Understanding Young Commuters’ Mode Choice Decision to Use Private Car or Public Transport from an Extended Theory of Planned Behavior

Abstract

The research examines young commuters’ (age 18–25 years) decision-making process in choosing a mode to commute. The study presented here characterized young commuters’ travel decisions in relation to choosing to use a private car or public transportation for commuting travel. To contribute to greater understanding, the theory of planned behavior (TPB) was used to investigate commuting mode choice, particularly from a social-psychological perspective. Disaggregated travel data was collected from an in-depth survey of young commuters (n = 3171) who lived in inner and metropolitan Melbourne, Australia. Respondents were clustered based on their usual commuting mode (private car or public transportation). The questionnaire was designed to capture information on young commuters’ attitudes toward their commuting mode, social environment, and travel behavior. Linear regression analysis and binary logistic regression were used to predict intention to choose a commuting mode and the actual mode choice, respectively. Results suggest that young commuters who used a private car held a positive attitude toward the mode. On the contrary, public transportation users had few positive beliefs about commuting by public transportation. Convenience and flexibility were the most valued advantages that the private car offers as a commuting mode. For public transportation users, reliability of the commuting mode was more important than other factors, but they did not agree that public transportation provides reliability. The paper includes a discussion on how the implications of these results can influence government policies and strategies in an effort to increase commuting by public transportation.

Every day on Australian roads around four people die and 90 people are seriously injured. Among all road users, young adults (age 18–25 years) are the most vulnerable group. Around 25% of the drivers who lost their lives in road collisions in the Australian state of Victoria in the past 10 years were aged 18–25 years. In 2016, 19% of drivers who lost their lives were aged 18–25 years, whereas only 10% of license holders in Victoria belong to this age group ( 1 ). A broader systems-thinking-based approach is required to achieve the “Towards Zero” vision among young adults ( 1 ). Young adults are part of an important demographic, as they generally have greater access to a variety of transport modes than other age groups. Additionally, young adults tend to have a different lifestyle from other age groups, and they are in their formative years in relation to travel habits ( 2 ).

In relation to mobility management, recent studies suggest that substantial safety benefits can be achieved by shifting car travel to public transportation. Research based on U.S. data indicates that transit travel is associated with lower crash risk ( 3 ). Thus, safe alternative transport options, including use of public transportation, need to be encouraged among young adults. There has been little research consideration given to exploring young adults’ decision-making process when choosing a commuter travel mode, particularly from a social-psychological perspective.

The research presented here advances the understanding of young commuters’ travel behavior. Commuting may have been to work or school. It is expected that young commuters’ attitudes and their surrounding environment influences their mode choices. The theory of planned behavior (TPB) offers the theoretical framework for deeper understanding of commuters’ travel behavior ( 4 ). Previous research has acknowledged the effect of attitudes toward mode choice ( 5 – 8 ). However, previous research has not explicitly taken into account the role of social and personal norms in young commuters’ mode choices. The research presented here investigates the travel behavior and related attitudes and beliefs of young commuters who live in and around Melbourne, Australia. The social and psychological aspects of behavior, including attitudes, social norms, habit, and perceived behavioral control (PBC), were examined.

The study divided young commuters based on their usual commuting mode, for example, private car or public transportation (e.g., bus, tram, or train). The aim of this grouping was to assess the motivators and deterrents in the context of choosing private car or public transportation for commuting. Being able to understand the factors that influence the travel decisions of young adults will help significantly with designing strategies to encourage a shift from car travel to public transportation and, thus, reduced crash risk among young adults.

Literature Insights

The passenger car has great mobility benefits, but it has undesirable implications, including road collisions. Recent research indicates that young people are less likely to get a car license than in previous years and, even if driving, they drive less. Victoria and New South Wales, among other states of Australia, have the fastest changes, with the percentage of young adults having a driving license falling by around 1% per year ( 9 ).

There is the growing body of research that aims to enhance understanding of the reduction of car use among young adults. Zhou explored university students’ travel behavior using samples from Los Angeles, U.S., through an online survey ( 10 ). The study found that those who lived alone were more likely to commute by car, while those who lived with or near other students were more inclined to commute together via public transportation. Simons et al. found that college-educated young adults were more likely to use either public transportation or cycle than non-college-educated participants ( 11 ).

Delbosc and Currie conducted a survey of young adults in Melbourne, Australia, which indicated that young adults found a car to provide independence ( 8 ). However, they also believed that it was a big responsibility. Nondrivers believed that they could go where they wanted without driving. Young adults without a driver license or a learner’s permit indicated that they were more likely to use public transportation as it gave them the flexibility to use a smartphone or a tablet ( 8 ). Kuhnimhof et al. reported a significant reduction of automobile mileage and an increase in the use of public transportation among young Germans and Britons ( 12 ). The growing multimodal behavior among young travelers was anticipated to contribute to this behavioral change. Currie et al. examined young people’s perceptions of personal safety on public transportation ( 13 ). Psychological factors such as “feeling comfortable with traveling with unknown people” were associated with young people’s perceptions of safety.

