1. Introduction

Previous Literature

Modal choice modelling and choices between public transport and private transport are very relevant and influential part for the understanding and prediction of future travel behaviour [5,6,7]. Theories of social science have also been developed and used heavily to understand and assess travel behaviour [8,9,10,11,12]. There are many studies in transportation planning processes that applied both travel behavioural and social science theories [8,9,10,11,12,13,14,15,16,17,18,19]. In traditional modal split studies, the market would be divided mainly between private and public modes of travel. These models were presented as diversion curves that were essentially created on the basis of one or two characters of the trips that dominated the generalized cost functions of the choice models (eg. in vehicle travel time or cost) [20,21,22]. Essentially, S-shaped curves, appear to characterise the relationship between the cost difference and the proportion of trips made by each option [23,24]. Figure 1 shows a typical S-shaped curve representation of the split and cost difference between two options (1&2). The components of the cost difference between the two options have been developed since to include qualitative and quantitative level of service factors, social, and attitudes and behavioural factors.

As Figure 1 shows, most travellers have a choice between the two options (say PT and private travel options), with various levels of cost difference. In these cases, travellers assess the available travel options based on the level of service including travel time, travel cost, comfort, convenience, parking, walking, waiting times, age, preferences etc., and select the option that maximises their utilities [25,26]. Towards the two ends of the curve, there are the captives’ zones. At the right-hand side end, there are the captives to option 1 (say PT) with a very large cost difference; those who cannot access private cars because of the very high cost (they do not own cars, don’t possess driving license or physically impaired, and so on). At the left-hand side of the curve, there are the less usual type of travellers, those who are captives to option 2 (say private cars). Those are those travellers who do have access to the private cars but not to public transportation options because they live far from them for example or don’t perceive public transportation as a feasible option of travel for them as the case in Riyadh and many other similar cities. In most of the low-level of income countries’, there are more “captives” to public transportation options than to private cars.

In higher income cities and countries that have specific cultures, the captivity to the private car is experienced. For example, in the city of Riyadh and most cities in the KSA, almost all travellers are captives or confined to the private cars rather than being choosers between the private and public options of travel. This is a result of the relatively high-income levels in the country, very high level of car ownership and the very low level of provision of public transportation or in some cases the complete lack of it. This, in addition to the cultural norms of not accepting to ride on the public modes of travel, makes it a big challenge to expect travellers in these cities to use public transport.

The Kingdom of Saudi Arabia is working towards achieving more sustainable travel behaviour and attitudes by constructing very high level of service public transportation options including a state-of-the-art metro system, which is very near completion and operation. The local travellers, therefore, are aware of such options and aware of the need to shift some of their travel activities towards such options [1,4]. Moreover, a large proportion of the population have travelled abroad and to the neighbouring countries and acquired different travel experience and awareness of the public modes of transportation. All these factors are expected to lead to changing travel behaviour and attitudes towards public transportation systems.

The theory of random utility is based on the assumption that individuals make decisions that maximise their benefits or utilities and that they have full information on all available option in their choice sets. The theory is heavily used in travel behaviour and travel choices. The utilisation of the random utility, or discrete choice models that model decision-makers choices from a set of mutually exclusive options have been adopted in this study. The theory of random utility is based on the assumption that individuals make decisions that maximises their benefits or utilities and that they have full information on all available option in their choice sets. The theory is heavily used in travel behaviour and travel choices. The utilisation of the random utility, or discrete choice models that model decision-makers choices from a set of mutually exclusive options have been adopted in this study.

There are very few available publications on the modal choice and use of travel options in countries such as the KSA and other similar oil rich countries. Relevant studies on these focusses include who claim that high income and economic factors have contributed to shaping the lifestyle in Saudi Arabia that resulted in the high dependency on private cars [3]. A study reported on a stated preference investigation that assessed the potential commuters’ perception towards the proposed metro system in Riyadh, as a function of some socio-economic as well as some level of service variables [1,4]. Other studies include investigation of public perception on public transportation system in Dammam [28]. There is a lack of studies however, that are looking and investigating the public attitudes, perception and acceptability of public transportation systems in these countries.

