Analyzing Women’s Security in Public Transportation in Developing Countries: A Case Study of Lahore City

Abstract

Security concerns regarding women in developing nations are frequently highlighted due to the prevalence of harassment incidents, particularly within public transportation systems. In Pakistan, where women make up half of the population, this issue persists in various forms of harassment, both within local environments and public transportation systems. Therefore, this study aims to investigate the security challenges confronted by women within the public transportation system in the city of Lahore, Pssakistan. To achieve this, a user perception survey was designed to focus on women’s security during travel and relevant socioeconomic factors. The collected responses were analyzed using descriptive analysis and factor analysis methods. Exploratory factor analysis (EFA) revealed five latent variables, each encapsulating distinct aspects of women’s security within public transportation environments. Later on, a structural model of comfort of using public transportation at night was developed using the results of the exploratory factor analysis. Our study’s results propose that although many women express feeling safe during their travels, a prominent number have experienced instances of harassment. Generally, issues such as insufficient lighting during night travel and a lack of awareness about harassment come out as primary concerns within Lahore’s currently operated public transport. The structural model results revealed that the latent variables of harassment, harassment reaction, bus stop station facility, and public transportation safety are significant predictors of comfort of using public transportation at night, being statistically significant (p < 0.05). The findings emphasize the initiatives to reduce overcrowding, improve nighttime lighting and infrastructure, and strengthen awareness among users, along with prevention measures against harassment. This approach assures the females’ physical security and enhances the overall well-being and empowerment of women in urban surroundings.

1. Introduction

Gender-based violence and harassment in public transportation represent a pervasive urban safety challenge, particularly in developing countries where legal protections, enforcement mechanisms, and urban design interventions remain inadequate. Safe and reliable mobility is not only a matter of individual freedom but also a prerequisite for women’s economic empowerment, social participation, and overall well-being. When public transit systems fail to provide a secure environment, women’s ability to access education, employment, and civic life is severely restricted—undermining both individual potential and broader socio-economic development.

This challenge is acutely visible in Lahore, Pakistan’s second-largest metropolitan city and a major hub for commerce, education, and culture. Despite recent investments in urban transit infrastructure, including the Lahore Metrobus and Orange Line Metro Train, women continue to face harassment, physical violence, and “eve-teasing” on a routine basis. Surveys highlight the scale of the issue: research indicates that 85% of women in Lahore have experienced harassment while using public transport [1], while a security audit by UN Women and the Aurat Foundation found 90% of female respondents reporting feelings of insecurity during transit [2]. These experiences encompass verbal, physical, and non-verbal harassment, often occurring in crowded vehicles, poorly lit stations, or during first- and last-mile travel.

The consequences of such insecurity extend beyond immediate safety concerns. As feminist urban theory and transport equity frameworks suggest, mobility is deeply gendered, with access shaped by socio-cultural norms, patriarchal power structures, and unequal spatial planning [3,4]. In Pakistan, entrenched gender stereotypes and weak legal enforcement create an environment where women’s autonomy is systematically constrained. This aligns with global findings that violence and harassment in public spaces have multi-dimensional impacts—affecting physical safety, mental health, freedom of movement, and socio-economic participation [5,6]. Addressing this issue requires understanding the interplay between built environment factors, cultural norms, transit operations, and policy frameworks [7,8]. However, in Lahore, there remains a gap in empirical research that integrates these dimensions to inform gender-sensitive transportation policy.

Despite recent urban transit developments, Lahore’s public transportation system remains unsafe for women, with alarmingly high rates of harassment undermining their freedom of movement, economic participation, and well-being. The absence of gender-sensitive infrastructure, entrenched socio-cultural barriers, and weak policy enforcement collectively perpetuate a mobility gap, yet there is a lack of integrated, evidence-based research to address these intersecting challenges in the city’s transit environment.

This study holds significance at both the academic and policy levels. From an academic perspective, it addresses a critical research gap by integrating infrastructural, socio-cultural, and policy-related dimensions into a single framework for understanding women’s security in public transportation. This multidisciplinary approach—drawing from urban planning, gender studies, and transport policy—advances the theoretical understanding of how gendered mobility is shaped in developing-world cities. From a policy and practice perspective, the research generates context-specific evidence for Lahore, offering actionable recommendations to transit authorities, urban planners, and policymakers. By identifying the structural and cultural barriers that compromise women’s safety, this study provides a foundation for targeted interventions—such as gender-sensitive transit design, enhanced security protocols, and public awareness campaigns—that can transform public transportation into a safe, inclusive, and equitable urban mobility system.

The research aims to assess the factors impacting women’s security in Lahore’s public transit system, analyzing infrastructural, socio-cultural, and policy-related dimensions. It is guided by three core questions: What are the primary infrastructural and operational characteristics of Lahore’s public transit that influence women’s perceived and actual security? How do socio-cultural norms and gender dynamics shape women’s experiences of mobility and harassment in the city? What policy measures and urban design interventions can enhance women’s safety and inclusivity in public transport? By situating the issue within the broader discourse on gender, mobility, and urban safety, this study seeks to contribute both context-specific insights for Lahore and theoretical implications for gender-sensitive transport planning in similar urban contexts.