Nordfjaern et al. conducted research into active and public transportation as alternatives to motorized transport through a survey amongst a random sample of individuals living in Norway ( 14 ). The issue of practical barriers such as picking up and dropping off children at day-care, kindergarten, or schools was raised. Several respondents considered it impossible or nearly impossible to take care of such duties without the use of a private car. Those who prioritized the flexibility of their mode of transport were less likely to consider active transport as an alternative to traveling by car ( 14 ).

In light of the preceding review, a structural change in young people’s travel behavior has been observed. Previous studies indicate that attitudinal factors influence young commuters’ mode choices. To better understand young commuters’ decision-making process when choosing a mode, a more detailed examination of their travel choices, particularly from a social-psychological perspective, is of paramount importance. In addition, few studies have investigated young commuters’ travel behavior, particularly in the context of Australia.

Theoretical and Conceptual Framework

TPB provided the theoretical foundation for this project. TPB has been employed successfully to predict and explain a wide range of behavior, including voting choices, committing traffic violations and use of different modes of transportation, from a psychological perspective ( 4 , 15 , 16 ). TPB hypothesizes that behavior is the outcome of the combined effect of “intention” and “PBC.” Intention, in turn, is predicted by “attitude,”“subjective norm,” and PBC ( 4 ). TPB further assumes that these components are determined by beliefs for each component and evaluation of those beliefs. Attitude toward the behavior is defined as the degree to which a person evaluates a behavior as favorable or unfavorable. Subjective norm refers to a person’s perception of social expectation to perform a certain behavior. The underlying normative beliefs are about the perceived approval or disapproval of a certain behavior by significant others with whom the person is motivated to comply. The third predictor of intention is PBC, which represents a person’s perception of their ability to perform a behavior. Control beliefs concern the factors that will either facilitate or inhibit the performance of the behavior.

An extended form of TPB, as shown in Figure 1, is used here to predict both the intention and behavior of young commuters’ mode choice. The basic elements of the TPB model, namely “attitude” and PBC, are included into the model considered here. Two forms of subjective or social norms have been adopted, as suggested by Cialdini et al.: injunctive norm and descriptive norm ( 17 ). The injunctive norm refers to the expectation from a social group or society to perform a specific behavior, whereas the descriptive norm represents the visible behavior of others. It is expected that young commuters’ mode choice behavior would be affected by the preferences of various social groups. Again, neighboring people’s travel behavior may also influence their decisions. Schwartz’s norm activation model hypothesizes that individuals adopt the expectations of society and their surroundings first before they actually perform a behavior ( 18 , 19 ). Thus, individuals’ personal beliefs about others’ expectations may influence commuting behavior. Previous work has found “habit” to significantly contribute to mode choice behavior ( 5 , 7 ). Thus, “habit” is also included in the model to reflect young people’s typical response to commuting mode choice.

Figure 1.

Application of an extended theory of planned behavior (TPB) to predict the travel mode choice of young commuters.

The current research attempts to predict not only young commuters’ actual mode choice (did the person use the particular mode or not, for example, private car/public transportation?), but also their intention to use that particular mode before they decided on their behavior.

Overall, the study presented here attempts to advance understanding of young commuters’ mode choice behavior through examining the social and psychological aspects of behavior, including attitude, social and personal norms, habit and PBC.

Methodology

Procedures and Participants

A survey was considered to be the most suitable approach to achieve the research objectives, as this approach attempts to determine relationships among specific variables ( 20 ). An online survey was conducted to explore young commuters’ travel behavior from a social-psychological perspective. Victorians who held a car driver license, were aged between 18 and 25 years old, and lived in inner and middle suburbs of Melbourne were identified as potential participants for this study. This survey was sponsored by VicRoads ( 21 ). Under the Road Safety Act 1986, Section 90k (iii) (which allows disclosure of information for the purpose of research and publication of results in an unidentified form), a list of names and email addresses was obtained from the driver licensing database of VicRoads (the Victorian state driver licensing authority). Emails inviting qualifying recipients to participate in the survey were sent out. An invitation to participate in the survey was also publicized by the Royal Automobile Club of Victoria (RACV) and the Bicycle Network in their respective fortnightly e-newsletters ( 22 , 23 ). In recognition of participants’ contribution to the research, a prize draw was offered to win one of ten AUD$100 shopping vouchers.

The part of the survey results presented here concentrates on young commuters who travel by private car (e.g., ute [vehicle with a tonneau behind the passenger compartment], van, 4WD, truck, etc.) or public transportation. Detailed questions specific to each of the modes were asked following some general travel pattern questions. Questions were developed based on various constructs of TPB. These included: attitude, personal norms, descriptive norms, injunctive norms, habit, PBC, intention and behavior. For each of these constructs, there were several questions to be answered. Respondents’ socio-demographic data—age, gender, education level, income, level, and so forth—were also recorded.

Given that this research was part of the Summer Intern Program run by the Australian Road Research Board, the survey was conducted in January, which is during the holiday season for most university and high school students in Australia ( 24 ). As the research was designed to capture the travel behaviors of a variety of commuters, respondents were asked to identify the location to which they traveled to most often in a typical week. A typical week was defined as a week that is representative of the majority of weeks in the year. This eliminated the possibility of respondents whose primary role was that of a student from being excluded. However, there is the possibility that a student respondent may not remember his/her last day of commuting in great detail.

