Gender Differences in Influence of Socio-demographic Characteristics on Mode Choice in India

Abstract

The study aims to investigate differences between men and women in influence of various socio-demographic factors on choice of mode of transport. For this purpose, a binary logit model of choice probabilities is implemented on survey data of a developing country city. Results indicate women's choice of travel mode to be more environment-friendly than that of men. Well-educated, working and middle-aged individuals appear to be the most likely to choosing more-polluting modes of transport for frequent travelling purposes. Individuals in the sample who are the least socioeconomically well off are found the most likely to be promising for the environment. The findings of this study suggest the future transportation policies toward development of existing infrastructure of greener modes of transportation in the city such as, public transportation services and pedestrian lanes, so as to manage the rising issues of degrading environmental quality. The study highlights how the consideration and inclusion of socio-demographic factors is crucial for policy recommendation regarding curtailing the environmental damages contributed by transportation sector. Because mobility crucially affects all other indicators of empowerment, and women are the ones using green modes extensively, the city's transportation system should be so developed which gives their safety and security due importance.

1. Introduction

Climate change is a pressing global challenge demanding significant attention from economies around the world (Faradiba & Zet, 2020). Industry, agriculture, transportation, electricity and heat production are few sectors contributing to the global greenhouse gas emission, of which the contribution of the transportation sector is extensive (Lai, Huang, Siang, & Weng, 2020). According to World Health Organization (WHO) estimates released in 2010, the transportation sector contributes about 23% to global carbon dioxide emissions. Environment protection is crucial because of its life- giving properties and its contribution in sustainable growth of a nation (Khan, Imdadul, Haque, & Khan, 2020).

In India, the transportation sector, which is growing at an increasing pace (Vongurai, 2020), stands at the third place in the ranking of sectors by level of greenhouse gas emissions, of which road transport is found to play a dominant role (Singh, Mishra, & Banerjee, 2019). Six Indian cities hold a place in ten most polluted cities in the world (Broom, 2020). Air pollution is found to claim 4.2 million lives in the world every year (WHO, 2014), of which one million are Indian (Chatterjee, 2019).

According to Wilbur Smith and MoUD (2008), in most Indian cities, the share of public transport in total modal use has declined to less than 50%. The same study describes the modal share in Jaipur as: walking 26%, cycle 13%, motorized two-wheeler 26%, motorized four-wheeler 4%, public transport 22%, and intermediate public transport 4%. Another study by Jaipur Metro Rail Corporation Ltd. (2018) finds the following modal distribution in the city, approximately: walking 16%, cycle 6%, bus 18%, car and taxi 19%, two-wheelers 32%, and auto rickshaw 9%. The current study finds that approximately 26% of the population of the city walks, 2% use cycle, 37% take motorized two- wheelers, 8% use motorized four-wheelers, and 27% use public transport.

From the Census of India, 2011 data released in 2016 about the mode of transport taken to work, the share of public transport in total work trips is only 18.1% (Census of India, 2011c). The declining preference of the population to use public transportation in India can be credited to the multiple challenges that this mode faces like under-capacity to meet the needs of the entire population and excessive congestion, among others (IIHS, 2015). A report by Times of India (2019) claims that approximately 72% of the roads in the Jaipur do not have a footpath and the remaining roads where footpaths existed were mostly parked with vehicles.

There exist gender differences in the damage caused to environment, concern for it, willingness to adapt to changing conditions and vulnerability for it. Studies claim significant differences in level of carbon dioxide emissions from men and women’s mode of transport (Carlsson-Kanyama, Lindén, & Thelander, 1999). Women are usually more concerned for the changing environmental conditions and more willing to adapt to them (Momsen, 2000; Xiao & McCright, 2015; Yadav & Lal, 2018). This study examines the type of mode of transport used by men and women in the city to hint at the differences in pollution caused by them and the simultaneous differences in concern for environment.

The objective of the present study is: (i) to highlight the type of modes of transport used by different sections of the population; (ii) to analyze the influence of various socio-demographic factors on the choice of mode of transport; (iii) to evaluate the gender differences that exist in the impact of these factors.

