men travel more frequently and at longer distances to go

to work while most women usually stay at home for household-related chores.

2.1.4 Marital Status JasonCompleted

Numerous studies have examined the association between marital status and

vehicle ownership. A study by Nolan (2010) determines that there is a highly significant

effect between age 35-44, being wedded and the number of adults, and whether the

household owned a car in the previous year, thus, increasing the probability of current

household car ownership by 1.3%, 6.5%, and 1.9% respectively. The strength of this

relationship, however, differs depending on various factors, including the gender of the

individual, the country or region, and the availability of transportation.

2.1.5 Income Bea Completed

Higher-income households typically possess greater financial means, facilitating

them to spend on greater-value and exclusive items and services that satisfy their needs

and preferences. Similarly, those of low income would adjust their buying preferences

based on the limited financial resources they have. One factor that may affect the varying

income households is the purchase price and operating costs. Manski and Sherman

(1980) provide that purchase price is a detriment factor in choosing a vehicle, particularly
for lower-income households. Operating expenses of the vehicle type appear to be more

critical for low-income households than for those with higher incomes. The study also

showed that one possible outlier result is that low-education, high-income rural

households appear indifferent about operational expenses. This is supported by the fact

that compared to urban-situated families, these rural households obtain higher fuel

efficiency and thus pay cheaper gasoline expenses. In the study accomplished by Bansal

et al. (2018), it is also found that household income influences all three forms of

motorized vehicle ownership which includes owning four-wheel vehicles, two-wheel

vehicles, or both. The study suggested through the MNL model that the possibility of

owning a four-wheel vehicle increases as household income also increases. High-income

households are able to afford more vehicles and may have more occupations and

priorities outside of their home.

2.1.6 Location / Urbanization Bea Completed

Location undoubtedly serves as a factor in consumers' preference for owning a

motorcycle or a car. A study by Chiu et al. suggests that motorcycle ownership is greater

in countries with low urbanization than in those that are highly urbanized (2022). From

the perspective of consumers, the extent of economic progress at an early stage of

development is limited and less extensive than at more advanced stages of development.

As a result of these conditions, motorcycles are more commonly preferred and utilized to

meet the needs of only short-range travels. In line with this, another study using the

multinomial logit model has also found that households residing at moderate distances

from the center of the city are more probable to opt for motorcycles over other vehicle

types, and that preference over motorcycle use is most susceptible to trip distance than
the other modes of transportation (Fevriera et al., 2021). As the economy and

urbanization develop, the demand for long-distance travel increases, causing the

preference for cars to increase. However, in contrast to other studies, households in

Macao residing in an urban area with a larger population are less likely to own a car than

those residing in the islands (Wong, 2013).

2.1.7 Employment JasonIn Progress

According to Baum (2009), owning a vehicle can improve employment

opportunities to 30% points, and 13 working hours a week, that lead to greater wages.

Based on these studies, employment and vehicle ownership have an optimal relationship.

Owning a car can make it easier for employees to go to work and to other possibilities,

like education and health care. It's crucial to remember that these findings do not

necessarily show that owning a vehicle contributes to employment. It's likely that other

aspects, like wealth or education, influence both employment and vehicle ownership.

2.1.8 Effect of Socio-economic Factors on Vehicle Ownership BeaCompleted

Socioeconomic factors play an important role in influencing vehicle type

preferences, as they reflect the wider economic and social environment within how

people build mobility decisions. Numerous studies have found that socioeconomic

characteristics indeed impact the ownership of vehicles, through the use of a multinomial

logit approach (Wong, 2013; Zegras & Hannan, 2012; Tsang et al., 2011; Gomez-Gelvez

& Obando, 2013; Prabnask et al., 2011; Pasra et al., 2018; Joseph et al., 2017). Some

common socioeconomic factors that were found to affect ownership from the various

studies were income, household size, and urbanization. Some studies have also used the

probit model (Ding et al., 2016), however, results from Potoglou and Kanaroglou (2008)
 explicitly demonstrated that the multinomial logit model is a better option relative to the

use of ordered probit and logit models. In another vein, aside from using disaggregated

level of socioeconomic characteristics, Dargay & Gately (1999) and Ukonze et al. (2020)

have shown at an aggregate level how GDP and the country’s income largely impacts

vehicle ownership in the country.