A total of 3,171 responses were collected from those who usually commute by private car or public transportation. It should be noted that no reminder email or phone follow-up was administered because of time and budget restrictions.

Designs and Measures

Self-reported measures were considered to assess young commuters’ travel behavior in the context of TPB. The selection of the items for each theoretical construct was based on an extensive evaluation of social-psychological literature to ensure that all items were appropriate and supported by previous evidence in the literature. Scores of the items were also specified based on suggestions by Fishbein and Ajzen ( 25 ).

Attitudes

Items capturing attitudes were adapted from Heinen et al. ( 26 ). Attitude is defined as the sum of belief strengths of all attributes multiplied by the evaluative aspect of the belief, that is, the evaluation of each attribute ( 27 ). The key attributes considered are relaxing, offering environmental benefits, time-saving, flexible, money-saving, pleasant, offering personal space, healthy, convenient, and traffic safe. To evaluate each attribute, statements such as “Commuting by private car gives me flexibility” were used. Participants evaluated each attribute on a five-point Likert scale ranging from “Strongly disagree” (–2) to “Strongly agree” (+2). To measure the strength of the belief for each attribute, statements such as “It is important that my commute transport mode gives me flexibility” were used. Belief strength was rated on a five-point Likert scale ranging from “Very unimportant” (0) to “Very important” (+4).

Personal Norm

Statements such as “I feel bad when I commute by private car because of the negative impact on the environment” were assessed. The five-point Likert scale ranging from “Strongly disagree” (–2) to “Strongly agree” (+2) was used.

Injunctive Norm, Descriptive Norm, and Habit

To measure injunctive norms, three attributes were considered and responses for belief strength and the evaluation of each attribute were collected. The injunctive norm was calculated in the same way as attitude. Items from Terry et al. were used to measure injunctive norms ( 28 ). Each attribute was examined using statements such as “The Government and transport authorities would like to see more people commute by private car.” Participants evaluated the attributes on a five-point Likert scale ranging from “Strongly disagree” (–2) to “Strongly agree” (+2). To measure the belief strength, statements like “I want to do what the Government and transport authorities want me to do” were used. The belief strength was rated on a five-point Likert scale ranging from “Strongly disagree” (0) to “Strongly agree” (+4).

Descriptive norm was calculated in the same way as the Attitude and Injunctive norm. Three items from Nigbur et al. were adapted to measure descriptive norms using statements such as “Please provide a rough estimate of the proportion of neighbors around your place of residence who commute by public transportation” ( 27 ). Participants were provided with six options: “none” (+1), “a few” (+2), “some” (+3), “around half” (+4), “many” (+5), and “most” (+6). Statements used to measure the belief strength of injunctive norms were also used here to measure the belief strength. They were also rated on a five-point Likert scale ranging from “Strongly disagree” (0) to “Strongly agree” (+4).

In relation to habit, items such as “I do not commute by public transportation when the weather is bad (hot, cold, raining, etc.)” were used. Again, a five-point Likert scale ranging from “Strongly disagree” (–2) to “Strongly agree” (+2) was used.

Perceived Behavioral Control (PBC)

Young commuters’ perceived control over their commuting mode was examined. PBC is defined as the sum of belief strengths of all attributes multiplied by the evaluative aspect of the belief, that is, the evaluation of each attribute ( 27 ). PBC was measured using statements examining belief strength such as “Needing to carry heavy or bulky items would discourage me from commuting by public transportation.” These statements were scored on a five-point Likert scale ranging from “Strongly disagree” to “Strongly agree.” To evaluate each attribute, statements like “I did not have to carry any heavy or bulky items” were used. The evaluation was scored on a five-point Likert scale ranging from “Strongly disagree” (0) to “Strongly agree” (+4). For private car users, the statements of belief strength were positive statements (e.g., “Needing to carry heavy or bulky items would encourage me to commute by private car”), thus scored on a positive scale “Strongly disagree” (0) to “Strongly agree” (+4), whereas it was scored on a negative scale “Strongly disagree” (0) to “Strongly agree” (–4), for public transportation users (statements such as “Needing to carry heavy or bulky items would discourage me from commuting by public transportation”).

Intention

To measure intention, participants were asked “Prior to the most recent day you commuted to your most frequent destination, how likely were you to use a private car (public transportation for public transportation users) to commute on that day?” A five-point Likert scale was used ranging from “Extremely unlikely” (+1) to “Extremely likely” (+5).

Behavior

A binary measure of commuters’ actual mode choice on the most recent day of commuting was recorded.

Results

Descriptive Statistics

Table 1 presents respondent characteristics in relation to their gender, age, role in the workplace, and personal income level. The survey participants were asked to report their commuting mode on the most recent day they commuted to their most frequent destination. This has also been presented in Table 1. Females were more highly represented (48.8%) than males (29.1%). The study sample was not intended to represent the general population of Victoria, which was male and female in almost equal proportion (i.e., 50.9% female versus 49.1% male [ 29 ]).The largest proportion of the respondents identified as being professional (19.3%), followed by being a student (19.3%).

Table 1.