Findings of the current study call attention to the need to integrate the socioeconomic and socio-demographic factors such as gender, age, level of educational attainment, work status, and socioeconomic status along with other crucial factors while formulating policies to effectively mitigate the soaring global temperature. The study also displays how significantly lower number of people in the city are walking or cycling for short distances and using public transport for long distances. This brings us to questioning the adequacy of infrastructure in the city along with the lifestyle of people, which is more inclined towards leisure. It highlights that women are presently contributing lesser to environmental degradation than men. However, if the climate policies require that at least the ecological footprints by women be maintained, this will stand in the way of their empowerment. This paper proposes the need for an efficient and secured public transportation system in Jaipur to not only maintain the carbon footprints of women, but also reduce that of men by a significant amount. 

This section is followed by a section on review of literature on the topic. The next section introduces the city followed by a description of data and methodology used. The fifth section presents the major findings. The next section concludes and reflects on policy implications.

2. Literature Review

2.1. Gender Differences in Contribution to Environment Destruction

Gender mainstreaming of climate change is not new, but has long been an ignored topic of discussion. Men in the global South will be affected less adversely by climate change than women in those countries and that women in the global North pollute less than their male counterparts (Arora- Jonsson, 2011). Studies have been conducted in the past, which show that men consume more energy than women. CO2 emissions from men’s mode of transportation are higher than those from women’s mode of transportation in Sweden and Greece (Carlsson-Kanyama, Lindén, & Thelander, 1999; Räty & Carlsson-Kanyama, 2010). Men also use more energy for travel, eating out, alcohol and tobacco. Cohen (2014) finds that men work to a great degree in emission generating industries and also their choice of transport vehicles makes their CO2 emission rates greater than women’s.

The role of women in mitigation measures should not be under-estimated. While comparing the knowledge and concern of climate change of men and women McCright, (2010) observed that women have greater scientific knowledge and concern of climate change than men and they underestimate their knowledge of it more than men. Women also have demonstrated their role as climate change mitigators and conservators in movements like Chipko movement in India, Green Belt movement of Kenya, Love Canal Movement of New York, Women’s Pentagon Actions, Akwesasne Mother Milk Project at Mohawk, and Greening of Harlem Coalition at Harlem, etc. In a survey conducted in Australia it was observed that more number of females reported that they had made changes to the way they lived their lives due to the risk of global warming (Agho, Stevens, Taylor, Barr, & Raphael, 2010).

2.2. Socio-demographic Factors and Transport

2.2.1. Gender

The idea of engulfing gender into transport is not an old one (Duchène, 2011). Studies have identified differences existing in usage of different modes of transport. Women are reportedly ‘no choice walkers' because other modes of transport are either unaffordable or inconvenient (World Bank Group, 2012). They work within walking distance from their homes because the burden of responsibilities is uneven in the household (Anand & Tiwari, 2007). Cycling is not a mode of transport, which they usually make use of. Clothing requirements, domestic responsibilities, gendered perceptions of cycling and low-quality cycling environment are some of the grounds on which women disregard cycling in not only developing nations, but also developed nations.

Literature reveals women’s higher dependency on public transport than men. In Sweden, men travel longer distances in cars and women travel similar distances by public transport (Carlsson-Kanyama, Lindén, & Thelander, 1999). Women prefer public transport over private ones in Vishakhapatnam (Jain & Tiwari, 2019). With growth in household income in Rajkot, men switch over to motorized vehicles and women to public transport (Mahadevia & Advani, 2016). Elderly women have higher odds of choosing public transport than elderly men in Netherlands (Böcker, van Amen, & Helbich, 2017).

The pattern of use of motorized vehicles also shows signs of existing gender disparities (Srinivasan & Rogers, 2005). In households that own private vehicles, it is often only the men who get to drive with women being co-passengers (Carlsson-Kanyama, Lindén & Thelander, 1999). Women are much less likely to have access to motorized means of transport compared to men (Peters, 2002). This limited use of private motor vehicles can be accredited either to less access to economic means or unfavorable culture and customs of the society.