2.2 The Built Environment’s Impact on Vehicle Type Preference

Besides the internal factors which are attributed to the buyers of vehicles, the built

environment and surroundings contribute to vehicle type preference as well. The

aforementioned environment is comprised of a variety of intentionally constructed

and altered buildings and locations, diverse patterns of land use, public

transportation networks, and other elements of urban design which may impact the

decisions and actions of inhabitants (Handy et al., 2002). The built environment

differs from the natural environment in that it is a product of human civilization,

provides human activity with a temporal, spatial, and social context, and includes

elements related to land use, urban design, and transportation systems.

2.2.1 Traffic Congestion JoaquinCompleted

The way in which the built environment could affect vehicle type preferences

would be in road traffic congestion, which is a prevalent issue in the majority of large

cities, and it is essential to investigate the underlying causes to relieve congestion in

the roadways. Travel behavior including vehicle ownership preference are closely

related to the built environment, with a large variability in the effect of the built

environment on the moderate travel pace of the road (Pan et al., 2020). Key factors

have been studied such as bus stop density, residential community density, and business
building density which could affect roadway speed, with bus stop density having the

maximum and positive effect, while residential community density and business building

density having a negative influence (Chiou et al., 2015). The influence of healthcare

service density, sports and leisure service density, and parking entrance and exit density

are likewise influential, providing positive effects throughout. It has also been seen that

the influence of road segment attributes such as grade and length on the average travel

speed is rather small. All of these factors demonstrate the relationship between the

built environment and adjacent road traffic efficiency, thereby providing

information and direction for urban planning and transportation system

optimization (Tomer et al., 2020).

2.2.2 Road Accidents JoaquinCompleted

A major consequence of the built environment on vehicle owners and passengers

would be in terms of road traffic injuries, which are a leading reason of death and

disability nationwide. This is further affected by built-environment factors that

potentially alleviate or aggravate traffic safety problems in urban areas. Particularly

vulnerable roadway users are frequently involved in collisions resulting in serious

injuries in low- to middle-income countries, making injuries sustained in traffic

accidents a significant global public health concern. Annually, thousands of

individuals are killed and countless others are disabled in road traffic accidents

(Clifton et al., 2009). Due to accelerated motorization, the absence of roadway safety

culture, poor road conditions, and a lack of road safety training, more than 90

percent of these fatalities and injuries occur in countries with low to middle

incomes. The results of a study (Zimmerman et al., 2015) show that in addition to road
network and traffic characteristics, built-environment factors play a role in the number of

crashes as well as the chance of a crash occurring . It is noticeable that impacts of the

built-environment factors are more dominant in KSI (killed and severe injury) crashes

compared to PDO (property damage only) crashes. Moreover, it was observed that newly

developed urban areas built after 1990 that have different spatial designs are safer for

road users (Schepers, 2018). The increase in land-use diversity is seen to be significantly

associated with its positive impact on improving traffic safety, and a more elevated land-

use density was seen to have negative effects on traffic safety. Population density, on the

other hand, does not have significant effects although it was expected to have an impact

on frequency and the probability of traffic crashes. Therefore, the overall results

suggest that the features of the traffic and land use in neighboring areas are closely

connected with the level of road safety.

2.2.3 Access to Public Transportation JasonIn Progress

Vehicle ownership and access to public transportation are strongly correlated. In a

study conducted by Mulalic & Rouwendal (2020) in a metropolitan area, access to high-

quality public transportation can be a good substitute for owning a vehicle for some

households. The development of the transport system results in a 2-3% decline in vehicle

ownership, with an average compensating variance of about 3% of household income.

Rith et al. (2019) also emphasizes that having a car has somewhat become a necessity in

the city as a result of improper development planning. This is because using public

transportation can lessen the need for an automobile by offering a practical and cheap

means of transportation. The study also discovered a complicated relationship between

having access to public transportation and owning a car. This relationship can be

influenced by a wide range of variables, including the standard of public transportation,

 the accessibility of other modes of transportation, and the socioeconomic characteristics

of the population of the area.