Respondent Characteristics

Respondent characteristics	Proportion (%)^a
Gender
Female	48.8
Male	29.1
Other/no response	22.1
Age
18–21	28.4
22–25	71.6
Main role or occupation
Manager	4.3
Professional	19.30
Technician or trades worker	4.7
Clerical or administrative worker	9.3
Sales worker	8.6
Student	19.29
Machinery operator or driver	0.7
Laborer	2.3
Volunteer	0.3
Unemployed/seeking work	0.6
Other	9.8
No response	20.8
Personal income level
$1–$119 weekly ($1–$6,188 annually)	2.6
$120–$199 weekly ($6,240–$1,0348 annually)	4.1
$200–$399 weekly ($10,400–$20,748 annually)	10.3
$400–$599 weekly ($20,800–$31,148 annually)	12.9
$600–$799 weekly ($31,200–$41,548 annually)	14.8
$800–$999 weekly ($41,600–$51,948 annually)	14.2
$1,000–$1,399 weekly ($52,000–$72,748 annually)	12.9
$1,400 or more weekly ($72,800 or more annually)	4.3
Don’t know/no response	23.8
Own private vehicle (private car/motorcycle)
Yes	77.0
No	4.1
No response	18.9
Reported mode choice on the most recent commuting day
Private car	55.7
Public transportation	22.2
Other mode/no response	22.0

Because of rounding, percentages will not always sum to exactly 100.0%.

The statistics presented in Table 1 are based on the responses from the participants (n = 3,171) who completed the questionnaire.

Figures 2 and 3 plot the comparisons between responses of private car users and public transportation users with respect to attitudinal beliefs and the strength or importance of each belief, respectively. For better representation, the responses with respect to personal norm, descriptive norm, normative beliefs (injunctive norms), habit, and control beliefs of PBC are presented in tabular format (Tables 2 –6). Similar items were used while exploring attitudinal aspects of different users, whereas separate items specific to each mode were employed in the case of other components. As mentioned in section Methodology, each component was measured and scored differently, according to its purpose. To assist the reader, the scoring is again presented here as a footnote under each table and figure.

Figure 2.

Attitudinal beliefs. Statement: “Commuting by private car is reliable.” Items “Environmentally friendly,”“Healthy,” and “Risk of infection” were not asked to private car users. Items “Convenient,”“Pleasant,” and “Relaxing” were not asked to public transportation users.

Figure 3.

Strength of attitudinal beliefs. Question: “How important are each of the mentioned factors to you about the type of transport you use in your day-to-day commuting decisions in a typical week?” Items “Environmentally friendly,”“Healthy,” and “Risk of infection” were not asked to private car users. Items “Convenient,”“Pleasant,” and “Relaxing” were not asked to public transportation users.

Table 2.

Personal Norms Component Evaluation

Personal norms component evaluation^a	Mean	SD
Personal norm beliefs toward commuting by private car
Feel good because private car is most efficient	1.4	0.8
Feel bad because of negative impact on environment	0.1	1.3
Feel wrong to use private car every time	−0.7	1.2
Personal norm beliefs toward commuting by public transportation
Feel good because of positive impact on environment	0.7	1.0
Feel bad when I cannot commute by public transportation	−0.4	1.2
Feel wrong not to use public transportation to commute regularly	0.1	1.3

Note: SD = standard deviation.

Statement: “I feel good when I commute by private car because it is the most efficient way for me.” Range: “Strongly disagree” (–2) to “Strongly agree” (+2).

Table 3.

Descriptive Norm Components Evaluation

	Private car		Public transportation
Descriptive norm beliefs toward commuting by the usual commuting mode^a,b	Mean	SD	Mean	SD
Perceived proportion of neighbors who commute by…	5.4	0.9	3.9	1.4
Perceived proportion of people you work or study with who commute by…	5.3	1.1	5.0	1.4
Perceived proportion of family members who commute by…	5.5	1.1	2.7	1.6

Note: SD = standard deviation.

Statement: “Please provide a rough estimate of the proportion of your family members (of driving age) who commute by private car.” Range: “None” (1) to “Most” (6).

Identical statements were used for both private car and public transportation users.

Table 4.

Injunctive Norm Components Evaluation

	Private car		Public transportation		Normative belief about commuting mode^b,c	Private car		Public transportation
Motivation to comply to social groups^a,c	Mean	SD	Mean	SD		Mean	SD	Mean	SD
Compliance to government and transport authorities	1.5	1.1	1.7	1.1	Perception about government and transport authorities’ views of commuting by…	−0.9	1.0	1.0	0.9
Compliance to neighboring people	1.2	1.1	1.5	1.0	Perception about neighbors’ views of commuting by…	−0.6	0.9	0.4	0.9
Compliance to people you work or study with	1.5	1.2	1.7	1.1	Perception about people you work or study with views of commuting by…	0.2	1.1	0.3	1.0

Note: SD = standard deviation.

Statement: “I want to do what people around where I live think I should do.” Range: “Strongly disagree” (0) to “Strongly agree” (+4).

Statement: “People around where I live would like to see more people commute by private car.” Range: “Strongly disagree” (–2) to “Strongly agree” (+2).

Identical statements were used for both private car and public transportation users.

Table 5.