2.2.2. Age

Choice of mode of transport varies across age groups. Hjorthol, Levin, and Sirén (2010) find that elderly generation maintains the use of private cars in Denmark, Norway and Sweden. Use of cars by older people is soaring in UK (Li, Raeside, Chen, & McQuaid, 2012). The older generation is found to be less likely to using public bicycles to access rail transit in China (Ji et al., 2017). School-going children in Kanpur, India are depending on motorized vehicles due to absence of an efficient public transportation system in the city (Singh & Vasudevan, 2018).

2.2.3. Socioeconomic Status

One determinant of socioeconomic status, which is of paramount importance in travel behavior studies, is income (Ahmad & Puppim de Oliveira, 2016; Manoj & Verma, 2015; Diaz Olvera, Plat, & Pochet, 2015). It controls the ability to own and use motorized vehicles (Li, Raeside, Chen, & McQuaid, 2012). People with lower income are generally more inclined to using non-motorized vehicles or public transport (Buehler, Pucher, & Bauman, 2020; Manoj & Verma, 2015). As income improves, so does the socioeconomic status and with it is found to improve the usage and ownership of motorized vehicles (Bansal, Kockelman, Schievelbein, & Schauer-West, 2018; Jain & Tiwari, 2019). However, under-reporting of data is one flaw related to income. The presence of informal sector, self- employment, seasonal variability in earnings (Diaz Olvera, Plat, & Pochet, 2015), recall bias, difficulty in accounting for all income generating activities like rents, pensions, (Jain & Tiwari, 2019), and presence of corruption (Saha, Roy, & Kar, 2014) are some of the factors promoting under-reporting of income data. There is, therefore, a need to use a proxy of income (Jain & Tiwari, 2019) to achieve our desired objective.

2.2.4. Work Status

Those who are employed are more likely to use private vehicles than those who are not (Nkeki & Asikhia, 2019). A study conducted in Seoul, South Korea, reveals employed people’s stronger propensity toward car transportation than unemployed people (Ko, Lee, & Byun, 2019). There also exist gender differences in travel behavior among full-time employees in Germany due to disproportionate burden of household responsibilities on women (Nobis & Lenz, 2004).

2.2.5. Education

Sovacool, Kester, Noel and, de Rubens (2018) find that individuals with higher levels of education are more prone to buying low-carbon electric vehicles in Nordic region. Well- educated population prefers private modes of commuting in Benin, Nigeria (Nkeki & Asikhia, 2019). On the other hand, another study finds walking and cycling rates to be highest among the educated population in US (Buehler, Pucher, & Bauman, 2020).

3. Study Area

The study area, Jaipur, is the capital of the largest north Indian state. Spread across an area of 484.64 sq. km (Smart Cities India, 2016), this city is the largest in the state by area. It holds a place in the top ten most populous cities in the country with a population of 30,73,350 (Census of India, 2011a). Population in the city grew by approximately 32.30% in 2011 as against 2001 (Census of India, 2011a). The average literacy rate of the city is 83.33% and there are approximately 900 females for every 1000 male (Census of India, 2011a). Most of the population of the city follows Hinduism. This city is also one of the earliest planned cities of modern India. It makes a popular tourist destination in the country by being a part of the West Golden Triangle tourist circuit and being a UNESCO World Heritage Site. The economy of the city is triggered by tourism, gemstone cutting, manufacturing jewelry and luxury textiles, and information technology.

Jaipur Municipal Corporation is divided into eight zones comprising of 91 wards. The city buses are operated by Jaipur City Transport Services Limited of Rajasthan State Road Transport Corporation. The service operates more than 400 regular and low-floor buses. The total number of government and private city buses in 2007 were 327 (Wilbur Smith, & MoUD, 2008).

The 150km six-lanes ring road encircling the city and Jaipur Rapid Bus Transit System (BRTS) is proposed to solve the problem of traffic. Though Jaipur Metro has commenced its operation, our study finds limited use of the same in the city. Usage of metro reduces carbon emission per person by a significant amount. A report by Times of India (2019) finds that approximately 72% of the roads in the city do not have a footpath and the remaining roads where footpaths existed were mostly parked with vehicles. The usage of bicycle is minimal in the city. In India, approximately 81% of the total automobile domestic sales in 2018-19 are of motorized two-wheelers (Society of Indian Automobile Manufacturers, 2018). We observe a similar trend in Jaipur where the use of motorized two-wheelers is ample and frequent.