2.3 Notable Differences on Users of Different Vehicle Types

2.3.1 Two-Wheeled Vehicle JoaquinCompleted

The research focuses on two-wheeled vehicles as the first form of vehicle.

Two-wheeled vehicles, more especially motorbikes, which had begun to become an

important part of a household in the Philippines are the primary subject of this

research. Because there is a lack of a reliable public transportation infrastructure

in many regions, individuals have resorted to riding motorbikes as a mode of

transportation instead. Tricycles, habal-habals, and motorelas are just a few

examples of the many varied designs that motorcycle-powered vehicles may take

(ADB, 2020).

A study conducted by Fishman and Schepers (2016) to find the major causes

why respondents owned a two-wheeler found several factors of influence namely:

convenience, followed by commuting, price, comfort, and performance. Secondary

reasons, young adults are more concerned with expressing themselves through the

purchase of a motorcycle, but middle-aged persons are more concerned with

meeting social demands, having a peaceful ride, receiving a grant from the

government, and getting recommendations from friends. Furthermore, most of the

respondents in the surveys conducted were slightly satisfied with their two-wheelers

(Huang, 2018). In particular, user-perceived hedonic and pragmatic qualities of two-

wheeled vehicles are a significant contributor to customers' levels of contentment

and recommendations. intention, while riding experience, interactions, vehicle

designs, and interpersonal interactions all appear to be important factors.

Specifically, the core factors that influenced vehicle ownership were satisfaction,

positive emotions, and riding experience. The riding experience displayed low positive

correlation with the needs of dropping off/picking up a child from school, convenience,

and work demands. In addition, external motivation plays a big part in driving the

intention to purchase, while convenience, awareness of environmental concerns, vehicle

price, and overall performance all appear to be important factors as well.

2.3.2 Four-Wheeled Vehicle JoaquinCompleted

According to Ahanchian and Biona (2014), the number of people who own

their own cars has increased significantly over the past several years, particularly in

Metro Manila. The choice of whether or not a family should possess an automobile

is influenced by a wide range of factors, each of which may be characterized in

either a qualitative or quantitative manner. The automobile has a significant impact

on the financial condition of industrialized countries as well as emerging ones, and

this has a domino effect on a wide variety of activities. According to Ang and Fillone

(2018), the processes of building and manufacturing automobiles, as well as

subcontracting, dealerships, and recycling, all have a significant influence on the

economy of a country. Furthermore, the vehicle provides planners and engineers with

crucial data for road network construction, maintenance, and management. Capital

movements associated with the export and import of vehicles and parts also have a

significant impact on the balance of trade, with petroleum imports and taxes such as

fuel taxes constituting an important driver of government income.

Despite the rising costs of vehicles, household income continues to play a

significant role in assessing the probability that a household owns at least one
vehicle, with the probability increasing as household income rises (Grise et al., 2019).

It is also intriguing to observe the inverse relationship between household income

and the total number employed adults, which indicates that an increase in the

number of employees in a household is not associated with a rise in household

income. The job of the household head is also a significant determinant, with

government officials, managers, and supervisors being more likely to own a vehicle.

It was also observed that younger acting as household heads are more likely to own

a vehicle. This is due to the reason that younger people have a greater need for

mobility than elderly people, which a vehicle can fulfill (Jiang et al., 2017). Another

factor that could be of note was the relatively better accessibility provided by mass transit

structures such as trains to workplaces, with improved public transportation within

the main commercial district and proximity to job opportunities. Additionally, more

advanced employment types positively affect car ownership, which implies that jobs

with higher salaries increase the likelihood of car ownership overall (Liu & Cirillo,

2016).
 III. Theoretical and Conceptual Framework

3.1 Theoretical Framework

This study focuses on looking at the probabilities of different vehicle types being chosen

based on the socioeconomic characteristics of the consumers such as household income, sex, age,

and employment, through the multinomial probit model. The theoretical underpinning of this

study is the Random Utility Theory, which states that the likelihood of an individual preferring

or selecting a given alternative is a function of the individual's socioeconomic characteristics and

the choices' relative enticing qualities (Zhong et al., 2022); The concept of utility, which is

derived from a set of attributes of the alternatives as perceived by the individual, is used to depict

the attractive qualities of alternatives. Furthermore, the theory states that individuals would

generally always opt for the option where they can get the highest utility.