Habit Evaluation

Habit evaluation^a	Mean	SD
Habitual behavior toward commuting by private car
Commute by private car when the weather is bad	1.5	0.9
Commute by private car when there is a more convenient transport mode available	0.7	1.2
Commute by private car when feeling tired	1.4	0.9
Commute by private car when it is night-time or dark	0.9	1.2
Commute by private car when public transportation is crowded	1.5	0.9
Commute by private car when public transportation is unreliable or disrupted	0.6	1.3
Commute by private car regardless of situation	−0.5	1.4
Habitual behavior toward commuting by public transportation
Do not commute by public transportation when the weather is bad	−0.9	1.2
Do not commute by public transportation when there is a more convenient mode of transport available	0.8	1.2
Do not commute by public transportation when feeling tired	−1.0	1.1
Do not commute by public transportation when it is night-time or dark	−0.3	1.5

Note: SD = standard deviation.

Statement: “I commute by private car when public transportation is crowded.” Range: “Strongly disagree” (–2) to “Strongly agree” (+2).

Table 6.

Perceived Behavioral Control (PBC) Components Evaluation

PBC components evaluation^a	Mean	SD
Control beliefs toward commuting by private car
Unfavorable weather would encourage to commute by private car	3.3	1.0
Unavailability of a cheaper mode of transport would encourage to commute by private car	3.3	1.0
Needing to carry heavy or bulky items would encourage to commute by private car	3.7	0.7
Having to take a different route would encourage to commute by private car	3.3	1.0
Shortage of time would encourage to commute by private car	3.7	0.7
Having activities before arriving at the destination would encourage to commute by private car	3.5	0.8
Having activities after leaving the destination, but before arriving back home, would encourage to commute by private car	3.6	0.8
Predictability in the availability of a car would encourage to commute by private car	3.5	0.8
Availability of carpooling would encourage to commute by private car	2.3	1.2
Availability of parking at the destination would encourage to commute by private car	3.2	1.0
Unavailability of car parking at the station would encourage to commute by private car	2.5	1.4
Control beliefs toward commuting by public transportation
Unfavorable weather would discourage from commuting by public transportation	−1.7	1.3
Availability of a cheaper mode of transport would discourage from commuting by public transportation	−3.3	1.0
Availability of a faster mode of transport would discourage from commuting by public transportation	−3.2	1.0
Needing to carry heavy or bulky items would discourage from commuting by public transportation	−3.4	0.9
Having to take a different route would discourage from commuting by public transportation	−3.4	0.9
Shortage of time would discourage from commuting by public transportation	−3.0	1.1
Having activities before arriving at the destination would discourage from commuting by public transportation	−2.3	1.2
Having activities after leaving the destination, but before arriving back home, would discourage from commuting by public transportation	−2.2	1.2
Predictability in availability of a car would discourage from commuting by public transportation	−2.1	1.3
Lack of public transportation services near residence would discourage from commuting by public transportation	−3.4	1.0
Lack of public transportation services near destination would discourage from commuting by public transportation	−3.4	0.9
Lack of frequent services during time of travel would discourage from commuting by public transportation	−3.3	0.9
Insufficient car parking at destination would discourage from commuting by public transportation	−2.0	1.6

Note: SD = standard deviation.

Statement: “Unfavorable weather conditions would encourage me to commute by private car.” Range: “Strongly disagree” (0) to “Strongly agree” (+4).

In relation to attitudinal beliefs toward private car use (Figure 2), results indicate that, overall, respondents who usually use a private car for their commuting purpose held positive beliefs about the attributes of commuting by private car. The highest score corresponds to “convenience” of using a private car (1.8), followed by the features “flexibility” (1.7), “time saving” (1.6), “offering personal space” (1.6), and its “reliability” (1.6). The lowest scores correspond to “money saving” (–0.2) and “security from traffic” (0.4). In relation to attitudinal beliefs toward public transportation (Figure 2), respondents did not hold a very positive attitude toward commuting by public transportation. The lowest score corresponds to “reliability” of using public transportation to commute (–0.3), followed by “flexibility” (–0.1), “security from people” (–0.1), and “time saving” (0.01). Respondents felt positive about public transportation as being environmentally friendly (1.3) and being secure from traffic (0.9). In relation to the attributes of the commuter travel mode (Figure 3), private car users attached more importance to “reliability” (3.8), “flexibility” (3.7) and “convenience” (3.7). For public transportation users, “reliability” (3.6), “money saving” (3.5) and “time saving” (3.5) were more important factors than others.

In relation to personal norms (Table 2), private car users put a higher score on the statement “I feel good to use private car because it is most efficient” (1.4). They disagreed that they “feel wrong to use a private car every time” (–0.7). In relation to public transportation users, to some extent, they agreed that they felt good when using public transportation because of the positive impact on environment (0.7).

In relation to the descriptive norm beliefs (Table 3), private car users assigned the highest score to the proportion of family members who commute by private car (5.5). Public transportation users perceived that “many” of the people who they work or study with commute by public transportation (5.0).