Dubai is a neighbouring city to KSA, that was the first Arabic Gulf city to introduce integrated transport planning, implement travel demand management and build a metro system in order to achieve shifting car traffic to public transport traffic. The aim was also to reduce negative traffic impacts and achieve the sustainability goals. Such shift of demand for public service entails not only market motivations to bringdown private vehicle ownership and increase the public transport services, but also an understanding of how much peoples are willing to use and pay for improved public transport services. As well as building public transportation systems, Dubai has introduced financial incentives and disincentives, legislative measures and raising awareness programs in order to facilitate the shift. This approach has resulted in success to shift some of the car trips to public transportation, walking, cycling as well as increasing acceptability of such options [29,30].

Oman is another Middle Easter country that has identified the importance of public transportation modes of travel. In a public opinion study that was carried out in Oman, it was reported that public transport facilities in the city are minimal and do not meet the demand, and that the private car is still the main dominant mode of travel in the cities [31]. Marketing of public transport services is constrained by certain issues including the socio-cultural and physical environments. The study produced recommendations to policy makers in Oman to establishing viable options to public transport solutions. Doha city in Qatar, addressing the rapid-growing demand on transportation in the city and in order to reduce the traffic congestion on the roads, the country has built a new metro system to connect all the major points in the country. The metro network has four lines with an overall length of approximately 340 km and 98 stations [32].

Understanding issues related to travel behaviour and mode choices have dominated many researchers and scholars’ interest. Many studies have been performed on different aspects of travel patterns, choices and decisions. The factors that are included in the cost functions have been developed further on to include more characteristics such as social factors, modes’ characteristics such as comfort, reliability and attitudinal factors [20,21,22,23].

Impacts of social economic factors on travel behaviour and their influence on the modal choices are evident in the literature. A number of studies reported some gender differences in terms of accessibility and travel experiences in a case study of Sofia, Bulgaria and Germany [32,33,34]. They report that women have less accessibility in comparison with men who use identical travel options. Another study investigated and reported on significant differences of the social implications for using different types of public transportation, including metro and buses in South Asia [35]. Three studies looked at impact of gender and income, levels of education, age, employment status and car ownership on travel behaviour and choices with a specific interest on Saudi women [35,36,37]. An investigation of mobility patterns and its impacts on culture and the structure of the household and relation to the transport patterns in terms of gender differences has been reported [38]. The study concluded on the variables’ impact of cost of travel and time duration on the choices of modes of travel in relation to employment destinations. Commuters’ perceptions on the employment decentralization and impacts on their mode choices in Chine has been reported in [37].

Swedish women attitudes towards sustainability have been investigated and reported [39]. The outcome of the study shows that Swedish women are more affirmative towards sustainability and are more likely to reduce their car travel in comparison with men. A study reported on a recent UK research that young women tend to travel more than young men. The authors also concluded that young men travelled substantially less than their counterpart young men in previous generations, which could indicate that different generations face different socio-economic contexts which impact on mobility trends. Similarly, Germany studies have found that gender impacts also affect travel behaviour.

Household structure and patterns of mobility on modal choices in Paris have been investigated and reported [40]. The impacts of position in the family on work trips behaviour and travel in Oslo have been assessed and stated [41]. In a comparative assessment, assessed the lower income categories of women and their levels of accessibility to public transportation in China and India.

A study on the demographic variables in Libya such as age and gender has shown a significant contribution to explain mode choice behaviour. In addition, the study showed that men were less likely than women to shift to public transport [43]. They also found that women tend to travel by car more often as passengers, while men tend to be the drivers. A study found that for students in Beirut, gender is one of the main factors affecting mode choice. Compared to females, males were more likely to use the bus. A study investigated the travel characteristics of female college students in Saudi Arabia. They found that 57% travelled as car (or van) passengers and 39% travelled by bus and over half (53%) were captives to their current mode of transport.

There has also been significant research on gender differences and travel patterns in the West with less research has been done in other part of the world including the Middle East, where cultural and background factors are significantly different than they are in the West, which could have impacts on travel behaviour [38,39,40]. For example, some studies have attempted to examine gendered differences in travel behaviour in the Arab world [41,42,43,44,45]. In such countries, women’s travel attitudes and behaviour are affected by economic and geographical factors as well as other social and cultural differences and restrictions.