Another important theory that supports the study is the rational choice theory originally

proposed by Adam Smith in 1776, which suggests that an individual’s own self-interests and

agendas are what make up their decision-making choices. In addition, this theory demonstrates

how individuals create specific choices based on relevant costs and benefits they will get from it.

Relating it to the study, when an individual resides in a highly urbanized area where traffic

congestions are rampant, he would be more likely to choose a motorcycle over a four-wheeled

car if his goal is to be able to move as fast on the congested roads. However, if the individual’s

goal is to prioritize his and his passengers' comfort and safety over the time being wasted on the

road, then he is more likely to get a car.

Furthermore, one of the most significant socioeconomic factors that are said to affect

vehicle choice, as stated in the literature and studies above, is the individual’s income. In

microeconomics, consumer choice theory would say that an individual would choose the choice
that would give them the highest utility which is subjected to their budget constraints. Hence,

this theory suggests that higher-income individuals would choose to own a car more, and those

with lower income would be more likely to purchase a motorcycle vehicle, given that two-

wheeled vehicles are way more inexpensive than their four-wheel counterparts. Aside from the

unit price, two-wheeled vehicles are also cheaper to maintain with a lesser amount needed to

allocate for fuel expenses and other related transaction costs.

3.2 Conceptual Framework Joaquin Enrico Manuel

 IV. Data and Methods

This chapter will discuss the researchers’ process of relevant data collection and analysis

using the appropriate econometric methods that will help answer the research questions of the

study.

4.1. Data Collection

4.1.1. Data and Variable Specification

The variables that will be used for the methodology and data analysis will be

specific variables found in the 2018 FIES LFS merged dataset that are based on the

conceptual framework and research objectives. Firstly, the researchers will need data on

different socio-economic factors of households in the Philippines. Table 1 lists the

different socio-economic variables that the researchers will be collecting data on.

Variables Description

fsize Family Size

hs001002_age Age

hs001001_sex Sex

hs001003_ms Marital Status

toinc Total Income

urb Urban/Rural

w_regn Region

hs001005_job Household Head has a job

Table 1. Description of variables identified as socio-economic factors for households

Secondly, the researchers will identify the variables that will show the choice of

the household to own a two-wheeled vehicle, four-wheeled vehicle, and owning both

types of vehicles. Data for the number of two and four-vehicles that the household owns
is present in the 2018 FIES LFS merged dataset. The variable “h150120_car_qty”

showing the number of cars, jeeps, and vans owned and will be used to represent four-

wheeled vehicle ownership for his study. Likewise, the variable “h150126_motor_qty”

shows the number of motorcycles and tricycles owned by the household and will

represent two-wheeled vehicle ownership for this study. Additionally, the researchers

created new variables that represent the choices of the households in owning each type of

vehicle. Table 2 shows the existing and new variables that will be used in the study that

shows the choices made by the household, as well as their measurement.

Variable Variable Label Measurement

h150120_car_qty Car, Jeep, Van No. of units owned

h150126_motor_qty Motorcycle/Tricycle No. of units owned

have_car Have Car 1 = owns Car, 0 otherwise

have_mtr Have Motorcycle 1 = owns Motor, 0

otherwise

have_car_mtr Have Car and Motorcycle 1 = owns Car and

Motorcycle, 0 otherwise

choice Household Choice 1 = owns Car, 2 = owns

Motorcycle, 3 = owns Car
and Motorcycle

Table 2. Variables Representing the Choice of Households

Of the household respondents surveyed in the Philippines, 5,600 households, or

3.79% of households, own only cars, 45,459 households, or 30.77% of households, own

only motorcycles, and 5,390 households, or 3.65% of households, own both cars and

motorcycles.