4. Data

For the purpose of sampling, a sample proportionate to each of the eight zones into which Jaipur Municipal Corporation is divided is selected. The 600 individuals, between the age of 18 and 60 years, comprising the sample are stratified on the basis of sex ratio and average literacy rate in the city. Data on socioeconomic characteristics, asset ownership of the household and individual information was collected from each surveyed individual.

The study defines a trip as a one-way intra-city trip made within the boundaries of the city. Access and egress trips are considered part of the main trip. The mode of transport used to travel the longest distance within the trip is considered the mode of the whole trip.

The study categorizes an individual’s choice of mode of transport into two categories. First category is green modes, where individuals choose walking and cycling to travel short distances (up to 5km) and opt for using public transport for long distances (more than 5 km). Non-green modes make up the second category where either mode such as motorized two-wheelers, motorized four-wheelers or hired taxis are used for both short and long distances or public transport is used for short distances. The reason why we use the term ‘green’ owes to the existing differences in carbon footprint per person of the two types.

The socio-demographic and socioeconomic factors make up the explanatory variables of the study. These include age, educational attainment, work status and socioeconomic status of individuals. There are three age groups into which the sample is divided. First is of individuals in the age group of 18-25 years who are the youngest, second of those in the group of 26-39 years of age and the last category is of individuals belonging to the eldest age group of 40-59 years. 

This study uses Census of India (2011b) definition of work, which defines it as participation in any economically productive activity with or without compensation, wages or profit. This definition of work also differentiates between workers and non-workers. The non-workers include unemployed, students, and housewives or homemakers who are attending to daily household chores. Workers, on the other hand, include those who are self-employed in non-agricultural activities, earning regular wage/salary or working as casual labors.

There are three categories into which individuals are divided based on the level of education attained by them. Those having no formal schooling or having studied till class 4 make the first category, individuals having studied till class 5 to class 12 comprise the second category and those who have attained education above class 12 comprise the third category.

The socio-economic groups are defined as follows: 

SEG1: This category comprises of individuals belonging to low socio-economic background. The monthly per capita consumption expenditure of people belonging to this group is less than or equal to INR2,500. In case of missing values for consumption expenditure, individuals whose household lives in houses, which they have occupied without paying any rent or own a kutcha house are adjusted in SEG1.

SEG2: This group corresponds to individuals belonging to middle socio-economic background where monthly per-capita consumption expenditure is within the range of INR2,501 to INR7,000. Individuals where the consumption expenditure data is missing, ownership of semi-pucca houses or where the household lives in rented houses are considered in SEG2.

SEG3: Individuals belonging to this group classify as those belonging to a high socio-economic background. The monthly per capita consumption expenditure of individuals in this group is above INR7,000. For missing values, individuals whose household owns a pucca house are considered to be a part of SEG3.

While the primary focus of this study is to examine the gender differences in the impact that socio-demographic factors have in selection of the type of mode of transport, one important variable that is expected to affect this decision is cost of travel (Vrtic et al., 2010). Therefore, cost of travel for corresponding purpose is assumed as a control variable in this study.

This study has the following limitations: (i) it does not take into account the actual per-capita income values for calculating the socio-economic categories; (ii) it fails to calculate the combined effect of two or more socioeconomic factors on choice of travel modes; and (iii) it does not take into account the impact of travelling for different purposes on choice of travel modes.

5. Methodology

Few studies determine choice of mode of transport using discrete choice models (Li, Raeside, Chen, & McQuaid, 2018; Liu, 2007; Ng, Law, Wong, & Kulanthaya, 2013; Singh & Vasudevan, 2018; Srinivasan & Rogers, 2005; Szeto, Yang, Wong, Li, & Wong, 2017). Logit models estimate the probability of a certain event for a linear combination of variables. This study uses binary logit model to identify the impact that different socio-demographic and socioeconomic variables have on choice of green modes of transport. This analysis has been done separately for men and women to analyze the gender differences in the impact of these factors on mode choice. The dependent variable is a binary response variable having choice of green modes of transport (choosing walking/cycling to travel short distances and taking public transport for long distances) and choice of non-green modes of transport (opting to use motorized two-wheelers/ motorized four-wheelers/hired taxi to travel short or long distances and taking public transport for short distances) as the two categories. For the respondent’s travel decision of green mode, this model takes a value of 1 with a probability of p and when the choice is made for non- green mode, a value of 0 with probability 1-p is assumed. Socio-demographic and socioeconomic variables, which are the explanatory variables in the model, have been used as categorical variables. The equation of the binary logistic model can be as follows:

\(ln(\frac G {1-G}) = \beta_0 + \sum ^n_{i=1} (\beta_i X_i)\) 

where G is the probability of outcome variable equal to 1, i.e., choosing a green mode of transport for frequent travelling purposes, \(\beta_0\) is the model constant, X_i is a categorical explanatory variable, and \(\beta_i\) is the coefficient of regression which is estimated.

6. Results and Discussion

6.1. Descriptive Statistics

This study discusses the varying impacts created by different socio-demographic and socioeconomic factors on choice of type of mode of transport. With an increase in age, while men switch from green modes to non-green ones, the opposite is true for women. The only group for which the usage of green modes exceeds that of non-green ones are women aged 40-59 years. Women’s use of less-polluting modes is either equal or double men’s use of the same, irrespective of their age (Figure 1).

Figure 1: Age group-wise travel mode choice of men and women of Jaipur

Figure 2: Travel mode choice of men and women of Jaipur segregated by the level of education attainment

Figure 3: Travel mode choice of men and women of Jaipur segregated by their work status

Figure 2 shows that with an improvement in years of education attained, there is a shift from usage of green modes to non-green modes, irrespective of their gender. It is only the men with the least years of education attained who depend more on green modes and less on non-green ones. However, it is not only the women in the least educated group, but also those in the moderately educated group who depend more on the green ones. Within each education group, the proportion of women users of green modes is higher than men. It is the group of most educated people who are the highest users of the most polluting modes.

Figure 3 studies the usage of green and non-green modes segregated by people’s work status. Working individuals, both male and female, depend less on the green modes. However, working men’s dependency on polluting modes is much higher than that of working women. It is the men who use polluting modes more extensively than the women, irrespective of the work status.

The usage of green modes is found to decline with improvement in socioeconomic status of individuals. Within each socioeconomic group, men always outpace women in usage of less climate-friendly modes (Figure 4).

Figure 4: Travel mode choice of men and women of Jaipur segregated by their socioeconomic status 

Table 1: Brief summary of sample characteristics

Note: a INR is Indian Rupee, US$1~INR 71.285 as on February 2020.

Table 1 presents a summary of sample characteristics. Overall, men are more dependent on polluting modes of transport for frequent traveling purposes and women, on the other hand, use green modes predominantly. The sample comprises of almost equal proportions of men and women belonging to different age groups. The sample comprises of a large proportion of individuals who have attained education above class 12^th and a small proportion of individuals having attained education less than class 5^th. Approximately four- fifths of men are working and three-fifths of women are non-working in the sample. The table also displays the different proportions of population belonging to different socioeconomic groups based on an individual’s monthly per capita consumption expenditure. For missing data of consumption expenditure, data on house ownership and house type has been considered.

Table 2: Logistic regression results for impact of socio-demographic and socioeconomic factors on choice of green modes of transport by men and women

Note: *** p < 0.01; ** p < 0.05; * p < 0.1.

6.2. Logit Models of Choice of Green Modes of Transport

The descriptive statistics presented in the previous section depict existing relationship between travel mode choice and socio-demographic and socioeconomic variables. An explanatory model of choice probabilities may complement the exploratory analysis.

Table 2 displays the results of logit model of mode choice of men and women separately. This model attempts to analyze the impact that various socio-demographic and socioeconomic variables have on choice of green modes of transport by male and female.