2. Literature Review

Women’s mobility and access to public spaces are significantly constrained by gender discrimination, a problem that persists globally. According to Advancing Learning and Innovation on Gender Norms (ALIGN), gender-based violence in public spaces—including harassment and physical assault—is widespread, with studies showing that a high proportion of women experience violence in urban areas [9]. This persistent risk not only restricts women’s freedom but also limits their ability to participate fully in public life. Across contexts, research consistently identifies a combination of socio-cultural, infrastructural, and institutional factors as drivers of insecurity during commutes. Patriarchal social structures and male-dominated environments perpetuate gender-based harassment [10], while infrastructural shortcomings—such as poor lighting, overcrowding, and insufficient surveillance—amplify vulnerability [11,12]. Institutional weaknesses, including inadequate legislation, lax enforcement, and limited victim support services, further exacerbate the issue [2]. Kacharo et al. (2022) point to six key predictors of women’s exposure to violence in public transport: age, marital status, type of transport, travel time, available facilities, and system management [13].

A review of global studies confirms that gender-based violence in public transport is particularly acute in developing countries, where legal, social, and infrastructural safeguards are limited. United Nations (2023) data indicate that 70% of women worldwide have encountered sexual harassment in public spaces, including transit systems [2]. The World Bank (2022) emphasizes that the provision of safe public transportation is essential for women’s empowerment and social inclusion [14]. Empirical studies from diverse contexts—such as India, where one-third of young women experience harassment on urban public transport [15], and Hawassa, Ethiopia, where over half of women report multiple forms of violence [13]—illustrate the universality of the problem. In Pakistan, both quantitative surveys and NGO reports show similar patterns: 85% of women in Lahore report harassment on public transport [1], and 73% nationwide experience harassment in public spaces [16]. These statistics reveal not only the scale of the issue but also the pervasiveness of fear among female commuters, which directly limits their mobility and independence [6].

While much of this literature converges on the conclusion that safety concerns reduce women’s transit use, the factors shaping these concerns are context-dependent. In Bangladesh, for example, barriers include poorly lit bus stops and inadequate security measures [6,17,18,19], whereas in other contexts, overcrowding has been identified as the single most critical factor, with the highest negative effect size (t = −12.74, p < 0.01) in studies of harassment risk [20]. Importantly, comparative research shows a consistent gender gap in perceived safety, with women reporting greater insecurity than men in identical environments [7,21]. This disparity is not limited to high-crime contexts; even in relatively safer environments, women’s travel choices are shaped by fear, prompting route changes, avoidance of night travel, and reliance on costlier private modes [22,23].

Proposed interventions in the literature range from short-term operational measures, such as gender-segregated seating and the deployment of security personnel, to longer-term strategies including CCTV installation, infrastructural redesign, and staff training [24]. However, there is limited evidence on the sustained effectiveness of these measures, particularly in developing countries. For example, while the Punjab Safe Cities Authority’s Women Safety App represents an important technological initiative, there is insufficient empirical evaluation of its capacity to reduce harassment in Lahore’s transit environments [25].

Emerging frameworks, such as Ding et al.’s (2020) “whole journey” approach, emphasize the importance of addressing women’s safety across all trip stages, from first-mile access to last-mile connections [26]. Similarly, global studies call for urban transport planning to adopt a gender-sensitive lens, ensuring that infrastructure design, policy, and service delivery integrate women’s perspectives [19,27]. In Chile, for instance, high insecurity levels among women and older commuters highlight the need for context-specific design interventions [28]. Reports such as ALIGN’s stress that gender-based violence is a structural problem requiring systemic policy reform rather than isolated interventions [9,29]. Innovative approaches, including ANN-based risk prediction models, have been proposed to integrate safety considerations into operational decision-making [30].

In Lahore, however, public transport infrastructure remains largely gender-neutral in design, lacking features such as secure, well-lit waiting areas and clearly demarcated women’s spaces. Evidence from local studies shows that safety is the strongest determinant of women’s satisfaction with public transport, as demonstrated by ANOVA (F = 159.72, p < 0.001) and regression analysis (β = 0.7104, p < 0.001) [31]. Yet, current policy frameworks rarely include dedicated provisions for women’s security, and broader societal attitudes continue to limit women’s mobility despite technological and infrastructural improvements [4].

The Crime Prevention Through Environmental Design (CPTED) framework offers a promising lens for addressing these challenges, advocating strategies such as improved lighting, unobstructed sightlines, clear territorial demarcations, and controlled access points [32,33]. Evidence shows that women’s perceptions of safety are closely linked to these environmental features, and their absence exacerbates fear and restricts mobility [34]. While CPTED principles have been applied in various global contexts, their systematic application to Lahore’s public transport system remains limited, leaving a gap in understanding how environmental design can be tailored to reduce harassment risk in this specific cultural and infrastructural setting.

Overall, the literature points to three key insights: first, gender-based harassment in public transport is a pervasive global problem with locally specific drivers; second, while interventions exist, their long-term effectiveness and cultural adaptability remain under-researched; and third, in Lahore, empirical evidence linking socio-cultural norms, infrastructural design, and women’s perceived and actual safety is scarce. This gap underscores the need for integrated, context-sensitive research that not only documents women’s experiences but also identifies actionable strategies to enhance safety in public transport, thereby contributing to more equitable and inclusive urban mobility systems.

3. Research Methods

3.1. Selection of Study Area and Its Characteristics

Lahore, the capital of Pakistan’s Punjab province, is the country’s second-largest metropolitan area, with a population exceeding 11 million people. Its expanding urban population relies heavily on diverse modes of public transport, including buses, rickshaws, the Orange Line Metro Rail Transit (OLMRT), and the Metro Bus System (BRTS) [35]. While these systems have improved mobility options, persistent safety concerns—especially for women—limit their accessibility and inclusiveness. Reports of verbal, physical, and non-verbal harassment remain high [2], making women’s security a pressing research and policy issue.