Logistic regression analysis is a very extensively utilised analysis tool to analyse relationships between a dependent and a number of independent variables, using the concept of random utility. In regression analysis, the dependent variable is often discrete, with one or more possible options, and is considered binary or multiple classes. This type of analysis uses linear regression estimation and results in a logit function that best describe the function form between the dependent and independent variables. In regression analysis when the dependent variable Z is a continuous variable, given by eqn. (1):

Z = α + β1 x1 + β2 x2 + · · · + βk xk + ε

(1)

where f (x) is linear in the parameters function and ε is the error term that can follow a distribution, and a model will result based on the type of utilised distribution ([25,26]. While the multinomial logit models allow for more than two categories of the dependent variable, the binary logit models allow only for two categories. The calibration of a binary or multinomial logistic regression model (with dependent variable classes) involves the calibration of the independent coefficients that predict the probability of the outcome of interest. One possible form of the function is the linear in the parameters form with probabilities are given by the following expression in eqn. [46,47]:

ln (prob.(an event occurring))/(1-prob.(an event occurring))

(2)

The coefficients in the logistic regression give information as to how much the logit changes depending on the independent variable values. This is applicable for two events, but when there are more than two events this binary logistic regression can be extended. The term on the left-hand side is referred to the log of the odds that an event occurs. The Logit model (binary or multiple) is mostly estimated using maximum likelihood estimation methods. These methods estimate the values of the parameters of an assumed probability distribution, given some observed data [25]. Regression techniques are also used to evaluate many applications in the area of discrete choice analysis [25].

Where the dependent variable is ordinal, the logistic regression analysis is referred to as ordinal logistic regression (OLR). When there is a systematic order in the de-pendent variable classes or categories, the ordinal logistic regression can be used to predict the dependent variable as a function of independent variables. To calibrate the coefficients of an ordinal logistic regression model, the levels of the dependent variable have to be defined [25,26,27]. The mathematical formula of the odds of proportional of the OLR can then be written as in eqn. (3):

logit [P(Y≤ d)] = αd – ΣβnXn

(3)