In relation to injunctive norms (Table 4), respondents were asked the extent to which they complied with the perceptions of various social groups including government and transport authorities, neighboring people, and colleagues (whom they work or study with). Private car users had a higher score in the case of “Government and their transport authorities” (1.5) and “people they study or work with” (1.5). Thus, the respondents were slightly disagreeing about complying with the views of different social groups, with “Government and their transport authorities” and “people they study or work with” being more influential than neighbors. The table also illustrates that, overall, respondents somewhat disagreed with the idea that the surrounding social groups would like to see more people use a private car to commute to work. In the case of public transportation users, respondents held more of a neutral opinion about complying with the abovementioned social groups. However, they somewhat agreed that the government and transport authorities would like to see more people commute by public transportation (1.0).

While evaluating commuting habits (Table 5), private car users strongly agreed that they were more likely to commute by private car when the weather was bad (1.5) and when public transportation was crowded (1.5). Public transportation users somewhat agreed that they did not commute by public transportation when there was a more convenient mode of transport available (0.8).

Table 6 shows the statistics for belief strengths of PBC for private car users and public transportation users. Private car users strongly agreed that “shortage of time” (3.7), “needing to carry heavy or bulky items” (3.7), “having activities after leaving the destination” (3.6), and “predictability in the availability of a car” (3.5) would encourage them to commute by private car. Public transportation users somewhat agreed that lack of public transportation services near residence (–3.4) and destination (–3.4), “needing to carry heavy or bulky items” (–3.4), “having to take a different route” (–3.4), and “lack of frequent services during my time of travel” (–3.3) would discourage them from commuting by public transportation.

Prediction of Intention

In line with previous work, a linear regression model was used to predict commuters’ intention to choose a particular mode to travel ( 7 ). The dependent variable is “Intention of using private car/public transportation to commute to work on the most recent day of commuting” and a five-point scale was used ranging from “Extremely unlikely” (+1) to “Extremely likely” (+5) to score it. Attitudes toward commuting by the particular mode (private car/public transportation), personal norm, injunctive norm, descriptive norm, habit, and PBC were all included into the first model as explanatory variables. The backward elimination method was then used to finalize the model which yielded the best fit. Correlation among the explanatory variables was examined but no significant correlations were observed. The proportion of missing values was less than 5% and those cases were excluded from analysis as suggested in the literature when those cases are of such low incidence ( 30 ). The models which produced the best fit, estimated using SPSS software, were finally selected. The results for the models specific to private car and public transportation are presented separately in Table 7.

Table 7.

Prediction of Intention to Choose a Mode to Travel

	Private car				Public transportation
Predictors	df	F	Beta	p-value	df	F	Beta	p-value
Attitude	3	11.51	0.05	0.05	3	24.84	na	na
Descriptive norm			0.11	0.000			0.06	0.07
Perceived behavioral control (PBC)			0.05	0.05			0.22	0.000
Habit			na	na			−0.24	0.000

Note: na = not applicable.

The R² value for the models specific to private car and public transportation are 0.02 and 0.1, respectively. Thus, the model explained 2% and 10% of the variance of young commuters’ intention to use private car and public transportation, respectively. Several variables were found to be statistically significant in the models presented above. Results indicate that attitude, descriptive norm, and PBC of young commuters significantly contributed to the prediction of their intention to use a private car. In the case of young commuters who usually use public transportation, descriptive norm, PBC, and habit predicted their intention to use public transportation for commuting.

Prediction of Behavior

Consistent with previous research, a binary logistic regression model was used to examine young commuters’ decision to choose a travel mode ( 7 ). In the case of the model specific to “private car,” the dependent variable “Mode to work” takes a value of “1” for those who reported they used a private car to travel to work on the most recent day of commuting and “0” where the respondents reported not having used private car. A similar approach was taken in the case of the model specific to “public transportation.” Again, SPSS software was used to estimate the models. All the previously mentioned variables were assumed to be influential on young commuters’ mode choice behavior and thus included into the first model. As discussed earlier, backward elimination method was then used to finalize the model which yielded the best fit. Correlation among the explanatory variables was examined but no significant correlations were observed. The proportion of missing values was less than 5% and those cases were excluded from analysis as suggested in the literature when those cases are of such low incidence ( 30 ). The models which gave the best results were finally selected. The results for the models specific to private car and public transportation are presented separately in Table 8.

Table 8.

Prediction of Mode Choice for Travel to Work

	Private car			Public transportation
Predictors	B	Wald	Exp(B)	B	Wald	Exp(B)
Intention	0.95	52.32^***	2.58	0.78	64.63^***	2.18
Attitude	0.02	5.22^**	1.02	na	na	na
Descriptive norm	0.14	6.89^***	1.15	na	na	na
Habit	0.12	7.81^***	1.13	−0.13	10.28^***	0.88
Perceived behavioral control (PBC)	na	na	na	0.02	8.45^***	1.02
Constant	−3.17	12.74^***	0.04	−1.28	11.59^***	0.28

Note: na = not applicable.

p-value < 0.1; **p-value < 0.05; ***p-value < 0.01.