4.1.2. Operational Framework

 Figure 2. Operational Framework

Figure 2 shows the operational framework of the research study, which shows the

process of developing the multinomial probit model of the study. To determine the

factors that influence the ownership of household of different vehicle types, the

researchers identified common socio-economic factors between each vehicle type owner

of the household, namely:

1. Age

2. Sex

3. Income

4. Employment

5. Marital Status

6. Urbanization

7. Family Size

4.2. Data Analysis

In the study, the researchers will develop a probit model based on socio-economic factors

that may affect vehicle type choice, which were discussed in the operational framework (Figure
3). Using Stata allows the researchers to estimate propensity scores for each outcome (Have Car,

Have Motorcycle, and Have Car and Motorcycle). After the estimation, the propensity scores of

similar households will be matched with one another before comparing outcomes, accounting for

the average treatment effect between the outcomes. The researchers will continue to use other

statistical tools to analyze and estimate the average treatment effect.

4.3. Model Specification

4.3.1. Components of Probit Model

As the researchers will be using a probit model, the components of the model

have been identified, which includes:

1. Dependent Variable - refers to the variable in the model that is being predicted.

2. Independent Variable - refers to variables that will be used in predicting the value

of the dependent variable of the model.

3. Probit Function - refers to the standard normal cumulative distribution function

that is used for estimating the probability of the dependent variable.

4. Coefficients - refers to the parameters that reflect the magnitude and direction of

the relationship between the dependent variable and the selected independent

variable.

5. Error Term - refers to the term in the model that represents the variability in the

dependent variable that is not captured by the chosen independent variables.

In estimating the probability of a household being assigned to a certain treatment

group, the propensity score equation is shown below:

PSi=P (T i=1∨x i);

 where PSi is the propensity score for the household i, T iis the treatment status for

respondent i, and x iis the vector of the independent variables.

4.3.2. Propensity Score Matching

For propensity score matching, households with similar propensity scores will be

matched with one another. To identify the propensity score of the household, the general

propensity score equation will be used:

PSi=Φ (β 0 + β i x i +... βn x n); for i=1 ,2 , ... ,n ;

where where PSi is the propensity score for the household i , Φ is the cumulative

distribution function of the standard normal distribution, β i is the vector of coefficients

estimated from the probit model for household i , and x i is the vector of covariates for

household i . Using this general propensity score equation and the socio-economic factors

identified in the Figure 3, the study’s multinomial probit model is as follows:

PSi=Φ ( β 0 + β i x i +... βn x n); for i=1 ,2 , ... ,n ;

4.4. A priori Expectations for Identified Independent Variables

Table 3 shows the key independent variables identified by the researchers in the

operational framework, as well as the a priori expectations for how each variable will affect the

choice of the household in owning two-wheeled vehicles, four-wheeled vehicles, and both types.

Variable Variable Label Relationship

fsize Family Size Negative (-)

age Age Positive (+)

sex Sex Positive (+) and Negative (-)

ms Marital Status Positive (+)

inc_dec Income (in decile) Positive (+)

 urb Living in Urban Area Positive (+)

region Region Positive (+)

job Household Head has a job Positive (+)

Table 3. A priori Expectations for Independent Variables of Households Affecting Vehicle

Ownership

1. Family size may decrease the likelihood of owning a vehicle, as more members may indicate

higher expenses, which could lead to cost-cutting.

2. Higher values for the age of the household head may lead to greater chances of vehicle type

ownership, with the probability increasing as age increases.

3. The sex of the household head may have varying effects depending on the vehicle type.

4. The household head may have a higher probability of vehicle ownership if they are or have

been married.

5. As the household head increases their income to a different income bracket, so does their

chances of vehicle ownership.

6. Living in an urban area rather than a rural area may increase the chances of vehicle ownership.

7. Living in regions where business activities are more centrally located increases the probability

of vehicle ownership.

8. The employment status of the household head has a great chance of increasing the probability

of vehicle ownership.
References:

ADB. (2020). Growth of motorcycle use in metro manila: Impact on road safety. Asian
Development Bank.

Ahanchian, M., Biona, J.B. (2014). Energy demand, emissions forecasts and mitigation
strategies modeled over a medium-range horizon: the case of the land transportation
sector in metro Manila Energy Policy, 66, pp. 615-629.