3. Methodology

4. Analysis of Survey Data

5. Ordered Logit Regression Model

6. Conclusions and Recommendations

7. Limitations

References

1. Omar Alotaibi & Dimitris Potoglou (2017) Perspectives of travel strategies in light of the new metro and bus networks in Riyadh City, Saudi Arabia, Transportation Planning and Technology, 40:1, 4-27. [CrossRef]
2. Alqahtany, A. 2014. The Development of a Consensus-Based Framework for a Sustainable Urban Planning of the City of Riyadh. Cardiff: Cardiff School of Engineering, Cardiff University.
3. Mubarak, F. A.2004.“Urban Growth Boundary Policy and Residential Suburbanization: Riyadh, Saudi Arabia.”Habitat International28 (4): 567–591.
4. Youssef, Zaher, Habib Alshuwaikhat, and Imran Reza. 2021. "Modeling the Modal Shift towards a More Sustainable Transport by Stated Preference in Riyadh, Saudi Arabia" Sustainability 13, no. 1: 337. [CrossRef]
5. Bhat, C. R., A generalized multiple duration proportional hazard model with an application to activity behaviour during the evening work-to-.
6. home commute, Transportation Research Part B 30, 1996, 465-480.
7. Tim Schwanen, Patricia L. Mokhtarian, What affects commute mode choice: neighborhood physical structure or preferences toward. n: Mokhtarian, What affects commute mode choice. [CrossRef]
8. neighborhoods?, Journal of Transport Geography, 13, 2005, 83–99.
9. Sofia Martin-Puerta, Commuting mode choice: Motivational determinants and road users profile, Transport Research Arena, 2014, Paris.
10. Boudon R (1982) The Unintended Consequences of Social Action. London: Macmillan Press. Coleman JS (1990) Foundations of Social Theory. Cambridge, MA: The Belknap Press of Harvard University Press.
11. Aguiar F, de Francisco A (2009) Rational choice, social identity, and beliefs about oneself. Philosophy of the Social Sciences 39(4): 547–71.
12. Coleman JS (1990) Foundations of Social Theory. Cambridge, MA: The Belknap Press of Harvard University Press.
13. Coleman JS (1986) Social theory, social research, and a theory of action. Am J Sociol 91:1309–1335.
14. Liebe, U., Preisendörfer, P. (2010). Rational Choice Theory and the Environment: Variants, Applications, and New Trends. In: Gross, M., Heinrichs, H. (eds) Environmental Sociology. Springer, Dordrecht. [CrossRef]
15. Saleh, W.; Malibari, A. (2021) Saudi Women and Vision 2030: Bridging the Gap? Behav. Sci. 2021, 11, 132. [CrossRef]
16. ang, X., Day, J. E., Langford, B. C., Cherry, C. R., Jones, L. R., Han, S. S., & Sun, J. (2017). Commute responses to employment decentralization: Anticipated versus actual mode choice behaviors of new town employees in Kunming, China. Transportation Research Part D: Transport and Environment, 52, 454-470. [CrossRef]
17. A Vega, Amaya, Reynolds-Feighan, Aisling (2008)Employment Sub-centres and Travel-to-Work Mode Choice in the Dublin Region. Journal of Urban Studies. 1747-1768, Volume 45, Number 9.
18. Yang, X., Day, J. E., Langford, B. C., Cherry, C. R., Jones, L. R., Han, S. S., & Sun, J. (2017). Commute responses to employment decentralization: Anticipated versus actual mode choice behaviors of new town employees in Kunming, China. Transportation Research Part D: Transport and Environment, 52, 454-470. [CrossRef]
19. Polk, M. Are Women Potentially More Accommodating Than Men to a Sustainable Transportation System in Sweden? Transp. Res. D Transp. Environ. 2005, 8, 75–95.
20. Bernard, A.; Seguin, A.; Bussiere, Y. Household Structure and Mobility ‘Patterns of Women in O-D Surveys: Methods and Results Based on the Case Studies of Montreal and Paris; The National Academies of Sciences: Washington, DC, USA, 2013.
21. Hjorthol, R.J. Same city--different options: An analysis of the work trips of married couples in the metropolitan area of Oslo. J. Transp. Geogr. 2000, 8, 213–220. [CrossRef]
22. Frank, L., Bradley, M., Kavage, S. et al. (2008) Urban form, travel time, and cost relationships with tour complexity and mode choice. Transportation 35, 37–54. [CrossRef]
23. Roya Etminani-Ghasrodashti, Mahyar Ardeshiri, (2015) Modeling travel behavior by the structural relationships between lifestyle, built environment and non-working trips, Transportation Research Part A: Policy and Practice, Volume 78, Pages 506-518, ISSN 0965-8564. [CrossRef]
24. Narisra Limtanakool, Martin Dijst, Tim Schwanen (2006) The influence of socioeconomic characteristics, land use and travel time considerations on mode choice for medium- and longer-distance trips, Journal of Transport Geography, Volume 14, Issue 5, Pages 327-341, ISSN 0966-6923. [CrossRef]
25. Salter, R.J. (1976). Modal split. In: Highway Traffic Analysis and Design. Palgrave, London. [CrossRef]