When the model was used to predict the probability that a young commuter would use private car to travel to work on the survey day, the model correctly predicted 98.5% of the cases and it was 92.0% for those who used public transportation to work. A probability of 0.5 was used as the cut value. Intention, attitude, descriptive norm, and habit were found to be statistically significant when predicting young commuters’ decisions to commute by private car. PBC was not a significant factor for those commuters who use private car, but it was a significant factor for those who traveled by public transportation. In addition, intention and habit also contributed to young commuters’ decisions to choose public transportation to commute.

Discussion and Conclusion

The research was conducted to investigate young commuters’ travel behavior. TPB was adopted to provide scope to explore a wide range of factors which may help to explain variations in young commuters’ travel choices, particularly from a social-psychological perspective.

Overall, commuters who usually travel by private car held positive beliefs about attitudinal beliefs (e.g., convenience, flexibility, time saving, offers personal space, and reliability) about commuting by private car. However, convenience and flexibility were the most valued advantages that private car offers as a commuting mode. In relation to the attributes of the commuter travel mode, private car users attached more importance to “reliability,”“flexibility,” and “convenience.” For public transportation users, reliability of the commuting mode was more important than other factors such as flexibility and time saving. However, they did not agree that public transportation as a commuting mode provides reliability. The results suggest that more young commuters could be encouraged to use public transportation by making it more reliable.

Private car users indicated that they did not feel wrong to use a private car every time. Public transportation users agreed that they felt good about using public transportation because of the positive impact on the environment. The results suggest that more commuters could be encouraged to use public transportation by means of positive feedback about the environmental benefits of this mode.

With respect to injunctive norms, overall, both private car and public transportation users held more of a neutral opinion about complying with different social groups, including government and their transport authorities, people they study or work with, and neighboring people. Public transportation users agreed to some extent that government and transport authorities would like to see more people to commute by public transportation.

Private car users strongly agreed that they commuted by private car when public transportation was crowded. Also, shortage of time and predictability in the availability of a car would strongly encourage them to commute by car. However, having a more convenient mode of transport available would make public transportation users to shift to alternative modes. Lack of public transportation services near their residence and destination and lack of frequent services during the time of travel would discourage public transportation users from using public transportation more often. Thus, increasing public transportation services at or near major destinations and improved predictability may lead to more young people traveling by public transportation.

Modeling results showed that the construct “attitude” significantly affected young commuters’ intention only in the context of choosing to commute by private car. Descriptive norm and PBC contributed to the intentions of commuters who chose private car or public transportation to commute. However, habit influenced only those who intended to use public transportation. Intention to use a particular commuter travel mode appeared to be the principal factor in forming the decision to actually use the mode to travel. A positive attitude toward the benefits of using a private car was significantly associated with the decision to use a private car, whereas it was insignificant for public transportation users. Thus, considerable scope exists to increase the likelihood of using public transportation by publicizing the benefits of using these modes to commute. PBC was found to significantly influence the decisions of young commuters who traveled by public transportation. Thus, the more commuters have control on their needs such as the need to carry items, frequent and timely services, cost effective service, and so forth, the more likely the commuters would be to use public transportation. End-of-trip facilities such as lockers may assist. More frequent and more reliable public transportation services would encourage young commuters to use public transportation more often.

While this research has enhanced understanding of young commuters’ travel behavior, it is important to be aware of its limitations. With regard to the models for “Prediction of intention” (Table 7), a reasonable number of predictors were found significant but the models had low R² values (0.02 and 0.1 for models of private car and public transportation use, respectively). Including other attributes such as infrastructure to explain commuting mode choice from a different perspective may improve the fit of the models. However, this lies beyond the scope of this study. The influence of socio-demographic attributes (such as gender, income level, etc.) were not analyzed. There would be merit in examining these data to identify the group-wise (male/female) variation, if there is any. From a transport policy perspective, it would be useful for future research to recruit participants to conduct a longitudinal panel survey to examine travel choices on a day-to-day basis. It would be helpful to better understand the motivational factors or the deterrents of choosing a commuting mode. Future research may also examine how young commuters incorporate trip chaining into their decision-making process.

Footnotes

Acknowledgements

The authors wish to thank VicRoads, RACV, and Bicycle Network for their support to this project.

Author Contributions

The authors confirm contribution to the paper as follows: study conception and design: F. Ahmed, J. Catchpole; data collection: T. Edirisinghe; analysis and interpretation of results: T. Edirisinghe, F. Ahmed; draft manuscript preparation: F. Ahmed, J. Catchpole. All authors reviewed the results and approved the final version of the manuscript.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: We received partial financial support from VicRoads, Australia to conduct the research.

References

Transport Accident Commission. https://www.tac.vic.gov.au/. Accessed May 2017.

Simon

Clarys

Bourdeaudhuij

I. D.

Geus

B. D.

Vandelanotte

Deforche

Why Do Young Adults Choose Different Transport Modes? A Focus Group Study. Transport Policy, Vol. 36, 2014, pp. 151–159.

Litman

A New Traffic Safety Paradigm. Victoria Transport Policy Institute, 2017.

Ajzen

Theory of Planned Behaviour. Organizational Behaviour and Human Decision Processes, Vol. 50, 1991, pp. 179–211.

Muñoz

Monzon

Lois

Cycling Habits and Other Psychological Variables Affecting Commuting by Bicycle in Madrid, Spain, Transportation Research Record: Journal of the Transportation Research Board, 2013. 2382: 1–9.