Ang, D., Fillone, A. (2018). Road capacity imbalance at exit points of the Metro Manila
skyway and adjoining roads. Proceedings of the 25th Annual Conference of the
Transportation Science Society of the Philippines (2018)

Bansal, P., Kockelman, K. M., Schievelbein, W., & Schauer-West, S. (2018). Indian vehicle
ownership and travel behavior: A case study of bengaluru, Delhi and Kolkata. Research
in Transportation Economics, 71, 2–8. https://doi.org/10.1016/j.retrec.2018.07.025

Chiou, Y.-C., Jou, R.-C., & Yang, C.-H. (2015). Factors affecting public transportation usage
rate: Geographically weighted regression. Transportation Research Part A: Policy and
Practice, 78, 161–177.

Chu, M. Y., Law, T. H., Hamid, H., Law, S. H., & Lee, J. C. (2022). Examining the effects of
urbanization and purchasing power on the relationship between motorcycle ownership
and Economic Development: A panel data. International Journal of Transportation
Science and Technology, 11(1), 72–82. https://doi.org/10.1016/j.ijtst.2020.12.004

Clifton, K. J., Burnier, C. V., & Akar, G. (2009). Severity of injury resulting from pedestrian–
vehicle crashes: What can we learn from examining the built environment?
Transportation Research Part D: Transport and Environment, 14(6), 425–436.

Fevriera, S., de Groot, H. L. F., & Mulder, P. (2021). Does urban form affect motorcycle use?
evidence from Yogyakarta, Indonesia. Bulletin of Indonesian Economic Studies, 57(2),
203–232. https://doi.org/10.1080/00074918.2020.1747595

Fishman, E., & Schepers, P. (2016). Global bike share: What the data tells us about road safety.
Journal of Safety Research, 56, 41–45.

Fry, R. (2013). Young adults after the recession: Fewer homes, fewer cars, less debt. Pew
Research Center’s Social & Demographic Trends Project.
http://www.pewsocialtrends.org/2013/02/21/young-adults-after-the-recession-fewer-
homes-fewer-cars-less-debt/

Gomez-Gelvez, J. A., & Obando, C. (2013). Modeling car ownership in urban areas of
developing countries: Case study of Bogoota, Colombia. Transportation Research
Record, 2394(1), 111–118
 Grisé, E., Boisjoly, G., Maguire, M., El-Geneidy, A. (2019). Elevating access: comparing
accessibility to jobs by public transport for individuals with and without a physical
disability. Transp. Res. A, 125, pp. 280-293.

Ha, T. V., Asada, T., & Arimura, M. (2019). Determination of the influence factors on
household vehicle ownership patterns in Phnom Penh using statistical and machine
learning methods. Journal of Transport Geography, 78, 70–86.
https://doi.org/10.1016/j.jtrangeo.2019.05.015

Huang, F.-H. (2019). Understanding Users’ Experiences of Riding a Two-Wheeler

Vehicle and their Intentions of Purchasing Electric Two-Wheelers. PROMET - Traffic &
Transportation, 31(5), 503–512.

Jiang, Y., Gu, P., Chen, Y., He, D., Mao, Q. (2017). Influence of land use and street
characteristics on car ownership and use: evidence from Jinan, China. Transp. Res. D, 52,
pp. 518-534.

Joseph, O. O., Eromietse, E. J., Emmanuel, D., & Olufunke, A. (2017). Multinomial logit model
estimation of household characteristics influencing car ownership in Akure, South West
Nigeria. International Journal for Traffic and Transport Engineering, 7(2), 203–215.

Kurz, C., Li, G., & Vine, D. (2016). The young and the carless? The demographics of new
vehicle purchases. FEDS Notes, 2016.

Liu, Y., Cirillo, C. (2016). Small area estimation of vehicle ownership and use. Transp.
Res. D, 47, pp. 136-148.

Manski, C. F., & Sherman, L. (1980). An empirical analysis of household choice among motor
vehicles. Transportation Research Part A: General, 14(5–6), 349–366.
https://doi.org/10.1016/0191-2607(80)90054-0

Mulalic, I., & Rouwendal, J. (2020). Does improving public transport decrease car ownership?
Evidence from a residential sorting model for the Copenhagen metropolitan area.
Regional Science and Urban Economics, 83, 103543.
https://doi.org/10.1016/j.regsciurbeco.2020.103543

Nolan, A. (2010). A dynamic analysis of household car ownership. Transportation

Research Part A-policy and Practice, 44(6), 446–455.
https://doi.org/10.1016/j.tra.2010.03.018

Pan, Y., Chen, S., Niu, S., Ma, Y., & Tang, K. (2020). Investigating the impacts of built
environment on traffic states incorporating spatial heterogeneity. Journal of Transport
Geography, 83, 102663.