Lahore’s transportation network, though extensive, has design and operational gaps that disproportionately affect female commuters. Overcrowding, insufficient lighting at stops, and inadequate security infrastructure have been identified as key risk factors [36]. Furthermore, deep-rooted socio-cultural norms in Pakistan, particularly in large cities such as Lahore, reinforce gendered mobility constraints and increase vulnerability in public spaces [37].

While recent initiatives, such as the Punjab Safe Cities Authority’s surveillance systems and the Women’s Safety App, represent progress, studies indicate that these measures are insufficient in isolation [38]. Addressing the issue requires a holistic approach that integrates infrastructure improvements, socio-cultural awareness, and policy reforms. Lahore’s scale, diversity of transport modes, and gender-specific safety challenges make it a representative and critical case study for understanding and improving women’s experiences in South Asian urban transport systems [14].

3.2. Questionnaire Design and Survey Instrument

A structured questionnaire was developed to systematically capture women’s experiences with public transport in Lahore, focusing on both safety perceptions and actual incidents of harassment. The survey instrument was divided into five sections—user general information, safety about public transport, safety while using public transport, harassment while using public transport, and reaction to harassment—structured around six key analytical dimensions: socio-demographic and travel profile, system-level safety perception, trip-stage safety perception, harassment exposure and typology, coping and reporting behavior, and contextual and environmental factors. Variables were derived from a review of relevant literature and empirical studies to ensure contextual relevance [3,22], while also considering Lahore’s local socio-economic, cultural, and infrastructural conditions during design.

The socio-demographic and travel profile section recorded baseline variables such as age, marital status, income, education, profession, travel mode, trip frequency, and time of use [4], enabling the analysis of how personal attributes influence mobility and vulnerability. System-level safety perception questions assessed respondents’ overall sense of security while using public transport in Lahore [5], whereas trip-stage safety perception items examined security at specific journey phases—walking to stops, waiting, riding, and transferring—at different times of day [6].

Harassment exposure and typology items identified perceived acts of harassment, including verbal abuse, physical contact, unwanted staring, and misuse of digital devices [5]. Coping and reporting behavior questions explored post-incident actions, such as confronting perpetrators, notifying authorities, or altering travel patterns [6]. Contextual and environmental factors (e.g., crowding, lighting, presence of CCTV/security staff, and transit stop design) were embedded in multiple sections to assess situational influences [8].

To ensure robustness and minimize measurement error, multiple steps were taken. A Likert scale from “strongly disagree” (1) to “strongly agree” (5) was used for attitudinal items [4], allowing for standardized and comparable responses. Prior to main data collection, a pilot survey was conducted to refine question clarity, eliminate ambiguities, and verify cultural appropriateness. Surveyors—comprising trained undergraduate transportation engineering students—were instructed on interview protocols, ethical considerations, and non-leading questioning to reduce interviewer bias.

The sampling strategy targeted frequent public transport users by conducting surveys at high-traffic transit points along the OLMRT corridor (Orange Line), the BRTS route (Green Line), and key feeder routes (Figure 1). Onboard surveys were also conducted to capture real-time, in-journey perspectives. Data collection took place in November–December 2024, covering peak and off-peak periods on weekdays and weekends to ensure temporal representativeness. Using Cochran’s formula, the minimum sample size was calculated as 384; however, to improve statistical power and reflect near real-time conditions, the sample was increased to 600 respondents.

To enhance replicability and transparency, the methodological process—from site selection and instrument design to pilot testing, main survey administration, and data validation—is presented in Section 4.2. This mapping of sequential stages and decision points ensures clarity for future replication and provides an auditable pathway from raw data collection to analytical readiness.

4. Data Analysis and Results

4.1. Description of Sample Statistics

The research data provide an insightful look into various demographic and socio-economic characteristics of women who use public transportation and their experiences regarding security.

Table 1. Socio-economic Factors.

Variable	Description	Distribution (%)
Age	Less than 18	11
	18–25	41
	25–50	40
	Greater than 50	8
Education	Uneducated	6
	Under matric	18
	Matric	26
	Graduate/Postgraduate	50
Marital Status	Married	42
	Unmarried	23
	Engaged	16
	Preferred not to tell	19
Married Years	Less than 1	4
	1–5	31
	5–10	7
	Greater than 10	10
	Preferred not to tell	48
Number of Children	Not applicable	55
	None	4
	1	8
	2	15
	Greater than 2	18
Employment	Work from home	8
	Job/Business	20
	Student	39
	Housewife	33
Income Level (PKR)	No income	58
	Less than 30,000	25
	30,000–70,000	16
	Greater than 70,000	1