The reference category for age in the model is the group of individuals belonging to the age group of 40-59 years. For men, the odds of choosing a green mode of transport by individuals belonging to the age group of 18-25 years is significantly higher than the individuals in the age group of 40-59 years. Men in the age group of 26-39 years have about 25% lower odds of opting for a green mode than men in the age group of 40-59 years. Therefore, it turns out that men in the youngest age group have the highest odds of choosing a green mode, followed by the eldest age group. The middle- aged men have the lowest odds of choosing a green mode among men in all other age groups. On the other hand, women in the youngest age group have approximately 68% lower odds of choosing a green mode than women in the eldest age group and women in the middle-aged group have 64% lower odds of choosing the same. This implies that in the case of women, the greenest behavior is displayed by the eldest age group, followed by the group of women in the middle-age group. The youngest group of women is found to have the lowest odds of choosing a green mode of transport. Women are found to become environment-sensitive with age.

The reference category for educational attainment is of individuals having studied above class 12^th. The results for this variable are also significant and uniform across gender. Those belonging to the least educated group have the highest odds of choosing a green mode as compared to the other groups. The most educated have the lowest odds of choosing the same. This implies that with increase in the years of schooling, the odds of choosing less-polluting modes goes on declining, hinting at the fact that simple increase in the formal years of schooling has nothing to do with being more considerate for the environment.

The results for work status are no different for male and female. It has a significant influence on choice of mode of transport. Working individuals, irrespective of their sex, have lower odds of choosing sustainable green modes of transport. The results of this variable help in possibly explaining why the middle-aged individuals and the most educated group were considerably more polluting. It is usually individuals in the age of 26-39 years and those having attained educational qualification above class 12^th, who are more likely to be working.

Socioeconomic status of individuals significantly impacts the decision of mode selection. The reference category for socioeconomic status in the model is the group of most privileged individuals, i.e., SEG3. In the case of men, SEG1 individuals have higher odds of choosing a green mode and SEG2 individuals have approximately 14% lower odds of choosing a green mode as compared to SEG3 individuals. Therefore, in the case of men, the least privileged have the highest odds and the individuals in the middle socioeconomic status group have the least odds of choosing a green mode of transport. However, in the case of women, as socioeconomic status is found to improve, the odds of choosing green modes of transport are found to decline, i.e., the most privileged are the most polluting and the least privileged are the least polluting.

Cost of travel, which is incorporated as a control variable in the study, does not significantly influence the dependent variable. As the cost of travel increases by one unit, the odds of choosing a green mode declines by 0.10% for men and 0.25% for women.

7. Conclusion

Using survey data for a developing country city, this study explores the gender differences in influence of socio- demographic factors on mode choice. With reference to the research issues listed at the beginning of this paper, the findings of the study assert the following: First, women’s mode choice are more environment-friendly than men, irrespective of the existing socio-demographic and socioeconomic differences. Second, middle-aged individuals have the highest odds of opting for a polluting mode of transport. Women’s use of less-polluting modes is either equal or double men’s use of the same, irrespective of their age. Third, increase in formal years of schooling negatively impacts the concern for environment, irrespective of the individual’s sex. Fourth, working individuals are more likely to choosing polluting modes of transport for frequent travelling purposes as compared to the non-working individuals. Fifth, in the case of men, the least privileged have the highest odds and the individuals in the middle socioeconomic status group have the least odds of choosing a green mode of transport. On the other hand, for women, the most privileged are the most polluting and the least privileged are the least polluting.

The results of the study draw attention to the varying damages created by different sections of population on the environment. The study highlights how the consideration of some or all of the factors listed in the study is crucial for effective policy recommendation regarding curtailing the environmental damages contributed by transportation sector. It highlights how choosing sustainable travel behavior is becoming rare in the city, hinting at questioning the reasons behind the same. Few possible explanations include the leisure-seeking, time-saving attitude of individuals and inadequate walking/cycling/public transport infrastructure in the city. The study, therefore, calls for a need for improvement of pedestrian and cycling lanes along with efficient public transportation system in the city.

Another aspect of understanding the results of the study call our attention to women’s display of greener travel behavior as compared to the men. This means that, to maintain the ecological footprints by women, public transportation services and pedestrian lanes in the city should be improved in such a way to keep in mind their demands of safety and security. This will facilitate smooth access of educational, health, financial, etc., services by them and not stand in way of their empowerment.