26. Lathrop, G. T., Hamburg, J.R., and Young, G. F. Opportunity-Accessibility Model for Allocating Regional Growth. Highway Research Record 102, 1965, pp. 54-66.
27. De Dios Ortúzar, J. & Willumsen, L. G. (2011) Modelling transport, John wiley & sons.
28. W.H. Greene, D.A. Hensher (2003) A latent class model for discrete choice analysis: contrasts with mixed logit. Transportation Research Part B: Methodologica.
29. Al-Rashid, M.A.; Nahiduzzaman, K.M.; Ahmed, S.; Campisi, T.; Akgün, N. (2020) Gender-Responsive Public Transportation in the Dammam Metropolitan Region, Saudi Arabia. Sustainability, 12, 9068.
30. Al-Fouzan, S.A. (2012) Using car parking requirements to promote sustainable transport development in the Kingdom of Saudi Arabia. Cities, 29, 201–211. [CrossRef]
31. Chaudhry, G.A., 2012. Evolution of the transportation system in Dubai. Network Industries Quarterly, 14(1), pp.7-11.
32. Abouelhamd, I., Radwan, I. and Abed, O., 2020. Conceptual Development Solutions for Traffic Jams & Accidents in Dubai City.
33. Rakesh Belwal, Shweta Belwal, (2010) Public Transportation Services in Oman: A Study of Public Perceptions, Journal of Public Transportation, Volume 13, Issue 4, Pages 1-21, ISSN 1077-291X. [CrossRef]
34. Kwan, M.-P.; Kotsev, A. Gender differences in commute time and accessibility in Sofia, Bulgaria: A study using 3D geovisualisation. Geogr. J. 2014, 181, 1–14. [CrossRef]
35. Best, H.; Lanzendorf, M. Division of labour and gender differences in metropolitan car use: An empirical study in Cologne, Germany. J. Transp. Geogr. 2004, 13, 109–121. 2004; 13, 109–121. [CrossRef]
36. Boarnet, M.G.; Sarmiento, S. Can Land-use Policy Really Affect Travel Behaviour? A Study of the Link between Non-work Travel and Land-use Characteristics. Urban Stud. 1998, 35, 1155–1169. [CrossRef]
37. Zhou, J.P., Sustainable commute in a car-dominant city: factors affecting alternative mode choices among university students, Transportation Research Part A – Policy and Practice 46. 2012, 1013–1029. [CrossRef]
38. Saleh, W.; Malibari, A. (2021) Saudi Women and Vision 2030: Bridging the Gap? Behav. Sci. 2021, 11, 132. [CrossRef]
39. Yang, X., Day, J. E., Langford, B. C., Cherry, C. R., Jones, L. R., Han, S. S., & Sun, J. (2017). Commute responses to employment decentralization: Anticipated versus actual mode choice behaviors of new town employees in Kunming, China. Transportation Research Part D: Transport and Environment, 52, 454-470. [CrossRef]
40. A Vega, Amaya, Reynolds-Feighan, Aisling (2008)Employment Sub-centres and Travel-to-Work Mode Choice in the Dublin Region. Journal of Urban Studies. 1747-1768, Volume 45, Number 9.
41. Polk, M. Are Women Potentially More Accommodating Than Men to a Sustainable Transportation System in Sweden? Transp. Res. D Transp. Environ. 2005, 8, 75–95.
42. Bernard, A.; Seguin, A.; Bussiere, Y. Household Structure and Mobility ‘Patterns of Women in O-D Surveys: Methods and Results Based on the Case Studies of Montreal and Paris; The National Academies of Sciences: Washington, DC, USA, 2013. The National Academies of Sciences: Washington, DC, USA.
43. Hjorthol, R.J. Same city--different options: An analysis of the work trips of married couples in the metropolitan area of Oslo. J. Transp. Geogr. 2000, 8, 213–220. [CrossRef]
44. Sirnivasan, S. A Spatial Exploration of the Accessibility of Low-Income Women; Chengdu, China and Chennai, India. In Gendered Mobilities; Uteng, T.P., Cresswell, T., Eds.; Ashgate: Aldershot, UK, 2008; pp. 173–192.
45. Abuhamoud, M.; Rahmat, R.; Ismail, A. Transportation and its concerns in Africa: A review. Soc. Sci. 2013, 6, 21–63. [CrossRef]
46. Al-Ahmadi, H.M. Travel Characteristics of Female Students to Colleges in the Kingdom of Saudi Arabia. J. Eng. Res. 2006, 3, 79–83. [CrossRef]
47. Hamed, M.; Olaywah, H. Travel-related decisions by bus, service taxi, and private car commuters in the city of Amman, Jordan. Cities 2000, 17, 63–71. [CrossRef]
48. Mangham, L.J., Hanson, K. and McPake, B., (2009). How to do (or not to do?). Designing a discrete choice experiment for application in a low-income country. Health policy and planning, 24(2), pp.151-158. [CrossRef]
49. Ledolter, J., (2013). Multinomial logistic regression. Data Mining and Business Analytics with R, pp.132-149.
50. Gentleman RC, Carey VJ, Bates DM, Bolstad B, Dettling M, Dudoit S, Ellis B, Gautier L, Ge Y, Gentry J, Hornik K, Hothorn T, Huber W, Iacus S, Irizarry R, Leisch F, Li C, Maechler M, Rossini AJ, Sawitzki G, Smith C, Smyth G, Tierney L, Yang JY, Zhang J. Genome Biol 2004: 5(10); R80.