shows that the majority of respondents are young, with 41% between 18 and 25 years and 40% in the 25–50 years age range. This suggests that public transportation is heavily utilized by younger women, likely due to their higher mobility needs related to education, employment, or other activities. A smaller percentage (11%) is under 18 years old, and only 8% are over 50 years old, indicating a lesser dependency or preference for public transport among these age groups. Education levels are skewed toward higher qualifications, with 50% of respondents having a graduate or postgraduate degree, while 26% have completed matriculation. A smaller portion, 18%, are under matric, and 6% are uneducated. The high educational attainment among respondents could imply that educated women are more inclined to use public transportation, possibly due to job or study requirements; 42% of respondents are married, and 23% are unmarried. 16% are engaged, and 19% prefer not to disclose their marital status. This distribution suggests that married women are still a significant group among public transportation users, but a notable segment chose not to disclose their marital status, which may reflect privacy concerns. Among those married, 31% have been married for 1–5 years, with 10% married for over 10 years. This reflects a diversity of experience and potentially different security concerns based on family commitments. 55% indicated “Not applicable,” which could correlate with the younger, unmarried group or those without children. Furthermore, among married or engaged respondents, 15% have two children, and 18% have more than two, which may influence the frequency and timing of their public transport usage. Employment status shows 39% as students, the largest segment, followed by 33% as housewives. 20% are in jobs or businesses, and 8% work from home. This suggests that students and homemakers are significant users of public transport, possibly due to limited access to private vehicles or flexible travel schedules. Income levels are generally low, with 58% reporting no income and 25% earning less than 30,000. Only 1% have an income greater than 70,000, indicating that most public transport users fall within lower income brackets, potentially affecting their ability to afford alternative transportation options.

Figure 2 shows that public transport is the primary mode of commuting, with 72% relying on it, while 25% use a combination of public and private transport. Only 3% use solely private transport. This high reliance on public transportation highlights the need for enhanced security measures for women in these settings. The primary modes of public transport include the Speedo/BRT (51%) and the Orange Train (43%). Qing qi usage is lower, at 6%. This breakdown may reflect the perceived security or accessibility of these services. Moreover, regarding the frequency of using public transport, 52% of respondents use public transport daily, with 35% using it sometimes and 7% rarely. This regular usage underlines the critical role of public transportation in these women’s lives and the importance of consistent security measures. The majority travel in the morning (64%), likely related to commuting for work or study. Noon and evening travel account for 22% and 13%, respectively, while 1% travel late at night. This distribution underscores the need for security measures, particularly in the morning when most women use public transportation.

Table 2. Correlation of socioeconomic and transport-related variables.

Characteristics	Age	Education	Marital Status	No. of Children	Employment	Income	Mode of Transport	Frequency of Use	Time of Use
Age	1
Education	−0.72	1
Marital Status	−0.457 **	0.016	1
No. of Children	0.693 **	−0.080 *	0.761 **	1
Employment	0.249 **	−0.127 **	−0.401 **	0.269 **	1
Income	−0.296 **	0.016	0.402 **	−0.282 **	0.229 **	1
Mode of Transport	0.026	−0.062	0.248 **	−0.282 **	−0.247 **	0.145 **	1
Frequency of Use	0.239 **	−0.357 **	−0.109 **	−0.012	0.172 **	−0.117	0.018	1
Time of Use	0.129 **	−0.224 **	−0.141 **	0.199 **	0.043	−0.153 **	−0.066	0.354 **	1

Note: ** Correlation is significant at the 0.01 level (2-tailed), * Correlation is significant at the 0.05 level (2-tailed).

presents Pearson correlation coefficients among various demographic and travel-related variables. Significant positive correlations were observed between age and number of children (r = 0.693 **), and between marital status and number of children (r = 0.761 **), indicating logical associations in family structure. Income level shows a significant negative correlation with age (r = –0.296 **) and number of children (r = –0.282 **), possibly reflecting economic disparities. Frequency of public transport use is negatively correlated with education (r = –0.357 **) and income (r = –0.117 *), suggesting women with higher socioeconomic status use public transport less frequently. Meanwhile, time of use is positively correlated with frequency of use (r = 0.354 **) and age (r = 0.129 **), implying older and more frequent users tend to travel during specific hours. Most variables are statistically significant at the 0.01 level, reflecting robust associations that can inform interventions for improving women’s transport security in Lahore.

The survey findings highlight public perceptions and concerns regarding security, crime, and harassment in public transport (Figure 3). A strong majority (72%) believe that public transport is generally safe during the day, but opinions shift at night, with 49% viewing it as unsafe and 41% remaining neutral, indicating increased concern and uncertainty after dark. Additionally, 52% of respondents agree that overcrowded transport increases the risk of crime, while 34% disagree and 11% remain neutral. When asked about gender-segregated transport, responses were mixed: 49% support the idea, but a significant 37% oppose it, reflecting differing views on how to improve security. Regarding harassment, only 10% acknowledged experiencing it, while 65% denied it, suggesting a reluctance to speak about such issues in a society like Pakistan. Moreover, 47% gave a neutral response when asked about reacting to harassment, with only 10% agreeing they respond and 43% disagreeing, further highlighting social and cultural barriers that may discourage open discussion or action.

The data present a demographic where young, educated, lower-income women, primarily students or homemakers, frequently use public transportation. Security concerns are particularly relevant during peak morning hours and at night as well, and for services such as the Orange Train and Speedo/BRT. Furthermore, data show that crimes are more likely to occur in overcrowded transport, but still, women are reluctant to report these crimes. Enhanced security measures could significantly impact this group, as public transportation is an essential and frequently used service among these women.

4.2. Exploratory Factor Analysis (EFA)

The flowchart describing the process of handling collected data for factor analysis is presented in Figure 4. The flowchart outlines the key steps from raw data to factor analysis and result interpretation.