Ahmed

Rose

Jacob

Examining the Impact of Changes in Weather on Commuter Cyclist’s Travel Behaviour. Transportation Research Record: Journal of the Transportation Research Board, 2013. 2387: 76–82.

Ahmed

Rose

Jakob

Commuter’s Cycling Behavior in Response to Weather Variation: Insight from an Extended Theory of Planned Behavior. Presented at 95th Annual Meeting of the Transportation Research Board, Washington, D.C., 2016.

Delbosc

Currie

Impact of Attitudes and Life Stage on Decline in Rates of Driver’s License Acquisition by Young People in Melbourne, Australia. Transportation Research Record: Journal of the Transportation Research Board, 2014. 2452: 62–70.

Delbosc

Delay or Forgo? A Closer Look at Youth Driver Licensing Trends in the United States and Australia. Transportation, Vol. 44, No. 5, 2016, pp. 1–18.

10.

Zhou

Sustainable Commute in a Car-Dominant City: Factors Affecting Alternative Mode Choices among University Students. Transportation Research Part A: Policy and Practice, Vol. 46, No. 7, 2012, pp. 1013–1029.

11.

Simon

Bourdeaudhuij

I. D.

Clarys

Cocker

K. D.

Geus

B. D.

Vandelanotte

Cauwenberg

J. V.

Deforche

Psychosocial and Environmental Correlates of Active and Passive Transport Behaviors in College Educated and Non-College Educated Working Young Adults. PLoS One, Vol. 12, No. 3, 2017, p. e0174263.

12.

Kuhnimhof

Buehler

Dargay

A New Generation: Travel Trends for Young Germans and Britons. Transportation Research Record: Journal of the Transportation Research Board, 2011. 2230: 58–67.

13.

Currie

Delbosc

Mahmoud

Factors Influencing Young Peoples’ Perceptions of Personal Safety on Public transport. Journal of Public Transportation, Vol. 16, No. 1, 2013, pp. 1–20.

14.

Nordfjærn

Simsekoglu

Ö.

Rundmo

Active Transport, Public Transport and Electric Car as Perceived Alternatives in a Motorized Norwegian Sample. Transportation Research Part F: Traffic Psychology and Behaviour, Vol. 42, 2016, pp. 70–79.

15.

Conner

Armitage

C. J.

Extending the Theory of Planned Behavior: A Review and Avenues for Further Research. Journal of Applied Social Psychology, Vol. 28, No. 15, 1998, pp. 1429–1464.

16.

Heath

Gifford

Extending the Theory of Planned Behavior: Predicting the Use of Public Transportation. Journal of Applied Social Psychology, Vol. 32, No. 10, 2002, pp. 2154–2189.

17.

Cialdini

R. B.

Kallgren

C. A.

Reno

R. R.

A Focus Theory of Normative Conduct: A Theoretical Refinement and Re-Evaluation of the Role of Norms in Human Behavior. Advances in Experimental Social Psychology, Vol. 24, 1991, pp. 201–234.

18.

Schwartz

S. H.

Normative Influences on Altruism. In Advances in Experimental Social Psychology ( Berkowitz

, ed.), Vol. 10, Academic Press, London, 1977, pp. 221–279.

19.

Schwartz

S. H.

Howard

J. A.

Explanations of the Moderating Effect of Responsibility Denial on the Personal Norm–Behavior Relationship. Social Psychology Quarterly, Vol. 43, No. 4, 1980, pp. 441–446.

20.

Black

T. R.

Understanding Social Science Research, 2nd ed. SAGE Publications Ltd, London, 2002.

21.

VicRoads. https://www.vicroads.vic.gov.au/.

22.

Royal Automobile Club of Victoria. https://www.racv.com.au/.

23.

Bicycle Network, Australia. https://www.bicyclenetworkvictoria.com.au/.

24.

Australian Road Research Board. https://www.arrb.com.au/.

25.

Fishbein

Ajzen

Predicting and Changing Behaviour: The Reasoned Action Approach. Psychology Press, New York, NY, 2010.

26.

Heinen

Maat

Wee

B. V.

The Role of Attitudes Toward Characteristics of Bicycle Commuting on the Choice to Cycle to Work over Various Distances. Transportation Research Part D: Transport and Environment, Vol. 16, No. 2, 2011, pp. 102–109.

27.

Nigbur

Lyons

Uzzell

Attitude, Norms, Identity and Environmental Behaviour: Using an Expanded Theory of Planned Behaviour to Predict Participation in a Kerbside Recycling Programme. British Journal of Social Psychology, Vol. 49, No. 2, 2010, pp. 259–284.

28.

Terry

D. J.

Hogg

M. A.

White

K. M.

The Theory of Planned Behaviour: Self-Identity, Social Identity and Group Norms. British Journal of Social Psychology, Vol. 38, No. 3, 1999, pp. 225–244.

29.

Australian Bureau of Statistics. Australian Capital Territory, Canberra. http://www.abs.gov.au. Accessed July 30, 2020.

30.

Little

R. J. A.

Rubin

D. B.

Statistical Analysis with Missing Data, 2nd ed. Wiley, Chichester, 2002.