Pasra, M., Ramli, M. I., & Dharmowijoyo, D. E. (2018). Car ownership of households in sub-
urban area in Makassar city. In M. Raza & UI Mustafa (Eds.), MATEC Web of
Conferences 203,05005, ICCOEE (pp. 1–9). EDP Sciences
Potoglou, D., & Kanaroglou, P. S. (2008). Modelling car ownership in urban areas: A case
study of Hamilton, Canada. Journal of Transport Geography, 16(1), 42–54.
https://doi.org/10.1016/j.jtrangeo.2007.01.006

Prabnask, J., Taylor, M. A. P., & Yue, W. L. (2011). Modelling household vehicle ownership in
Asian medium-sized urban area: A case study of Khon Kaen City. Thailand. Journal of
the Eastern Asia Society for Transportation Studies, 9, 196–210.

Rith, M., Abad, R. P. B., Fillone, A. M., Doi, K., & Biona, J. B. M. (2019). Understanding the
impact of urban form attributes on household vehicle ownership and choice in metro
Manila: Modeling, simulation, and application. Engineering and Applied Science
Research, 46(3), 238–247. https://doi.org/10.14456/easr.2019.27

Ritter, N., & Vance, C. (2013). Do fewer people mean fewer cars? Population decline and car
ownership in Germany. Transportation Research Part A-policy and Practice, 50, 74–85.
https://doi.org/10.1016/j.tra.2013.01.035

Schepers, P., Lovegrove, G., & Helbich, M. (2018). Urban form and road safety: Public and
active transport enable high levels of road safety. In Integrating Human Health into
Urban and Transport Planning (pp. 383–408). Springer International Publishing.

Thompson, D., & Weissmann, J. (2012). The cheapest generation. The Atlantic.
http://www.theatlantic.com/magazine/archive/2012/09/the-cheapest-generation/309060/

Tomer, A., Kane, J., & Vey, J. S. (2020). Connecting people and places: Exploring new
measures of travel behavior.

Tsang, F., Daley, A., & Milthorpe, F. (2011, September 28–30). Forecasting car ownership in the
Sydney area. In W. Young & G. Dolman (Eds.), Australian Transport Research Forum
proceedings (pp. 1–14). Adelaide, Australia: EDP Sciences

Ukonze, F. I., Nwachukwu, M. U., Mba, H. C., Okeke, D. C., & Jiburum, U. (2020).
Determinants of vehicle ownership in Nigeria. SAGE Open, 10(2), 215824402092297.
https://doi.org/10.1177/2158244020922970

Wong, K. I. (2013). An analysis of car and motorcycle ownership in Macao.

International Journal of Sustainable Transportation, 7(3), 204–225.
https://doi.org/10.1080/15568318.2013.710057

Yang, M., Li, D., Wang, W., Zhao, J., & Chen, X. (2013). Modeling gender-based differences in
mode choice considering time-use pattern: Analysis of bicycle, public transit, and car use
in Suzhou, China. Advances in Mechanical Engineering, 5, 706918.
https://doi.org/10.1155/2013/706918

Zegras, P. C., & Hannan, V. H. (2012). Dynamics of automobile ownership under rapid growth.
Transportation Research Record, 2323, 80–89.
Zhong, M., Hunt, J. D., Abraham, J. E., Wang, W., Zhang, Y., & Wang, R. (2022). Advances in
integrated land use transport modeling. Advances in Transport Policy and Planning, 201–
230. https://doi.org/10.1016/bs.atpp.2021.10.002

Zimmerman, K., Jinadasa, D., Maegga, B., & Guerrero, A. (2015). Road traffic injury on rural
roads in tanzania: Measuring the effectiveness of a road safety program. Traffic Injury
Prevention, 16(5), 456–460.