Several variables were selected for analysis and classified into two categories: endogenous (latent) and exogenous (observed) variables. Six endogenous latent variables were identified alongside thirty-six exogenous variables, which were included in the analysis to capture women’s experiences regarding their security on public transportation in Lahore. These variables were considered for factor analysis using SPSS 19 software. To enhance transparency in the EFA, this study used Principal Component Analysis (PCA) for factor extraction, with six factors retained based on Kaiser’s criterion (eigenvalues > 1) and confirmation from the scree plot. Varimax rotation, an orthogonal method, was applied to simplify interpretation by maximizing variance across factor loadings, ensuring that observed variables loaded distinctly onto specific factors. This approach helped identify clear, meaningful constructs such as harassment, reaction, and infrastructure-related concerns. All retained factors demonstrated strong internal consistency, with Cronbach’s alpha values ranging from 0.703 to 0.885, indicating reliable groupings of observed variables and a robust factor structure.

Cronbach’s alpha values were calculated for the five extracted factors (

Table 3. Latent Variables with Associated Observed Variables and Cronbach’s Alpha.

Observed Variables	Mean	Latent Variables with Factor Loading
		Harassment	Harassment Reaction	Safe Public Transportation	Bus Stop Facility	Comfort of Using Public Transport at Night
I face harassment often.	1.60	0.780
I have faced visual harassment.	1.62	0.764
I have experienced violent physical attacks.	1.50	0.762
I suffered from physical harassment.	1.62	0.757
I faced verbal harassment.	1.59	0.755
I carry a security weapon with me.	1.59	0.750
I have been stalked.	1.52	0.720
I have been harassed on public transport.	1.68	0.679
Harassment stops me from using public transport.	1.66	0.654
I stopped using that stop where I got harassed.	1.79	0.567
I have to wait longer at the bus stop.	2.59	0.501
I usually don’t speak against it.	2.67		0.835
I always register complaints to security at the stop.	2.77		0.852
It does not offend me anymore.	3.26		0.833
I wear a mask to hide myself while traveling.	3.66		0.803
I speak out loud in public.	2.53		0.773
I gave a verbal reaction to the harassment.	2.54		0.761
I confront the harasser when he humiliates me.	2.60		0.503
I feel safe walking alone during daylight.	4.05			0.647
Few seats should be reserved for ladies.	4.21			0.621
The bus environment is comfortable.	4.07			0.616
Staff behavior with people is good.	4.11			0.560
There is proper security at the station.	4.44			0.555
The bus is punctual.	4.24			0.536
I don’t worry about using public transport.	4.47			0.509
Crimes against women are less in overloaded buses.	3.55				0.724
Bus stops are safe for a young girl.	4.03				0.635
Bus stops are safe for women.	4.07				0.623
Transportation infrastructure is functional.	4.25				0.514
I am restricted from travelling alone from home.	3.40					0.833
I feel comfortable using public transport at night.	3.50					0.643
I feel Safe walking alone at night.	3.10					0.565
I feel safe using public transport.	4.07					0.537
The streetlight is working properly.	4.13					0.527
Cronbach’s Alpha		0.885	0.871	0.825	0.802	0.703
Percentage variance explained		20.50	25.86	16.61	19.07	17.95

). This value evaluates the reliability and consistency of variables [39]. The larger the alpha coefficient, the greater the chances of variables being interconnected.

Nunnally and Bernstein (1994) suggested 0.70 as an acceptable reliability coefficient. The test values were 0.885, 0.871, 0.825, 0.802, and 0.703 for the five latent variables, respectively [40]. These values indicated that there was a significant consistency among participants in evaluating the observed variables. These variables were consistent in estimating their impact on women’s security within public transportation in the city of Lahore.

Table 3 also shows the latent variables with the associated observed variables. The first latent variable, Harassment, encompasses eleven observed variables, such as frequent harassment, visual and verbal harassment, physical attack, presence of security weapons, stalking, stopping using public transportation, and changing the bus stop, and waiting time at the bus stop. Research indicates that extended waiting times for public transportation are associated with an increased risk of harassment for passengers. This research supports the findings of Abeona et al. (2018) that prolonged exposure at bus stops or transit areas often leaves individuals, especially women, vulnerable to harassment [41]. Similarly, Ceccat & Loukaitou-Sideris’s (2022) work indicates that the frequent harassment on public transport has led to behavioral changes, with many women opting to stop using these services after experiencing incidents [42]. The result shows that various types of harassment have been reported, including physical assault, visual harassment, verbal abuse, and, in some cases, stalking, which supports the findings of Kabir & Islam (2023) that such situations create a hostile environment for female passengers [43]. The constant risk of harassment has driven some respondents to carry self-defense tools for personal security while traveling, a measure that underscores the level of insecurity faced by these individuals [44]. This aligns with findings that suggest a correlation between the perception of public transport as unsafe and the decreased likelihood of usage, particularly among women [45].

The second latent variable, Reaction, is associated with seven observed variables including lack of reaction, complaint filing, confrontation, and a few more along with the use of face masks as a protective measure, which allows them to conceal their identity, thus minimizing the likelihood of unwanted recognition which supports the finding of Duri at al. (2024) whose study claims the face mask as a security tool against harassment at public transportation [46]. This latent variable captures varied responses to harassment, where some women choose to confront or vocally respond to the harasser, making their distress visible to others in hopes of receiving assistance. Vocal reaction techniques are effective in mitigating harassment incidents by drawing public attention [47]. Additionally, some women opt to formally report incidents of harassment to the transportation authorities, an action associated with increased perceptions of justice and deterrence [48]. However, not all women respond actively; some may choose to ignore the harassment, which is common in societies like Pakistan. It is a behavior that, Ceccato & Loukaitou-Sideris (2021) suggest, is often a coping mechanism when the likelihood of successful involvement is perceived as low [49].

The third latent variable, Safe Public Transport, is tied to seven observed variables addressing perceptions of security while walking during the day, seat reservation, environmental comfort, staff behavior, station security, punctuality, and overall comfort in public transport. This variable reflects women’s overall perceptions of public transportation services. Research shows that perceptions of security are crucial in determining whether women feel comfortable using public transportation [50]. The result shows that some women mentioned feeling safe and secure while traveling, especially when the station and bus environment met acceptable security standards. Efficient security measures and punctuality improve the optimistic perception of public transport for most women, supporting the studies that highlight the value of reliable and secure transport systems that increase their usage [51]. Moreover, the availability of seating and respectful staff interactions contributes to the positive perception of public transport, which is consistent with studies mentioning that supportive infrastructure and humble staff behavior positively affect women’s comfort in transit systems [42].

The fourth latent variable, Bus Stop Facility/Infrastructure, includes variables, i.e., security at bus stops for women and girls, security at crowded stops, and infrastructure quality. The variable contributes to understanding women’s sense of security within certain facilities of public transportation. It is reported that the women who perceive bus stops as safe are more likely to feel safe while commuting on public transport buses than those who feel unsafe at stops [52]. Many studies show the importance of safe, well-designed public transport infrastructure in reducing crimes and improving women’s perception of security [53]. For example, crowded bus stops with operational infrastructure are perceived to be safer, possibly discouraging criminal activity through heightened visibility and a stronger social presence. The findings show that one of the essential concerns is assuring the security of young girls as well as women at bus stops. Well-designed infrastructure, such as well-kept shelters, proper and sufficient lighting, and surveillance, can create an environment in which women feel safer and less vulnerable to harassment or assault. This supports the findings by Badiora et al. (2015), who underline that those structural attributes play an essential role in minimizing crime rates against women [54].

The fifth latent variable, “Comfort using public transport at night,” consists of four observed variables: family restrictions on night travel, comfort zone, security in the surroundings (walking), security within public transport, and adequacy of street lighting. Concerns regarding public transport usage at night often relate to security, as many users find public transport secure if they feel comfortable using it at any time. Moreover, the variable reflects the perceptions of security and comfort that affect women’s decision to use public transportation at night. Many researchers indicate that perceptions of security are not restricted to the transport system itself but also include the surrounding infrastructure, i.e., the walkways leading to public transport facilities, where proper lighting and security measures are essential [55]. The findings show that public transport systems prioritizing these elements can increase user satisfaction and promote greater accessibility [56]. A common issue involves family-imposed limitations on women traveling alone, especially at night, due to security concerns. This is supported by Gardner and Cui et al. (2022), who highlight that those perceptions of being unsafe and insecure in public transport environments can lead to limited mobility for women, as family members may impose limitations on their nighttime outings [57]. For some women, these restrictions originate after personal experiences or knowledge of security risks, reflecting a protective reaction by families who are concerned about harassment. Conversely, other women feel comfortable and safe while walking alone at night when proper lighting and secure infrastructure are present. This finding supports the work of Park & Garcia (2020), who found that features such as bright street lighting and visible security measures improve women’s comfort in public spaces and increase their likelihood of using public transport services [58].

4.3. Structural Equation Modeling

Wang et al. [59] note that a researcher is not constrained to a single modeling approach but can select the most appropriate method at each stage of analysis. In this study, Structural Equation Modeling (SEM) was employed to extend the exploratory factor analysis results and test the hypothesized relationships between key latent variables influencing women’s security in public transportation. The choice of SEM was guided by its ability to model complex interdependencies between observed and latent variables, account for measurement error, and provide a theoretically grounded assessment of causal pathways.

The structural model (Figure 5) was developed to examine the comfort of using public transportation at night, integrating the five extracted latent variables: harassment, harassment reaction, bus stop facilities, safe public transportation, and comfort of using public transportation at night. In addition, socio-economic variables such as frequency of public transport use, type of mode, and respondent demographics were tested but did not yield significant correlations with the model’s core constructs.

Operationalization of the latent variables was based on factor loadings from the earlier exploratory analysis. For example, harassment encompassed indicators such as frequency of harassment incidents and perceived vulnerability; harassment reaction captured respondents’ reported strategies for confronting or avoiding harassment; bus stop facilities reflected perceptions of lighting, cleanliness, and security measures; safe public transportation measured overall perceptions of safety during travel; and comfort at night reflected self-reported ease and willingness to use public transport after dark.

The model was tested against four hypotheses:

• Bus stop facilities influence both harassment and harassment reaction.
• Harassment affects the harassment reaction.
• Perceptions of safe public transportation are influenced by harassment, harassment reaction, and bus stop facilities.
• Comfort at night is influenced by safe public transportation, harassment, harassment reaction, and bus stop facilities.

Model results confirmed all hypotheses, with statistically significant paths (p < 0.05). Figure 5 illustrates these relationships alongside the corresponding path coefficients and fit indices. The negative relationship between bus stop facilities and harassment (β = –0.69, p < 0.001) suggests that even respondents with positive views of stop facilities still report experiencing harassment—indicating that infrastructure alone cannot eliminate harassment risks. Conversely, the positive relationship between bus stop facilities and harassment reaction (β = 0.68, p < 0.001) implies that respondents who perceive bus stops as secure are better prepared to respond effectively to harassment through active measures such as confronting the harasser, lodging complaints, or employing avoidance strategies.

The positive association between harassment and harassment reaction (β = 0.40, p < 0.001) indicates that exposure to harassment may heighten women’s readiness to respond, though this does not necessarily translate into higher perceptions of overall safety. In fact, both harassment (β = –0.40, p < 0.001) and low harassment reaction (β = –0.11, p = 0.016) are negatively associated with safe public transportation, underscoring that perceived safety declines when harassment risk is high or when women feel less capable of responding.

Bus stop facilities exhibit a positive link with safe public transportation (β = 0.46, p < 0.001), reinforcing the role of well-lit, secure, and clean waiting areas in enhancing perceived safety. Harassment also directly undermines comfort at night (β = –0.30, p = 0.016), while insufficiently secure bus stops further reduce night-time comfort (β = –0.21, p = 0.024). In contrast, harassment reaction (β = 0.89, p < 0.001) and safe public transportation (β = 0.62, p < 0.001) both strongly and positively influence comfort at night, highlighting the combined importance of personal agency and systemic safety measures.

From a theoretical standpoint, these findings align with socio-ecological models of safety, where individual behaviors, environmental features, and institutional safeguards interact to shape security outcomes. The operationalization of the variables in SEM allowed for a nuanced understanding of how infrastructural and behavioral factors jointly affect women’s comfort in nighttime travel.

However, it is important to acknowledge limitations. SEM fit indices reported in Figure 5 indicate acceptable reliability [60], yet the model remains sensitive to measurement error, particularly in self-reported variables prone to recall bias or under-reporting in sensitive contexts. Additionally, the cross-sectional design limits causal inference, and the error margins on certain paths, while within acceptable thresholds, warrant cautious interpretation when generalizing beyond Lahore.

Overall, Figure 5 does not merely represent statistical relationships but provides a conceptual map of how infrastructural, experiential, and behavioral factors converge to influence women’s mobility at night. The results underscore that strategies to improve security must integrate both physical interventions (e.g., bus stop upgrades) and social measures (e.g., harassment response training), with the understanding that infrastructure improvements alone are insufficient to eradicate harassment risks.

5. Policy Implications

The quality of the transport environment, including cleanliness, seating availability, and crowd management, was found to significantly influence women’s comfort and willingness to use public transport. To address these concerns effectively within the context of limited urban resources in Pakistan, a series of specific and practical interventions is proposed. Other studies have emphasized that the improvement in bus stop infrastructure has a significant impact on enhancing women’s safety and hence their mobility [33,61]. Improving bus stop infrastructure and the surrounding environment, including the provision of shelters, lighting, seats, etc. These measures can be implemented at a low cost. Security at night can be improved by installing solar-powered lighting at bus stops while minimizing electricity costs through the use of solar energy. The provision of basic sheltered seating with open visibility and integrating designated pedestrian paths or marked crossings can alleviate comfort and access without requiring major reconstruction. Routine cleanliness and maintenance can be managed by engaging local contractors or community organizations to ensure hygiene standards at stops and inside vehicles.

Crowd management and scheduling strategies can also make a significant difference. Increasing the frequency of buses during peak hours, introducing short shuttle services on high-demand routes, and adjusting service timings to align with the travel patterns of working women and students can improve the travel experience. At transit points, establishing women-only queues or boarding zones may help reduce discomfort during boarding, particularly during rush hours.

Security improvements should be made in a cost-sensitive and scalable manner. Deploying trained female personnel at high-traffic stations during peak periods can increase the perceived security of female passengers [62]. Where full-time security staff is not feasible, low-cost surveillance cameras can be installed at critical locations, paired with clear signage to serve as a deterrent.

Addressing societal attitudes is also crucial, as harassment hinders female mobility in the public transport system [4,63]. Education and awareness campaigns should be integrated into existing community programs and schools, as these soft measures can be useful policy interventions for behavioral modification to control the harassment and improve the sense of security for females [61,64]. The medium can include awareness posters, public announcements, and workshops to promote respectful behavior and increase awareness about harassment. Transport staff and drivers can receive basic education and training in gender sensitivity and conflict de-escalation. Longer-term change can be supported by including gender respect content in school curricula and public service messaging. Further, community engagement and proper legislation at the local level can play a significant part in controlling the violence and harassment against women [63,64]. The community-level activities, support, and awareness campaigns are useful policy interventions to address women’s issues in public places and services. These soft policies should be followed by a proper law and enforcement framework, and their implementation by the concerned authorities.

Together, these measures offer practical, scalable ways to improve women’s security and comfort in public transport systems. They can be implemented progressively, aligned with budgetary limitations and existing infrastructure, to create more inclusive and accessible mobility options. These findings serve as a foundation for gender-sensitive and responsive public transport planning and provide direction for policy and research in urban contexts similar to Pakistan.

The findings of this study carry significant policy implications for improving women’s security in public transportation across urban centers like Lahore. Policymakers should prioritize the integration of gender-sensitive planning into transport policies by enhancing infrastructure at bus stops and within transit systems, such as installing adequate lighting, surveillance cameras, and secure waiting areas to address security concerns, especially at night. Additionally, regular training programs for transport staff on respectful behavior and gender awareness, along with clear protocols for reporting and addressing harassment, are essential. Public awareness campaigns aimed at shifting societal attitudes and encouraging victims to report incidents can further empower women. Policies should also promote the inclusion of women in transport planning processes to ensure their needs are directly reflected in future interventions. By incorporating these measures, governments can create safer, more accessible public transport systems that encourage female participation in education, employment, and public life, contributing to broader goals of gender equity and urban development.

6. Conclusions

This study set out to address a well-documented gap in the literature: while women’s security in public transportation has been explored extensively at a regional scale in South Asia, there remains a shortage of city-specific, empirically grounded analyses that connect socio-economic characteristics, travel behaviors, and perceived safety. By focusing on Lahore, Pakistan, this research moves beyond generalized narratives and provides a fine-grained, data-driven understanding of the contextual factors shaping women’s mobility experiences.

The results confirm and extend findings from prior studies that harassment is the most critical determinant of women’s perceptions of public transport safety, but this study adds analytical depth by showing how this concern is interwoven with socio-economic status, infrastructure quality, and travel timing. Factor analysis identified harassment as the most influential latent construct (Cronbach’s Alpha = 0.885), with frequent harassment (0.780) and the ability to register complaints emerging as central. Structural Equation Modelling further revealed that women’s capacity to respond to harassment, combined with perceived safety at bus stops and during travel, significantly improves comfort when traveling at night. This nuance challenges the assumption common in both policy and scholarship that infrastructural upgrades alone are sufficient. Instead, the findings highlight the interplay of social norms, behavioral confidence, and physical design in shaping perceived safety.

The analysis also offers a counterpoint to the literature’s tendency to overemphasize nighttime risk by showing that daytime travel was consistently perceived as safer (loading = 0.647) and that crowd presence at bus stops (0.724) can enhance security perceptions, findings that resonate with urban safety theories. However, these results also caution against simplistic interpretations: crowd density may deter harassment in certain contexts but exacerbate discomfort in others, depending on crowd composition and surveillance measures.

By linking the empirical findings to the literature gap, this study advances three scholarly contributions. First, it operationalizes a multidimensional framework that integrates physical, social, and behavioral determinants of perceived security, an approach seldom applied in South Asian transport research. Second, it refines our understanding of the temporal dimension of safety, revealing the significant ways in which time of day interacts with other security factors. Third, it bridges gendered mobility studies with applied transport planning by translating statistical insights into targeted intervention points, such as well-maintained and monitored waiting areas, robust complaint mechanisms, and awareness campaigns to enhance women’s response capabilities.

For policymakers and practitioners, the findings underscore that improving women’s mobility requires more than isolated measures; it calls for coordinated interventions addressing harassment prevention, infrastructure quality, and socio-cultural barriers simultaneously. For scholars, the results illustrate the value of city-level, mixed-method research in capturing the heterogeneity of women’s security experiences, thereby offering a blueprint for similar inquiries in other urban contexts across South Asia.

7. Limitations and Future Research Recommendations

While this study advances the understanding of women’s security in public transport and addresses a key gap in city-specific research, it is not without limitations that should inform the interpretation of its findings. The reliance on perception-based survey data introduces potential biases: respondents may under-report harassment due to stigma or cultural norms, or overemphasize certain risks based on individual experiences rather than actual prevalence. This limitation is particularly pertinent in conservative societies like Pakistan, where social expectations often constrain open discussion of gender-based harassment. Furthermore, the cross-sectional design captures a single moment in time, limiting our ability to assess whether interventions or changing socio-political contexts alter women’s perceptions over the long term.

Another limitation lies in the demographic scope of the sample. While this study reflects the experiences of a significant portion of female public transport users—primarily younger, educated, low- to no-income women—it does not adequately represent older women, differently-abled individuals, or women from the most marginalized socioeconomic groups. This means that certain dimensions of vulnerability and resilience in public transport environments may remain unexplored.

Addressing these gaps presents several clear avenues for future research. Longitudinal studies should be employed to monitor how perceptions and behaviors evolve over time and in response to policy interventions, infrastructure changes, or social campaigns. Triangulating survey data with objective indicators—such as police records, transport authority logs, and observational audits—would provide a more robust evidence base for identifying and addressing security threats. Moreover, adopting mixed-methods approaches that combine quantitative modeling (e.g., SEM) with qualitative techniques (e.g., in-depth interviews, focus groups, and ethnographic observation) would deepen contextual understanding and help validate perception-based findings.

There is also a strong case for incorporating behavioral tools, such as situational judgment tests, to assess how women might respond in real-world harassment scenarios. Such tools could help identify gaps between perceived and actual response capabilities, offering targeted entry points for empowerment and training programs. Comparative research across cities—both within Pakistan (e.g., Karachi, Islamabad) and across the South Asian region (e.g., Dhaka, New Delhi)—would also be valuable for testing the transferability of the findings and for developing adaptable, region-wide strategies.

In sum, while the present study strengthens the empirical and analytical foundation for understanding women’s public transport security in Lahore, it also opens the door to more inclusive, methodologically rigorous, and comparative research. The interplay of socio-cultural, infrastructural, and behavioral factors uncovered here underscores the need for integrated, context-sensitive interventions. By addressing both the tangible environment and the socio-behavioral landscape of public transport, future work can contribute more directly to the creation of mobility systems that are not only efficient but also genuinely safe and empowering for women.